*

‘%’#i3\5
i?-
ECONOMKX
$1
()!)

Investment Decisions in Petroleum Exploration

and Production
IAN G. NORTHERN SOCOtW MOBIL OIL CO.
MEMBERAIME NEW YORK, N. Y.

Abstract cdl and environmental risk is frequently greater than in

most other forms of business enterprise. Therefore, as
Ecortol?2icevult(atiotz in [he exp[ctratiotz and produclian others have done before,’ let me emphasize that recourse
ettd of tile petrofeunz industry is quite diflereut because we to formal project evaluation cannot eliminate uncertainty
are dealing wirh alz ituiastry in which bodz the diversity and risk. What it can and should do is permit orderly
and extent of the technical and environnzentul risk is fre- consideration of the factors that are pertinent to an in-
quently greater than in most other fornzs of business etr(er- vestment opportunity and produce a brief economic point
prise. With tizis in mind, four profit indicators iire discus- of reference for management decisions.
sed, with their short-conzitzgs andadvaotages. Payou~ time,
vndiscountecl profit ratio, discounted cash jlow rate of re- Evaluation Criteria
turn and present value ratio or present value profit ratio
are tile foztr tnethods discussed with application to an Right at the outset, 1 would like to discard from general
actualdeve!opnzent drilling case hislory. Allowances jor application in an exploration and producing setting all
[he risk involved in development drilling and production those profitability measures that fall under the heading
are rzl.sodi.rcassed. OEaccounting methods. In this category I include all cri-
teria that involve investment and/or income averaging to
Introduction produce a percentage return on investment figure. Though
these older evaluation methods may have utility under
A business historian writing a number of years from conditions of more or less equal project life and level in-
now might WC1llook bizck to the middle years of the come stream between competing projects, neither of these
twentieth century and label them “the age of evaluation”. conditions is met in the usual range of exploration and
Only a few years ago evaluation procedures that are in producing investment opportunities. Instead, I would like
common use today were the somewhat mysterious pre- to focus on four profit indicators that I believe do have
serve of financial institutions irrzd theoretical eccznomists. utility in our end of the -business. These are:
The process of translzition tkom theory to business appli- 1, Payout time-a measure of the length of time re-
cation has not always been errsy, nor is the process of quired for accumulated net earnings from a project to
change complete. Though our industry has developed a equal the amount of project investment;
broad base of evaluation know-how, 1 think it is a fair 2. Undiscounted profit ratio — a simple expression of
assumption that the application of economics and mathe- ultimate net profit returned by a venture as a multiple of
matics to the field of business management is still in an the total investment in that venturg
expanding phase, and that certain aids to judgment will 3. Discounted cash-flow rate of return—the constant
become more effective with time, Greater etlectiverress rate of interest that can be earned on outstanding project
does not imply intricacy and involvement. A sound eval- investment while permitting exact amortization of the full
uation policy need not be complex, Whh a proper under- amount invested over the project’s useful life; and
standing of concepts, techniques and limitations, the ex- 4. Present value ratio or present value profit ratio-a
tent to which individual evaluations shozzldbe carried and
ratio of project-revenue to investment or project profit
the particular characteristics that should be captured in
to investment, with both investment and income reduced
profit indicator form become more a matter of common
to present value terms at a predetermined rate of interest.
sense than of going by the book,
I propose to enlarge uprtn these concepts by drawing
Whatever form evaluation progress adopts, however. upon an actual development driI1ing case history. This case
th~Urt&rlyirzg facts of life ss far as the exploration and ‘- -history -involves -40- separate developnlenL drilling ProPW
production business is concerned will remain essentially sals,’ each one evaluated on an after-tax basis, and each
unchanged. We are dealing with a branch of the petroleum without bearing any part of prior exploration expenditure.
industry in which both the diversity and extent of techni- The wells chosen for inclusion in this example are not
meant to synthesize the economies of development drill-
Orircinal manuscript reeeived in Society of Petroleum Ensrineers office
Feb. 16, 1964. Itevised mmmscrbt received June 6, 1664, Paper Zsve!?entcd
at AIME annuat Meeting in New York, Feb. 16-~0, 1964. Ittcfwenws uiven d. end d wwer.

JULY, 19rIt 727

 * ..

ing in any one area. A different 40-well sample wouki of development driliing. Where evaluation covers Ihc cx-
almost inevitably display different correlation figures, dc- ploratims as well as the development phase of operations,
p.ersding on the degree of uniformity in unit investment payouts of 10+ years may well be encountered, and dif-
and expense levels, the quantity and quaiity of reserves, ferences in the length of time that investment funds arc
the impact of proration control on production rate, etc, exposed may, if compounded with high technological, poii-
Payout Time tical or economic risk, assume considerable importance,
The rmme weakness is that it mewres only liquidity of If such is the case, it is frequently desirable to show more
investm&t. Itdoes not measure the extent to which the than u singie-figure expression of final investment liquid-
venture is in the red from the time of inception to pay- ity. Though two projects may display equal payout times,
out, nor does it provide any indication of either the the extent of maximum investment exposure and the un-
amount or timing of income receipt beyond the point of recovered baiance at different stages of the climb, toward
payout. Fig. 1 shows the individual plot points for the 40 payout might be quite ditferent, These differences can be
development ventures, with payout time in years shown effectively depicted -in graph form,
along the horizontal and the undiscounted profit ratio Undtseounted Profit Ratio
measured on the vertical. The proposal with the shortest
Absolute profit on its own telis us very little about pro-
payout, 1.4 years, ranks 32txi in terms of profit ratio,
“while the project with the longest payout, 5.5 years, ject worth. Absolute profit related to dollars invested, or
stands 16tit in profit ratio ranking. the “undiscounted profit ratio”, is an improvement, but
still suffers from the weakness of ignoring time value on
On Fig. 2, payout time is plotted against DCF rate of
both the outgo and income sides of the equation. Further
return, and somewhat better correlation is apparent, How-
complications may arise in arriving at a completely con-
ever, the relationship is still very imprecise, Assume for
sistent definition of what constitutes “investment”.
the moment that we adopted payout as a prime measure
of relative attractiveness, Within the fairly narrow payout The absence of a norm in the shape of project cash flow
range of 2 to 2.5 years we would have our choice of pro- automatically implies the absence of a standard relation-
jects that had a range in profit ratio terms of from 0.2 ship between undiscounted profit ratio and investment
to 2.3, while the spread in DCF goes from a low of 19 worth, when reduced to present vaIue terms. Even withirt
to a high of 46 per cent. the field of development drilling, where the bulk of in-
Despite its obvious limitation as a critical tool for vestment normally falls ciose to the time of project in-
selwting between competing investment opportunities, ception and the revenue stream normaJIy avoids erratic
payout time, on occasion, has its uses. Liquidity of in- movement, the relationship is very loose. This is borne out
vestment funds may be a factor of significance in financial in Figs. 3 and 4.
planning, in which case the decision maker may be pre- In the first of these,” the undiscounted profit ratio is
pared to give up something on long-term profit and earn- plotted against DCF rate of return and a broad fan effect
ings rate in order to recoup investment at an early date. results, The degree of dispersion is less in Fig. 4, in which
Additionally and realistically. investment choice will not undiscounted profit ratio is measured against present vaiue
always be confined to fairly short payout projects typical profit ratio, the latter embracing an arbitrary 10 per cent

I
m ‘1 0 ‘1
0

, t I 1 1

4 10 20,30 40 ~ 4
Payout In Years D.C.F. Ralie of Return
— l—Individual
Fig. plot-points for the 40 development Fig. 3—Individual pIot-points for the 40 development
ventures. ventures,

3.0

a
. *

$1.01 L no I 0 I I I

1 2. 3 4 56
‘ L“’:l’;”oo
0%1--
t
0.20 0,40 O.@
1

o ●t
1
‘
1.00
I

10$ Preisenk Worth Ratio

Payout in Years
Fig. 2--IndMdttal plot-pohtts for the 40 development Fig. &individual plot-pojn~ for the 40 development
ventures. ventures.

728 JOURNAL OF PETROLEUM TECIINOI.OCY

.-’ —
 —
.,
..-

,’

discount rate. The degree of dispersion is,, of course, a $1,400, with annual amounts decreasing from the first to
function of the discount rate employed, with deviation in- last years. Project B promises revenues that start low’,
creasing as the discount rate is raised, and vice versa. climb to a plateau through the middle years and then fall
As a completely independent criterion for project ap- away at the end, Total revenue from Project B is higher
proval or rejection, undiscounted profit ratio, may, in its at $1700. Using year-end discount factors, the .J3CF solu-
lower range of values or in instanws of unusual income tions approximate 46 per cent for A and 36 per cent for
patterns, lead to poor investment decisions, As an inde- B. The advantage to A extends to payout time also,—L7
pendent measure of relative attractiveness between com- years compared with 2.8 years for Project B. The rank-
peting projects, it will, because it excludes time value, fre- ings are reversed for the undiscounted profit ratio para-
quently result in less than optimum deployment of in- meter, B yiefding 2,7 to A’s 2,04. That a preference for
vestment capital. Obviously, therefore, it should not be Project A because of its superior DCF rate and shorter
employed to the “exclusion of all other profitability measu- payout could be misplaced is readily demonstrated by in-
res. Rather; it is, a very necessary complement to @e troducing the principle of reinvestment. Under the rea-
DCF rate of return in project evaluation, sonable assumption that funds returned to the corporate
kitty are undergoing a process of continual reinvestment,
DCF Kate of Return ‘ and with the premise in this example that they are ex-
Undiscounted profit’ ratio ignores time value, DCF rate pected to yield an average earnings rate of 10 per cent
of return frequently overemphasizes time value, with the per annum, the $280 returned by Project A at the end
extent of overemphasis increasing in the move from low of the first year will have compounded” to $660 by the
to high DCF solution rates. Disregarding risk, use of DCF end of the 10th year. The $240 received from A at the
rejection rates will assuredly avoid investment in projects end of the second year will, if reinvested at 10 per cent
thal promise a yield to maturity of something less than the compound interest for eight years, have grown to $515
cost of capital or whatever other minimum e.arnings-rate at the end of year 110. Following this simple exercise
objective may be set. It will not, however, if used in isola- through, as in Table 2, for the two projects, it is provable
tion from other selection criteria, provide a proper ranking that had we elected Project A the net worth of our initkrl
of investment opportunities; and may, as its prime danger, $460 would stand at just short of $2,600 after 10 years,
very well result in incorrect decisions between mutually while B would yield an accumulation of close to$2800.
exclusive, investment opportunities. Its failing is concep- Purely by way of interest, the equalizing rate, or rate at
tual, Discounted cash flow inherently requires, as an ab- which revenues from each projeet would have to be re-
solute measure of investment preference, that project earn- invested to yield equal cumulative worth at the end of
ings be reinvested at the solved rate of return. This be- 10 years, is in this example 17 per cent, I read into this
comes an unrealistic requirement when dealing with high ilhrstration the conclusion thtst DCF alone, especially in
DCF solutions. From a practical point of view, the poten- its higher ranges frequently encountered in the develop-
tial for erroneous ranking only looms large in cases where ment phase of E&P operations, can be an insufficiently -
the DCF sohstion is well above the rejection rate, and the critical selection tool. Prudence dictates that undiscounted
answer will frequently be that all such projects, irrespec- profit ratio needs concurrent consideration. This could
tive of correctness in ranking, are worth undertaking. Such leave us with the problem of informally weighing DCF
is normally the case in development drilling where the” and profit ratio or with the alternative of preparing labor-’
only problem of optimization would lie in a r~tioning of ious calculations of the type ussd in the example. Fortun-
capital that exhausted ready investment funds well up on ately, solutions based on the. present vahre ratio concept
the DCF ladder of attractiveness. The potentissl for poor skirt these problems in ti very stmightforward manner.
investment counseling through reliance on DCF is greater,
Present Value Ratio
though, in the area of mutually exclusive ventures. This
could, for instance. involve the relative economic merits Present value solutions to investment ,problems avoid
of alternative approaches to fieId pressure maintenance or the conceptual flaw inherent in ,the DCF rate of return
to alternative production facility installations. Because 1 approach by discounting revenues attributable to the proj-
believe that this is an area of considerable pmctical sig- ect under immediate review at a predetermined rate of in-
nificance in the petroleum business. I think it will be terest, The rate selected should preferentially be the anti-
worthwhile emphasizing the point with a short example. cipated’ average interest rate at which corporate funds can
be reinvested in future years. Use of the average rein-
Table 1 tabulates investments and a series of annual
vestment rate presumes that in most instances, there will be
revenues for two competing projects. Both involve an
a budgetary or organizational constraint on :he amount of
initial investment of $460 and both have a. 10-year life.
money that can prudently be invested in any time period.
Project A. however, promises a total lifetime revenue of
Were this not the case, then maximization of return on
..— shareholders’ equity would best be served by acceptirtg
TASLE I —COMPARATIVE PROJECT ANALYSIS all investment opportunities that promised an earnings
Pmiect A Proled
.-. —.—
B
.—— —--
. . . ..— — .—— -----
Ww P. w.
TABLE 2-COMPARATIVE PROJECT ANALYSIS INTRODUCING THE
RE-I NVES~MENT PRINCIPLE
Prc?ied A
— .--— . PrOiect
.—.———-S
c;a7::uj- C;a:gu::

End of End of
Year 10 Year 10
—. Year Net. Re.. I@Ioy. ~Net. Rev. Q“a”I Oy.
m“- ,,- ~.
1
.. ‘4!” ala -. 19V
It! ii z_. .–— ———40 i - 210 409 220 ‘~
180 319 220
To to I 460 1400 45s 460 1700 457 150 242 220 354
-—. PrOiect 6 120 176 220 322
A.
7 $ ;~ I 20 220 293
— ——.. B- S5 1so 21s
DCF Rctn of Return 45.7% 35.s% ., ; dO 44 100 110
Pawvt Time 1,7 yrs. 2,S yrs. 10 20 ..— 20 — 40 40
Undiscounted Profit Ratio 2.04 2.70 TQId 1400 “- “ 2590 I 700 27S2

JULY, 1964 729

—.
 ,’ .-

rate in excess of the marginal cost of capital, implying

that the marginal rate of borrowing and not the expected
average reinvestment rate should be employed for dis-
count purposes,
To revert briefly to our two-case example, Table 3 tab-
ulates the present worth of the respective revenue streams
when discounted at 10 per cent per annum.
The sum of the discounted annual revenues is then ex-
pressed as a ratio to the required investment to establish
a present value ratio.
Alternatively, and as has been adopted in the figt.rre in
this paper, the relationship may be expressed in terms of
present value pro5t. It should be noted that both rela-
tionships require that investment, if made in more than
one time period, also be reduced to present value, using
whatever opportunity cost dk.count rate is judged appro- 10$ Present Value Ratio
priate for the year in which the investment is made.
Fig, GDFC rwte of re:nrn w 10 per reot present vulsre
Results in terms of our 40-well inventory of develop- profit ratio.
ment drilling opportunities are also of interest. Fig. 5 pro-
vides a plot of DCF rate of return against 10 per cent
present value profit ratio. respective price-earnings ratios, current yield ond assessed
To state the obvious, correlation is poor. The project growth prospects, it is highly unlikely that the same issues
with the highest DCF rate of return, 53 per cent, ranks would fill the same positions by each method of rating.
only eighth in terms of 10 per cent present value ratio. ‘“”‘Nor could we expect that a nurnbcr of investors would
while the project with the highest 10 per cent PVR rates select the same issues for investment purposes. Each in-
seventh in DCF. Extending the comparison a stage furth- vestor would, if prudent, build his holdings to meet his
er, the order of finish for the five top-rated 10 per cent. particular needs, be they security, high yield, capital ap-
PVR projects in terms of DCF rate of return, undis- preciation or some mixture of these.
counted profit ratio and payout time is given in Table 4. Business management has similar problems in clecid-
One thing should be made clear. Technically, I believe ing how best to apportion corporate investment funds, and
that investment advice based on the present value method the optimum investment program is not necessarily that
of analysis is sounder than that derived from the DCF which picks only from the top of the present value rating
rate of return method, I do not suggest, however, that tree. This is where the contribution of the evaluation
approval- or rejection-type decisions or decisions involv- leaves off and management judgment begins, The divid-
ing a choice between alternatives follow automatically ing line is not clearly defined,. and at least in one respect.
from present value ratings, Quite apart from countless that of risk appraisal, is the cause of a deal of present-day
factors outside of the economic evaluation that have to controversy. The problem, and one that I would, in clos-
be considered, it would be an oversimplification to assign ing, like to look at briefly, is the extent to which allow-
a tlxed order or degree of importance to individual profit ance for risk and uncertainty should be formalized as a
indicators, Each of the four discussed has good points part of project evaluation,
and failings. Whh regard to the latter, the weaknesses in
payout time, undiscounted profit ratio and DCF rate of Allowance for Risk
return are a product of limited or unrealistic concepts.
The difficulty with present value ratio is partly one of Most exploratimr and production ventures are charac-
relative complexity, of concept and partly due to the neces- terized by relatively long life and a high degree of irre-
sw introduction ,of a reinvestment assumption that is vocability. Though the explanation of major changes from
susceptible to some marginal error of prediction. anticipated to actual economic performance will often be
Though the four profit indicators are not truly inde- found in physical factors—incorrect reserve estimation
pendent measures of project worth, they will inevitably or unexpected process performance—the combined im-
rank a cross section of investment opportunities differ- pact of poor environmental assumptions can also be of
ently. There is, I believe, a reasonable parallel in the considerable importance. This is, I think, particularly true
securities market. If we were to take a random sample of of .mverseas operations. Perhaps, fortunately, many en-
common stock issues and ladder them according to their vironmental factors can have their near-tcrtn uncertainty
expressed in a fairly narrow range of values. This is true
in dealing with allowabIes, prices, costs, and, domesti-
TABLE 3-COMPARATIVE PROJECT ANALYSIS 10 PER CENr PRESENT cally at least, taxes. Though not one of us can predict
VALUE RATIO
exactly how these items will trend in future years, I would
Prolect A Proiect S
guess that our range of opinion would be relatively small.
In. ln-
veM- Net l:% ve5t- Net I;;b It still remains important that within a company differ:
— Year — ment ——Rev. — ment Rev. ences of opfition should be resolved and a common set
460 280 254 ‘“— 460 — I 20 -——w—’ ,
: 240 19s 160 132 of environmental premises developed.
210 lsa 220 166
;------ -;:: -- 123 -220- .- 150
220 137
6 1;: :: 220 124
: 46 220 113 TASLE 4—COMPARATIVE PROFIT INDICATOR RANKING
70 33 180
10% PVPR DCF RQR Undibc. PR Payout Time
40 17 100 , !:
Iz 20 8 40 15 —1 St [0.93) ‘—— ---z-@z7r 1; f:.:
w z 1AOO m 2nd [0.S4) 5 [1 ,7s1 I
x 1700 W2 %~j
3rd (0.S1 I I (2.451 13 [2:6]
10% PV Ratio 99S/460 = 2.17 10721460 == 2.33 4th (0.79] 10th [3S00] 4 (2.03] 15 (2.s1
B/A = 2.33/2.17 = 1.074 S/A = 2782/2590 ‘= 1.074 5th [0.721 16th [2S !/o) 3 (2.17] 24 [3.6]

.730
 Though time will almost certainly prove errors m out- Though this paper has been concerned primarily with
look, at least the degree of error will be consistent, and the end product of the evaluation process, it is appropri-
a fairer basis for judging between competing opportuni- ate, in closing, to re-emphasize that the evaluator can no
ties established, Furthermore, though the margin of un- more convert dubious input data into sound ecmomic
certainty will seldom spell the difference between riches guidance than the m@leval alchemist could convert base
and ruin, the evaluator is under an obligation, where metal into gold. Regardless of what profit indicators are
first runs suggest a borderline decision or when problems considered significant under any given set of circwrs-
of estimation are acute, to indicate numerically or graphi- stanccs, a good 95 per cent of over-ali evaluation effort
cally the extent to which the project under review is sen- should be centered on proper project de5ition and the
sitive to plausible shifts in environmental or physical fac- construction of reaiistic physical and financial premises,
tors. I don’t know of any rules that spell out what to do This is the hard part. What follows is a mixture of under-
or when to do it. This should be left to the good judgment standing of methodology, proper comprehension of com-
of the evaluator, the main point beitig that uncertainties pany gods and standards, and a very large slice of com-
of the form discussed should, when appropriate, be made mon sense,
an explicit part of the evaluation approach.
The other part of the problem relates to the sort of Acknowledgment
risk undertaken every time funds are committed to a wild-
cat drilling venture. I have no quarrel with statistical The author is indebted to countless others who in re-
theory as h relates to risk and probability analysis, Ap- cent years have written in the field of evaluation theory
proaches that, for example, involve the establishment of and practice, Extending proper credit to all who have
a number of possibIe venture results, weight each by the published and to many whose contributions have not ap-
odds of attainment and arrive arithmetically at a value peared in print would involve many pages of references.
that would match the average result actually achieved if Whh apologies, therefore, for the many worthwhile ref-
the identical venture could be undertaken an irttlnite num- erences excluded, exceiient coverage on many nf the
ber of times are theoretically sound, I will agree, too, points raised in this paper may be obtained from the fol-
that the theories can be applied beneficially where the lowing short list.
number of possible results is limited, where the chances
of attaining each can bc set with some confidence, and Keferenms
where each possible result can be defined in reasonably
precise profitability terms. Some exploitation problems 1, Lutz, Friedrich and Lutz, Vera: The T/wury o/ investment o/
tke J’irm, Princeton U. Press (1951 I.
can be so defined. I do not believe, h’owever, that these 2. Solomon, ha: The Management Oj Corporate Capital; The
conditions are met very often in a petroleum exploration Free Press of Glencoe, 111, (1959).
setting, Obviously, the time at which evaluation advice % Hirshleifer, J.: “On the Theory of optimul Investment Deci-
would be of the greatest value’ in exploration operations sion”, Jour. of Political Economy ( Aug,, 1958 ),
is right at the start of the venture, before any substantial 4. Wade, F. R. and Evert D. E.: “Simplified Methods for Invest-
funds are committed. This also happens to be the time at ment Analysis in the ~%trolcmn Production Industry”, Paper
which knowledge, is letist perfect. By the time the cle- 1526-G presented at the SPE 35th AmlnaI Fall Meeting,Denver,
Oct. 2-5, 1960.
cision to drill or not to dril~ is reached, possibly some- 5. Pollack, G. A.: “TIw Capitai Budget Contrivers : Present
thing like 60 to 70 per cent of the total finding cost doJ- Value w Discounted Cash Flow Method”, BuU, N.i ,A. (Nov.,
Iar has already been committed, and all the evaluation 1961 ).
can proffer is advice on the wisdom of spending incremerr. 5, Johnson, Herbert W.: “Measuring the Earnin Power of Invest-
tal dollars, I do not use the word “aII” disparagingly, for ments-A Comparison of Methods”, Bull,, N, 1 ,A, (Jan,, 1962),
I am aware that this phase of the business involves hun- ?, McLean, John G.: “Techni nes fur the Appraisal of New Cap.
, dreds of millions of dollars each year, Exploration statis- ital Jnvestmcnfs”, Proc,, Sixt1 I World Petroleum Congres, Frank.
fort (June, 1963 ).
tics, however, bear mute testimony to the fact that our
pre.drilling krtowIedge is so imperfect that only a very
small percentage of all wildcat bets pay off with econortiic
discoveries.
Whether or not probability analysis in its many forms IAN G, NORTHERN is a senior pro-
pays off in an exploration and producing setting cannot ducing econotnist in the Exploration
be answered with art outright yes or no, My own view is and Producing Dept. of Socony Mobil
that, where the ~hvsical and financial alternatives are rea- Oil Co. in New York, A 19S3 graduate
sonable well defined. it has definite utilitv. Where t)hvsi- of the London School oj Economics,
cal and financial control is poor. I question wheth& ~he he spent five years with Socony’s Ca-
certain expenditure of time and temper wilI be rewarded nadian Div, before moving to New York
hy improved performance. in 1959.

.. -—

lULY, 1964

. ~ . ———_

,“. ---’, . . . . . . .. . . .. .