Jurnal 2 2016

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

EM174846 DOI: 10.

2118/174846-PA Date: 20-July-16 Stage: Page: 68 Total Pages: 10

Progressive Royalty Framework for


Oil- and Gas-Development Strategy:
Lessons From Nigeria
Joseph C. Echendu and Omowunmi O. Iledare, University of Port Harcourt

Summary specify high royalty rates by the host governments. Although this
This paper presents a progressive royalty framework and investi- payment does not begin until production begins, it is considered
gates the effects of the various kinds of royalty schemes on oil- to be regressive because it is dependent on gross value or pro-
and gas-development economics. The conducts and performances duction regardless of whether profit is made from the portfolio
of fixed, jumping, and/or sliding royalty schemes are evaluated. of assets.
Further, the paper reviews the different sliding-scale specifica- It is critical to note that high royalty rates do not necessarily
tions in fiscal systems and recommends the optimal boundary to equate to maximized royalty revenue (Lohrenz et al. 1981; Mer-
ensure efficiency and effectiveness. The proposed royalty scheme cier et al. 1997). Mercier et al. (1997) noted that especially for
recommended in the Inter Agency Team memorandum on the net-profit-sharing contracts, high royalty rates typically require
2008 Petroleum Industry Bill (PIB) of Nigeria (PIB 2008), which constant evaluation of operation details (costly administrative
was tied to terrain, geology, and value, forms the basis for the roy- overheads) and lead to increased financial liability. Not having a
alty design and modeling analysis evaluated in this paper. proportionate amount of increased benefits as a result of the
The logarithmic sliding-scale royalty scheme is generally per- increased financial risks could lead to lower production, which
ceived to perform better than other schemes (linear, jumping, or reduces the mineral owners’ royalty revenue and would not guar-
fixed scale), but our analysis shows that the effects of terrain and antee future oil production in the event of lower oil prices.
geology matter greatly. This implies that if the royalty scheme is Iledare (2014) affirms that attracting investments to any petro-
tied to geology, then marginal field operators with marginal pro- leum province requires having a dynamic, progressive, efficient,
duction rates would prefer the linearly sliding-scale mechanism and stable fiscal arrangement (Iledare and Echendu 2013).
to logarithmic scale. Flexibility to switch from one scheme to Dynamic and stable fiscal systems literally may require surrender-
the other offers incremental fiscal-regime progressiveness in the ing a great proportion of economic rent to investors to ensure the
quest for efficient, effective, equitable, and ethical energy- prospects of high rewards in countries with high exploration risk
resource development. and low prospectivity. Investors may have no interest in petro-
The paper recommends the use of the sliding-scale royalty leum-resource development if factors affecting the balance
scheme as a progressive bidding parameter and subsequent rent- between inherent risks and rewards are incoherent despite the
extraction instrument. The uncertainty in oil prices and its plum- promising economic opportunities.
meting trend—plausibly discouraging exploration and exploita- This paper applies the proposed royalty scheme of the Inter
tion of the not-easy-to-find hydrocarbons—requires progressive Agency Team memorandum on the 2008 PIB and the proposed
royalty schemes to create equitable performance outcomes for all 2012 PIB (PIB 2012) submitted to the 6th and 7th National As-
stakeholders in the emerging new petroleum era. sembly of Nigeria in its analysis. The purpose is to evaluate the
conducts and performances of fixed, jumping, and/or sliding roy-
alty mechanics. The effect of each scheme on oil- and gas-devel-
Introduction opment economics is investigated. Although the logarithmic
Royalty is a fundamental element in the international petroleum sliding-scale royalty scheme is generally perceived to perform
fiscal system through either concessionary or contractual fiscal better than the linear sliding or fixed scheme, to ensure future pro-
systems (Mian 2010). It is strategic for economic rent extraction duction, flexibility of a possible switch from one scheme to
by a host government toward maximizing the social well-being the other seems to be desirable in the quest for maximizing
of a petroleum-producing country. Consequently, investors and petroleum-resource-development economics. Further, the paper
petroleum-resource owners during the bidding process have nego- reviews the different sliding-scale methods and recommends the
tiated to maximize individually defined objective functions under- upper production boundary to ensure efficiency and effectiveness.
lying the exploration and development of endowed petroleum
resources in the host country.
Generally, economic rent reflects the difference between the Overview of Royalty Specifications in Nigeria
value of production and the costs of extraction (Iledare 2014). To The Petroleum Act of 1969 and the Deep Offshore and Inland Ba-
the investor, royalty, paid in cash or in kind, is a part of the cost sin Production Sharing Contracts Act (DOIBPSCA) of 1999 are
of extraction, including exploration, development, and operating some of the key legislation that provides the fiscal provisions gov-
costs. It is tax-deductible in oil and gas economic-performance erning petroleum operations in Nigeria. The Petroleum Act of
calculations. It is a mechanism through which the host govern- 1969 is the main legislation governing matters relating to petro-
ments secure upfront revenue as soon as production starts from leum exploration and production in Nigeria, and it vests full own-
the lease/block/concession. Economic rents extracted at the time ership and control of all petroleum in, under, or upon any lands
of transfer of rights, such as rentals or signature bonuses, and (including underwater) to the federal government of Nigeria. The
through royalties are not dependent on profits. In the design of the Petroleum Act of 1969 requires the holder of an oil-prospecting or
fiscal system and with the objective of the host governments to oil-mining license to pay royalties to the federal government of
place the highest value on their resources, there is the tendency to Nigeria as soon as production starts. This is usually in the form of
monthly cash payments at an agreed percentage of the quantity of
crude oil/gas produced, after making adjustments for treatment,
Copyright V
C 2016 Society of Petroleum Engineers
handling, and related expenses or by way of royalty oil. The
This paper (SPE 174846) was accepted for presentation at the SPE Annual Technical DOIBPSCA (1999) sets out the general framework for the opera-
Conference and Exhibition, Houston, 28–30 September 2015, and revised for publication.
Original manuscript received for review 25 June 2015. Revised manuscript received for tion of production-sharing contracts (PSCs), including applicable
review 7 March 2016. Paper peer approved 8 April 2016. royalties, tax regimes, costs treatment, and profits allocation. It is

68 July 2016 SPE Economics & Management

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 69 Total Pages: 10

Royalty Rates
Royalty Rates Average Daily Production, q Proposed Proposed
Water Depth, d (m) 1993 PSC 2005 PSC (1,000 BOPD) PIB 2008 PIB 2012
Onshore 20.00% – Onshore
0 ≤ d ≤ 200 16.67% – 0≤q≤2 5.00% 5.00%
200 < d ≤ 500 12.00% 12.00% 2<q≤5 12.50% 12.50%
500 < d ≤ 800 8.00% 8.00% q>5 25.00% 22.00%
800 < q ≤ 1000 4.00% 8.00%
q > 1000 0.00% 8.00% Shallow Water

0≤q≤5 5.00% 5.00%


JV 5 < q ≤ 20 12.50% 12.50%
Onshore 20.00% – q > 20 25.00% 20.00%
0 ≤ d ≤ 100 18.50% –
Deepwater
100 < d ≤ 200 16.67% –
200 < d ≤ 500 12.00% – 0 ≤ q ≤ 50 5.00% 5.00%
500 < d ≤ 800 8.00% – 50 < q ≤ 100 12.50% 12.50%
800 < q ≤ 1000 4.00% – q > 100 25.00% 18.50%

Table 1—Summary of royalty-rate specifications in Nigeria.

the enactment that governs the oil and gas companies operating in The royalty-rates frameworks in the proposed PIBs (PIB 2008,
the Deep Offshore and Inland Basin area under PSCs. Notably, 2012) to the 6th and 7th National Assembly in Nigeria identify
the royalty-rate specifications are fixed percentages tied to terrain/ three terrains as onshore/swamp, shallow offshore, and deep off-
water depth only for the existing fiscal regimes. shore (Iledare 2011). The royalty scheme specifies a sliding-scale
The royalty-rate specifications governing the joint-venture royalty rate tied to production rates and ranges from 5 to 25%
operations in Nigeria are typically a 20% fixed rate for onshore depending on terrain, as seen in Table 1. Although not part of the
production, an 18.50% flat rate for offshore production up to scope of this paper, it is noteworthy to mention that lower rates
100 m of water depth, and a 16.67% rate for offshore production are proposed for condensate from nonassociated gas fields and
beyond 100 m of water depth (PwC 2015). These royalty rates also for frontier and ultradeepwater basins. There is also a pro-
govern the standard agreement between the national oil company, posed royalty by value that triggers off when the crude oil and
represented by the Nigerian National Petroleum Corporation condensate price exceeds a threshold value of USD 50/barrel and
(NNPC), and the international oil companies with proportionate the natural-gas price exceeds USD 2/billion Btu. The proposed
contributions toward funding the joint-venture equity holdings. bills’ royalty rates would be dependent on production and value,
For the PSCs in Nigeria, the graduated royalty-rate specifica- and the effective rates would be calculated by use of a sliding-
tions are also by terrain/water depth, with 12% for water-depth scale mechanism and not the conventional fixed mechanism
areas between 201 and 500 m and 8% for water depth greater than (Echendu et al. 2014).
500 m but less than or equal to 800 m. For depths greater than The implications and effects of the proposed royalty rates by
800 m but less than or equal to 1000 m, a royalty specification of terrain and by use of different mechanics of applications to the
4% is stipulated, whereas there is no royalty payment for water rates, such as jumping and/or sliding scale, are evaluated and ana-
depths greater than 1000 m (Echendu et al. 2012). However, the lyzed in later sections. The possibility of applying a royalty
rate specifications are modified for the 2005 PSC, as seen in scheme and its effect as a bidding parameter are discussed here.
Table 1. In the PSC arrangements, the national oil company holds The effect of the proposed royalty rates by value with respect to
the concessions for the government and engages the international the oil price shall be undertaken in another study.
oil company or indigenous company as a contractor. Typical of
PSCs, the contractor takes on the financial risks and is entitled to
recover its cost if exploration is successful but cannot recover its Royalty-Rate-Design Mechanics
costs if exploration is not successful. Generally, gross revenue GRt forms the base for royalty payment,
ROYt, and this rate-base form of rent extraction is perceived to be
regressive because it is not tied to profits and it reflects a risk-
Royalty Rate aversion attitude of the host government. The royalty payment is
the product of the rate and the gross revenue as estimated here:

Fixed/Constant Nonfixed/Graduated ROYt ¼ Rð/Þ  ðGRt Þ: . . . . . . . . . . . . . . . . . . . . . . ð1Þ

The proposed royalty rate Rð/Þ depends on the terrain, pro-


duction rate at time t, and the incentive schemes, if any, in effect.
Sliding Scale Jumping The royalty rate Rð/Þ, 0  Rð/Þ  1, may be fixed or sliding
scale. The terms of the royalty rate, like many other production-
sharing-agreement factors, may be negotiable. The preferred
Nonlinear/ approach is to legislate royalty-rate terms, especially in petroleum
Linear Scale
Logarithmic provinces where transparency and accountability are deficient.
Fig. 1 illustrates two broad types of royalty-rate classifications
Fig. 1—Royalty-rate classification. that could be identified: fixed/constant-percentage royalty and

July 2016 SPE Economics & Management 69

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 70 Total Pages: 10

Average Daily Production, qd (BOPD) Royalty Rate (%) Average Daily Production, qd (BOPD) Royalty Rate (%)

qo ≤ qd ≤ qi Ri (φ ) qo ≤ qd ≤ qi Ri (φ )

qi < qd ≤ q j R j (φ ) qi < qd ≤ q j Ri (φ )

Table 2—Two-tranche royalty specifications (tranche level 5 i, j). qd > q j Ri (φ )

Table 3—Three-tranche royalty specifications (tranche level 5 i, j, k).


nonfixed/graduated royalty. The graduated royalty could be sub-
classified into jumping and sliding-scale royalties. The sliding-
scale royalty could be further subdivided into linear-scale and the proposed Nigeria PIB (Table 3), then a possible optimal
nonlinear-scale methods. boundary would require having qj < qd  qk .
The optimal production qk for the new tranche in a linear slid-
ing-scale scheme is
Fixed/Constant-Percentage Royalty Rate. Fixed- or constant-
percentage royalty was the most-common type of royalty in the Rk ð/Þ  Rj ð/Þ
1960s, whereby a fixed percentage of the gross revenue is paid ei- qk ¼ qj þ : . . . . . . . . . . . . . . . . . . . . ð6Þ
a1
ther in cash or in kind. Irrespective of oil price and/or production
rate, this percentage is applied to gross revenue. Examples of Nonlinear/Logarithmic Sliding-Scale Royalty Rate. It is also
countries that use a fixed-percentage royalty are Malaysia, Vene- assumed for the nonlinear-scale royalty rate that the first tranche
zuela, US, Ghana, Republic of Congo, and Gabon. Fixed-payment with instantaneous production has a constant effective royalty rate
royalty is a type of royalty in which a fixed payment/amount is up to the average production rate qi : The effective royalty-rate
paid to the mineral owner regardless of whether profit is made. It is progression is similar to that of a straight-line equation for each
no longer as commonly used as it was in the past (Iledare 2011). tranche, but inversely related to average daily production qd . The
effective royalty rate as a function of both the terrain and produc-
Nonfixed/Graduated Royalty Rate. Graduated royalty rates are tion for the logarithmic sliding scale is specified as
introduced to rectify the disadvantages of high fixed royalty (Pat- 8
terson 1979). It is used to account for uncertainties in field size, < Ri ð/Þ; q0  qd  qi ; q0  0
oil price, average daily production, geology, economics, engineer- R/ ðg; qd Þ ¼ 1
: a2 þ Rj ð/Þ; qi < qd  qj ;
ing, and so forth. Graduated royalty rates could be tied to R-fac- qd
tor, average daily production, cumulative production, oil price,                    ð7Þ
and project economic measures, as seen in petroleum fiscal sys-
tems. Nonfixed royalties could be jumping or sliding scale. In the where
jumping-scale royalty, the value or percentage to be paid is de-
pendent on the tranche (level) specified and is a flat rate at that a2 ¼ Ri ð/Þ  qi Rj ð/Þ: . . . . . . . . . . . . . . . . . . . . . . . ð8Þ
level, as in existing fiscal terms in Nigeria. In the sliding-scale
royalty, an effective value/percentage is calculated on the basis of
the tranche reached as being proposed. The effective royalty rate for a three-tranche terrain would be
An example of two-tranche royalty specifications is presented as follows:
in Table 2. 8
The effective royalty-rate-design mechanics tied to both ter- >
> Ri ð/Þ; q0  qd  qi ; q0  0
>  
>
rain and average-daily-production rate, R/ ðg; qd Þ, in this paper >
< a 1 þ R ð/Þ; q < q  q ;
2 j i d j
is discussed here for the two categories of royalty schemes pre- R/ ðg; qd Þ ¼ q
sented earlier. >  d
>
>
> 1
>
: a3 þ Rk ð/Þ; qj < qd  qk ;
qd
Jumping Graduated Royalty Rate. For the terrain g and aver-
                   ð9Þ
age-daily-production rate qd , the effective royalty rate as a func-
tion of both terrain and production is defined as
 where
Ri ð/Þ; q0  qd  qi ; q0  0
R/ ðg; qd Þ ¼ : . . . . . . ð2Þ a3 ¼ Rj ð/Þ  qj Rk ð/Þ: . . . . . . . . . . . . . . . . . . . . . . ð10Þ
Rj ð/Þ; qi < qd  qj ;

Methodology and Assumptions


Linear Sliding-Scale Royalty Rate. In the linear sliding scale,
Life-cycle oil- and gas-project evaluation is adopted to test the
the effective royalty-rate progression is similar to that of a
royalty-rate-design mechanism described previously. In this sec-
straight-line equation for each tranche or production-rate range,
tion, a discounted net-cash flow with the application of the previ-
but is assumed constant for the first tranche with instantaneous
ously discussed royalty-design mechanism is summarized. The
production. The effective royalty rate as a function of both the ter-
descriptive input variables and assumptions of the discounted net-
rain and production rate is given by
cash flow follow the pattern in Echendu et al. (2015). The model
 is formulated adapting the framework described by Mian (2010),
Ri ð/Þ; q0  qd  qi ; q0  0
R/ ðg; qd Þ ¼ ; . . . . . ð3Þ Johnston (2003), and Echendu et al. (2015).
a1 qd þ b; qi < qd  qj ; q0  0
Existing PSCs in Nigeria are currently governed with a royalty
where rate that is tied to water depth only. Echendu et al. (2015) eval-
uated the different PSCs in Nigeria, and although the rates are
Rj ð/Þ  Ri ð/Þ graduated as water depth increases, the application of the rates are
a1 ¼ . . . . . . . . . . . . . . . . . . . . . . . . ð4Þ fixed with respect to the water depth of production operations.
qj  qi
This study adopts the existing PSC of Nigeria with water depth
and b ¼ Ri ð/Þ  a1 qi ; . . . . . . . . . . . . . . . . . . . . . . . ð5Þ less than 200 m and uses a royalty rate of 16.67% as the base for
its evaluation. The adoption is informed by empirical evidences
where a1 is the constant slope of the linear scale. However, if suggesting that greater depths have prolific finds and their extrac-
another tranche specification is to be defined, which is the case in tions could not be classified as marginal production, and the fact

70 July 2016 SPE Economics & Management

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 71 Total Pages: 10

Variables

STOIIP* 250 Million bbl Estimated Reserves 100 Million bbl

% Cumulative production
Oil recovery 40 % 57.5% %
(≤15 MBOPD)

% Cumulative production
Buildup period 4 years 42.5% %
(> 15 MBOPD)

Initial oil production 950 BOPD Initial oil price 30 USD/bbl

Buildup pattern Linear buildup Capped oil price 50 USD/bbl

Oil peak rate 17100 BOPD Escalation factor 2% %

Plateau period 5 years Unit technical cost 15.00 USD/bbl

Production period 23 years Unit Capex 8.31 USD/bbl

Decline pattern Exponential Unit Opex 6.69 USD/bbl

Effective decline rate 0.06 Fraction Discount factor 10% %

Total wells to drill 18 Units Average well cost 20 USD million


*STOIIP = stock-tank oil initially in place

Table 4—Life-cycle project-input data.

that although most indigenous producers in Nigeria, with mar- Stochastic analysis on the input variable by use of the Monte
ginal reserves and productions, operate onshore, imposing the Carlo simulation approach is evaluated to account for inherent
lower royalty rates for a water depth of fewer than 200 m is more risks and uncertainties akin to oil and gas projects. Some key
conservative and encompassing for the study. The basic input var- input variables such as oil-recovery rate, which affects invest-
iables and assumptions are summarized in Table 4. ment, well rate, discount rate, oil price, and Capex, are selected
Typical of most petroleum-project life-cycles, the starting point on the basis of the high degree of inherent risks associated with
is the estimation of recoverable reserves. For the case study, a field them. Capital investment imposed in the study is a percent func-
with an estimated 100 million bbl in recoverable reserves is tion of the recoverable reserves and prevailing oil price; the
imposed with plateau-rate production of approximately 17,000 higher the recoverable reserves, the higher the expected capital
BOPD; this is typical of marginal operators in the case-study nation investment. The objective functions for the stochastic analysis are
of Nigeria (AOGR 2014). The model incorporates a price restric- the NPV, IRR, AGR, royalty revenue, tax payment, and govern-
tion of USD 30–50/bbl to be able to account for the dwindling oil- ment net-cash flow (NCF). However, the political-risk component
price shock that is currently being experienced, relative to the oil- is not captured in the model. Table 5 shows a summary of the
price boom that saw periods with oil price greater than USD 100/ project stochastic distribution.
bbl. The price restrictions and production rates give the expected
gross revenue. Capital expenditures (Capex) and operating expen-
ditures (Opex) are modeled as a percentage of the expected recov- Empirical Results and Analysis
erable reserves and oil price. This is premised on the fact that as oil After applying the royalty-rate-design mechanics described previ-
price dwindles, cost-cutting measures are adopted by investors to ously to royalty specifications in Nigeria as proposed, Figs. 2
have a return on investment; in addition, the invested capital in through 4 show the effects of royalty rates by volume on eco-
facilities and assets is also a function of expected production rate nomic-rent extraction. The jumping royalty rate as previously
and reserves capacity, with other conditions remaining the same. described is similar to the fixed royalty rate. However, it is only
In estimating the economic-performance measures such as net constant at a specified production tranche, unlike the fixed royalty
present value (NPV), internal rate of return (IRR), growth rate of rate, which would be constant irrespective of the production
return, discounted payout, present value ratio, discounted govern- tranche reached. For the first tranche, an effective royalty rate of
ment take, discounted contractor take, and access to gross revenue 5.0% is applied, whereas a constant effective royalty rate of
(AGR) that determine the bottom line, the existing PSC in Nigeria 12.5% is applied to the second tranche. The third tranches for the
with its fiscal terms and instruments are imposed so as to be able onshore, shallow, and deepwater terrains assume the rates as
to estimate front-loaded takes, tax calculations, and the proposed specified for the level. The first tranche attracts a rate of 5% irre-
royalty-design-specification applications and impacts. spective of the terrains under operation.

Variables Stochastic Distribution Minimum Mean Maximum

Oil recovery (%) Normal 32.4 50 68.5

Well rate (thousand BOPD) Normal 0.59 0.95 1.31

Discount rate (%) Log-normal 10.03 10.99 27.82

Initial oil price (USD/STB) Triangular 27.06 32.31 39.89

Oil price cap (USD/STB) Triangular 40.11 49.95 59.76

Capex (%)* Triangular 15 19 25


*Capex investment is a function of STOIIP, oil recovery, and oil price.

Table 5—Life-cycle project stochastic distribution.

July 2016 SPE Economics & Management 71

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 72 Total Pages: 10

Terrain: Onshore
25.00

Effective Royalty Rate (%)


20.00

15.00

10.00

5.00

0.00
3,000
6,000
9,000
12,000
15,000
18,000
21,000
24,000
27,000
30,000
33,000
36,000
39,000
42,000
45,000
48,000
51,000
54,000
57,000
60,000
63,000
66,000
69,000
72,000
75,000
78,000
81,000
84,000
87,000
90,000
93,000
96,000
99,000
-

Average Daily Production (BOPD)


Ons_Log Ons_Lin Ons_Jum
(a)

Terrain: Onshore
25.00

20.00
Effective Royalty Rate (%)

15.00

10.00

5.00

0.00
-
0

0
0

0
0

0
0

0
0

0
0

0
0

10 0
00
50

00

50
00

50
00

50
00

50

00
50

00

50

00
50

00
50

00

50
,0
1,

1,
2,

2,
3,

3,
4,

4,

5,
5,

6,

6,

7,
7,

8,
8,

9,

9,

Average Daily Production (BOPD)

Ons_Log Ons_Lin Ons_Jum


(b)

Fig. 2—(a) Onshore effective royalty rates, q £ 100 thousand BOPD; (b) onshore effective royalty rates, q £ 10 thousand BOPD.
Log 5 logarithmic; Lin 5 linear; Jum 5 jumping; Ons 5 onshore.

Within the specified variables, Fig. 2a shows the effective roy- remaining the same. At this last tranche, it is observed that the
alty rate for onshore terrain with jumping, linear, and logarithmic royalty rates progress in the logarithmic trend and become parallel
scales. The linear-scale royalty scheme linearly progresses with a to the upper royalty rate, Rk ð/Þ, but never reach it.
slope of a1 ¼ 0:000025 in the second tranche from an effective Fig. 2b reveals clearly that for the second tranche of the
rate of 5.0–12.5%. Ironically, it progresses with a different linear onshore terrain, the linear sliding scale gives a lower effective
slope in the third tranche. This could be attributed to the upper royalty rate up to a production rate of qd < 3; 000 BOPD, within
royalty rate defined for the third tranche, which is not linearly pro- the assumed specifications. This could be attributed to the linear
portionate with the rates between the first and second tranches. If rise that is less steep in the linear-scale mechanism, unlike the
a1 were to be the slope, it is expected to have an upper royalty steeper rise in rate associated with the logarithmic plot at the early
rate of Rk ð/Þ ¼ 20:0%, leading to a proportionate change stage of production. Consequently, marginal operators with pro-
between tranches, with other conditions remaining the same. Sub- duction rates less than 3,000 BOPD would tend to opt for the lin-
sequent production after the last tranche attracts a constant rate of ear sliding-scale mechanism rather than the others, with other
22.0%, as specified. conditions remaining the same.
Logarithmic sliding-scale royalty rates for the onshore terrain In Fig. 3, the shallow-terrain linear-scale royalty scheme
in Fig. 2a produce a logarithmic trend when plotted. It progresses ideally exhibits a linear progression with a slope of a1 ¼
logarithmically with a slope of a2 for the second tranche. It is 0:000005 for both the second and third tranches from an effective
observed that the royalty rate, Rj ð/Þ ¼ 12:5%, specified for the rate of 5.0–20.0%. This could be credited to the upper royalty rate
second tranche is not reached at the specified upper production defined for the third tranche, which is linearly proportionate with
rate, qj ¼ 5; 000 BOPD. The third tranche follows the same trend the rates between the first and second tranches. As highlighted in
but with a slope of a3 ¼ Rj ð/Þ  Rk ð/Þqj , with other conditions the onshore-terrain case, with a1 being the slope, the expected

72 July 2016 SPE Economics & Management

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 73 Total Pages: 10

Terrain: Shallow Waters


25.00

20.00

Effective Royalty Rate (%)


15.00

10.00

5.00

0.00
3,000
6,000
9,000
12,000
15,000
18,000
21,000
24,000
27,000
30,000
33,000
36,000
39,000
42,000
45,000
48,000
51,000
54,000
57,000
60,000
63,000
66,000
69,000
72,000
75,000
78,000
81,000
84,000
87,000
90,000
93,000
96,000
99,000
-

Average Daily Production (BOPD)

Sha_Log Sha_Lin Sha_Jum

Fig. 3—Shallow effective royalty rates, q £ 100 thousand BOPD. Log 5 logarithmic; Lin 5 linear; Jum 5 jumping; Sha 5 shallow.

upper royalty rate would be Rk ð/Þ ¼ 20:0%, as in this case. This tions. This could also be attributed to the uniform rise that is less
leads to a proportionate change between tranches, with other con- steep in the linear-scale mechanism, compared with the steeper
ditions remaining the same. Subsequent production after the last rise in rate associated with logarithmic plot at the early stage of
tranche attracts a constant rate of 20.0%, as specified. production. At qd ¼ 15; 000 BOPD, the equilibrium effective
The logarithmic sliding-scale royalty rates for the shallow ter- royalty rate for both linear and logarithmic sliding scales is
rain in Fig. 3 progress logarithmically with a slope of a2 for the sec- Rj ð/Þ ¼ 10%. Consequently, marginal field operators with pro-
ond tranche. It is also observed that the royalty rate, duction rates less than 15 MBOPD would tend to opt for the linear
Rj ð/Þ ¼ 12:5%, specified for the second tranche is not reached at sliding-scale mechanics rather than the others, with other condi-
the corresponding upper production rate, qj ¼ 20; 000 BOPD, that tions remaining the same.
is specified. The second tranche reached an effective royalty rate of However, because the limit of production rates is qd  qj ¼
Rj ð/Þ ¼ 10:63% at qj ¼ 20; 000 BOPD. The third tranche follows 100; 000 BOPD for the study, only two tranches for deepwater are
the same trend but with a slope of a3 ¼ Rj ð/Þ  Rk ð/Þqj , within observed in Fig. 4. As a result, the linear progress for the second
the assumed variables. At this last tranche, it is also observed that tranche has a slope of a1 ¼ 0:0000015 and rises linearly from a
the royalty rates progress in the logarithmic trend and become par- rate of 5.0 to 12.5%. There is no observed change in slope
allel to the upper royalty rate, Rk ð/Þ, but never reach it. because of the limit to two tranches. However, if the limit is
With the uniform linear sliding-scale slope for both the second extended beyond qd > qj ¼ 100; 000 BOPD, a change in slope for
and third tranches, the impact of the lower effective royalty rates the next tranche is expected and it will be observed to be steeper
from the linear sliding scale is more distinct for the second than the previous tranche. This could be attributed to the upper
tranche of the shallow terrain than the onshore terrain. The linear royalty rate defined as Rk ð/Þ ¼ 18:5% for the third tranche, which
sliding scale gives a lower effective royalty rate up to a produc- is not proportional with the rates between the first and second
tion rate of qd < 15; 000 BOPD, within the assumed specifica- tranches. As discussed previously, with a1 being the slope, the

Terrain: Deepwater
14.00

12.00
Effective Royalty Rate (%)

10.00

8.00

6.00

4.00

2.00

0.00
3,000
6,000
9,000
12,000
15,000
18,000
21,000
24,000
27,000
30,000
33,000
36,000
39,000
42,000
45,000
48,000
51,000
54,000
57,000
60,000
63,000
66,000
69,000
72,000
75,000
78,000
81,000
84,000
87,000
90,000
93,000
96,000
99,000
-

Average Daily Production (BOPD)

Deep_Log Deep_Lin Deep_Jum

Fig. 4—Deepwater effective royalty rates, q £ 100 thousand BOPD. Log 5 logarithmic; Lin 5 linear; Jum 5 jumping; Deep 5
deepwater.

July 2016 SPE Economics & Management 73

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 74 Total Pages: 10

Life-Cycle Project Case Study


18.00 18,000
16.00 16,000

Production Rate (BOPD)


14.00 14,000

Royalty Rate (%)


12.00 12,000
10.00 10,000
8.00 8,000
6.00 6,000
4.00 4,000
2.00 2,000
0.00 -
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
R%_Norm R%_Log R%_Lin R%_Jum BOPD

Fig. 5—Life-cycle project-royalty-rates profile. R%_Norm 5 existing jumping and fixed rate; Log 5 logarithmic; Lin 5 linear;
Jum 5 jumping.

expected upper royalty rate should be Rk ð/Þ ¼ 20:0%, leading to 200 m) with fixed royalty rate in the existing PSC, an undis-
a proportional change between tranches, with other conditions counted government take of 72.25% yielding a government NCF
remaining the same, but it is lower in the specified case. Subse- of USD 1875.74 million is achieved, whereas undiscounted gov-
quent production after the last tranche would attract a constant ernment takes of 71.40, 71.36, and 71.75% with corresponding
rate of 18.5%. government NCF of USD 1853.71 million, 1852.67 million, and
The logarithmic sliding-scale royalty rates for the deepwater 1862.99 million were achieved for the logarithmic, linear, and
terrain in Fig. 4 also progress logarithmically with a slope of a2 jumping royalty-rate schemes, respectively. Although the percent
for the second tranche. It is also observed that the royalty rate, change in effective royalty rate reduced by approximately 38, 34,
Rj ð/Þ ¼ 12:5%, specified for the second tranche is not reached at and 25% for the logarithmic, linear, and jumping rate, respec-
the corresponding upper production rate, qj ¼ 100; 000 BOPD, tively, with respect to the fixed rate of 16.67% of the existing PSC
that is specified. The second tranche reached an effective royalty (Fig. 5), the observed-percent-change decrease in government
rate of Rij ð/Þ ¼ 8:75% at qj ¼ 100; 000 BOPD. If a third tranche NCF is very minimal, at approximately 1.2% for both logarithmic
is introduced, extending the production rates beyond 100,000 and linear scales and 0.7% for jumping scale. The possible mini-
BOPD, it would also follow the same trend but with a slope of mal change in government NCF could be attributed to higher tax-
a3 ¼ Rj ð/Þ  Rk ð/Þqj , with other conditions remaining the same. able income and profit oil for the nonfixed rates, thereby
At this last tranche, it is noted that the royalty rates progress loga- guaranteeing more NCF for the government at the late stage of
rithmically and become parallel to the upper royalty rate, Rk ð/Þ, the project, with other conditions remaining the same.
but never reach it. Table 6 shows higher IRR and NPV for the sliding scale rates
The observed effect of lower royalty rates in deepwater terrain over the fixed rate. With a discount rate of 10%, the interest rates
shows that the logarithmic sliding scale has a lower effective roy- earned from the project for the logarithmic, linear, and jumping
alty rate throughout the second tranche, contrasting the onshore scales are 13.72, 13.74, and 13.41% respectively. These rates are
and shallow terrains where linear sliding scale gives lower effec- higher than the fixed-rate IRR of 12.82%. Although the eval-
tive royalty rates than the logarithmic scale to a certain production uated IRRs are higher than the discount rate, the marginal effi-
rate. The effect of the effective royalty rate in deepwater is quite ciency of capital is highest with the linear royalty-rate
marginal up, to an additional 3,000 BOPD in the second tranche, specification, within the assumed variables. This could be as a
before becoming more pronounced afterward, within the assumed result of much of the cumulative production being in the fewer-
variables. Consequently, deepwater producers with prolific basins than-15,000 BOPD production-rate periods, where the applica-
and nonmarginal production rates would prefer the logarithmic tion of linear royalty rates is more evident (Fig. 5), with other
royalty-rate scheme over the others. conditions remaining the same. The equity rate of return, which
is the growth rate of return, also shows a higher rate for the lin-
ear scale over the other rates and above the discount rate for all.
Deterministic and Stochastic Despite the heavy discounting effect as a result of more than 2
Petroleum-Project Analysis decades of production, the earning power and reinvestment rate
Applying the previously discussed royalty-rate-design mechanics are still encouraging.
to a typical life-cycle oil and gas project, Fig. 5 clearly substanti- Table 6 also shows royalty revenue in favor of linear, logarith-
ates the predicted behavioral patterns of the proposed royalty-rate mic, jumping, and existing (fixed) in increasing order, but a
application, with other conditions remaining the same. It is reverse order in terms of profit oil to government and tax revenue.
observed that, for a typical marginal operator within the shallow This expectation could be explained with respect to more AGR to
terrain, the linear royalty rate will produce the least rate whenever be shared when the royalty value is lower than when it is higher,
the production rate falls lower than the threshhold (qd < thereby leading to more proportionate profit oil to government
15; 000 BOPD). This is seen at the early-buildup and late-decline and tax economic rents. The higher profit oil to government and
stages of the field life. Fig. 5 shows a typical production profile tax for the lower sliding royalty rates is a progressive economic-
imposed over the different royalty-rate trends and the existing rent-extraction mechanism because it is profit-based and after cost
jumping and fixed pattern. recovery. This profit-based mechanism reveals a higher degree of
For evaluating economics-system-performance measures of earning to the government that is tied to profit; it is highest for the
the life-cycle project, Table 6 shows deterministic results for an linear scale with lowest nonprofit-based royalty and decreases in
existing PSC with a fixed royalty rate for the terrain under consid- the aforementioned order. The adoption of sliding-scale royalties
eration and corresponding economic indicators with the applica- in the petroleum-contract agreement, against fixed rate, could en-
tion of the proposed sliding-scale royalty schemes: logarithmic, courage investment because it leads to high AGR and early dis-
sliding, and jumping. For the imposed water depth (fewer than counted payout.

74 July 2016 SPE Economics & Management

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 75 Total Pages: 10

Existing Logarithmic Linear Jumping

Undiscounted government take 72.25% 71.40% 71.36% 71.75%

Discounted government take 87.21% 83.70% 83.63% 84.92%

Contractor IRR 12.82% 13.72% 13.74% 13.41%

Discounted payout (years) 11.18 10.34 10.36 10.60

NPV (USD million) 81.53 103.89 104.31 96.11

AGR 83.33% 90.66% 91.01% 87.58%

Growth rate of return 10.53% 10.66% 10.67% 10.62%

Profitability index 1.13 1.16 1.16 1.15

Royalty (USD million) 642.05 359.65 346.29 478.61

Profit oil to government 193.55 306.51 311.86 258.93

Tax (USD million) 968.51 1,115.92 1,122.90 1,053.83

Government NCF (USD million) 1,875.74 1,853.71 1,852.67 1,862.99

Table 6—Deterministic performance indicators.

Table 7 shows a summary of stochastic economic indicators of the linear scale over the other schemes, as seen in Table 8 for
for the mean, 50th (P50), and 90th (P90) percentile. It is observed production rates less than or equal to 15,000 BOPD, with other
that the mean and P50 IRR values for the sliding royalty rates are conditions remaining the same.
the same (12.70%) and greater than the fixed and discount rate. Table 8 shows Monte Carlo simulation result revealing the
This means, despite the inherent risks, that there is a 50% chance probabilistic finding cost of the field within the assumed variables.
or less of obtaining a mean value greater than the hurdle rate, There is a 90% certainty level of obtaining reserves between 104
thereby adding value to the investment. The IRR value increases to 145 million bbl with approximate finding cost between USD 15
for P90. The corresponding effect on NPV reveals a higher P50 and 21/bbl, and a mean estimate of approximately 125 million bbl
value of USD 65.82 million for the linear scale than the logarith- in reserves. There is also a 90% certainty level of obtaining
mic (USD 65.71 MM) or jumping scale. This marginal NPV between 44 and 100% of cumulative production less than and
increase could be credited to the larger proportion of lower rates equal to 15,000 BOPD; the mean value of obtaining cumulative

Existing Logarithmic Linear Jumping

NPV (USD million) Mean 27.13 59.36 59.35 47.32

P50 33.87 65.71 65.82 53.97

P90 95.39 124.87 123.87 112.98

IRR (%) Mean 11.81 12.70 12.70 12.36

P50 11.78 12.70 12.70 12.36

P90 13.56 14.52 14.49 14.15

AGR (%) Mean 83.33 90.64 90.93 87.51

P50 83.33 90.62 90.93 87.59

P90 83.33 90.90 91.59 87.67

Royalty (USD million) Mean 881.03 494.97 479.35 660.02

P50 877.23 492.65 476.06 656.51

P90 1,032.27 582.51 569.96 776.03

Tax (USD million) Mean 1,198.02 1,398.42 1,406.58 1,312.2

P50 1,186.93 1,386.87 1,394.83 1,301.34

P90 1,469.97 1,693.17 1,699.64 1,596.76

Government NCF Mean 2,408.32 2,374.11 2,372.87 2,387.42


(USD million)
P50 2,390.93 2,356.52 2,355.07 2,370.17

P90 2,856.66 2,819.05 2,817.1 2,833.33

Table 7—Stochastic economic indicator (mean, P50, and P90).

July 2016 SPE Economics & Management 75

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 76 Total Pages: 10

P90

Minimum Maximum Mean

% Cumulative production (≤ 15 MBOPD) 44% 100% 61%

% Cumulative production (> 15 MBOPD) 0% 56% 39%

Peak rate (BOPD) 14,317 19,884 17,903

Reserves (million bbl) 104 145 125

Average well cost (USD million) 21 32 25.85

Unit technical cost (USD/bbl) 15.66 20.75 18.04

Table 8—Stochastic reservoir and technical results (90% certainty level and mean).

production at a rate less than or equal to 15,000 BOPD is 61%, linear sliding scale, if properly structured, will actually encourage
whereas a mean value of 39% will result in cumulative production bidding and increase competition for marginal-field concession.
at rates greater than 15,000 BOPD. The estimated-mean peak-pro- Generally, it could be inferred that the linear sliding scale would
duction rate is approximately 17,900 BOPD, which is close to the perform better than the logarithmic sliding scale up to production
estimated deterministic value. rates that equal the range ðqj  qi Þ thousand BOPD in a three-
Consequently, the estimated mean, P50, and P90 royalty reve- tranche mechanism for both onshore and shallow terrains, with
nues are lowest for the linear royalty rate in comparison with the other conditions remaining the same.
other schemes. Conversely, the accrued tax as a result of the linear Another objective of the royalty-rate schemes is to lessen the
sliding-scale royalty approach is highest in comparison with the risk associated with speculative bidding and possible failure to de-
logarithmic, jumping, and existing fixed schemes. This could be velop smaller fields and/or early shutdown of production (Leland
attributed to the higher mean and P50 AGR of the linear scale et al. 1974; Lohrenz et al. 1981). This could be achieved by crys-
over the other schemes. tallizing the effects of the various types of royalty schemes on any
field development and possible switching of royalty schemes as
Conclusions production progresses to engender marginal-field development,
especially in the case of dwindling oil prices by mineral owners
A progressive royalty framework is presented in this paper with even as investors continue the cost-cutting measures of production
different royalty mechanics for oil- and gas-development econom- operations. As a result, mineral owners are encouraged to defer
ics. To ensure efficiency and effectiveness, the paper recommends perceived higher royalty rates by allowing the possible switching
an optimal/ideal production-rate boundary in a three-tranche ter- of royalty schemes at the early stage of development. The per-
rain as defined in the PIB (Table 3) to be qk , as defined in Eq. 6. ceived surrendered royalty would most likely be recouped at the
This eliminates ambiguity in the treatment of an additional later stage of production by other rent-extraction means such as
tranche with the application of linear sliding-scale royalty taxes; however, it becomes progressive because the extraction
mechanics. The effects of the various kinds of royalty schemes means would be profit-based, leading to a win-win situation for
evaluated show an interesting interplay that suggests a possible both the mineral owners and investors.
switch from one royalty scheme to the other as a plausible alterna-
tive toward the sustenance of future production. As deterministic
and stochastic result analysis portray, switching to a lower roy- Nomenclature
alty-rate scheme in the case of dwindling oil prices would result
d ¼ water depth, m
in greater access to gross revenue to be shared between stakehold-
g ¼ terrain
ers. The switch does not necessarily entail revenue loss on the
GRt ¼ gross revenue
part of mineral owners, but is a means of tying a larger portion of
q ¼ average daily production, BOPD
the economic rent to a profit-based mechanism. This could prob-
ROYt ¼ royalty
ably encourage sustained exploration and exploitation of the
R(/) ¼ royalty rate, %
resources and guarantee the accruing of future revenues by such a
a ¼ slope parameter
petroleum-producing nation.
Contrary to the perception that the logarithmic sliding-scale is
better than the linear sliding-scale system in view of fostering
competition (Patterson 1979), it is observed that the linear sliding Acknowledgments
scale for shallow-water terrain is better for a field producing at The authors sincerely acknowledge the research assistance pro-
most 15,000 BOPD in terms of lower effective royalty rate than vided by the Chirota and Emmanuel Egbogah Distinguished Pro-
logarithmic scheme. The linear sliding scale also performed better fessors Fund toward the presentation of this work at the 2015 SPE
than the logarithmic scale for a field producing at most 3,000 Annual Technical Conference and Exhibition in Houston. The
BOPD in onshore terrain, within the assumed variables. Thus, the support of the World Bank African Center of Excellence is
conclusion offered by Patterson (1979) is conjectural in the sense acknowledged as well. Several comments received from the par-
that it applies only to shallow-water production greater than ticipants at the 2015 SPE Annual Technical Conference and Exhi-
15,000 BOPD and onshore terrain greater than 3,000 BOPD, bition are greatly appreciated. All conclusions and errors are
within the assumed variables. However, for deep offshore terrain, those of the authors.
the conclusion relatively applies for all production rates.
Furthermore, it has been opined that the logarithmic sliding
system encourages the development of marginal prospects better References
than the linear sliding-scale system (Patterson 1979). Currently, Africa Oil and Gas Report (AOGR). 2014. Africa Oil and Gas Report 15
many marginal operators in Nigeria produce fewer than 15,000 (10).
BOPD; this implies that the linear sliding-scale system will en- Deep Offshore and Inland Basin Production Sharing Contracts Act
courage the development of marginal fields more than the loga- (DOIBPSCA). 1999. Deep Offshore and Inland Basin Production
rithmic sliding-scale system. As a result, it is inferred that the Sharing Contract Decree No. 9 of 1999.

76 July 2016 SPE Economics & Management

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004


EM174846 DOI: 10.2118/174846-PA Date: 20-July-16 Stage: Page: 77 Total Pages: 10

Echendu, J. C., Iledare, O. O. and Onwuka, E. I. 2012. Comparative Per- PwC. 2015. Nigeria Data Card 2014/2015, www.pwc.com/ng (accessed 5
formance Analysis of Petroleum Sharing Contracts in Angola, Equato- May 2015).
rial Guinea, Gabon and Nigeria. Presented at the SPE Nigeria Annual
International Conference and Exhibition, Lagos, 6–8 August. SPE-
162990-MS. http://dx.doi.org/10.2118/162990-MS. Joseph C. Echendu is a research fellow with the Chirota and
Echendu, J. C., Onwuka, E. I. and Iledare, O. 2014. Spreadsheet Modeling Emmanuel Egbogah Distinguished Professorship at the Emer-
and Simulation Analysis of Production Sharing Contract Terms and ald Energy Institute and a World Bank Africa Centre of Excel-
Instruments in Nigeria. Presented at the SPE Nigeria Annual Interna- lence Scholar at the University of Port Harcourt, Nigeria. He is a
tional Conference and Exhibition, Lagos, 5–7 August. SPE-172389- consultant and facilitates field-development plans and petro-
MS http://dx.doi.org/10.2118/172389-MS. leum-economics workshops on behalf of Shell Petroleum De-
velopment Company and the Department of Petroleum
Echendu, J. C., Iledare, O. O. and Onwuka, E. I. 2015. Comparative Eco-
Resources and Nigerian National Petroleum Corporation in Ni-
nomic-Performance Analysis of Production-Sharing Contracts in geria. Echendu has more than 10 years of oil and gas experi-
Angola, Equatorial Guinea, Gabon, and Nigeria. SPE Econ & Mgmt 7 ence in practice and research. His current interests are
(4): 157–166. SPE-162990-PA. http://dx.doi.org/10.2118/162990-PA. petroleum/energy-economics modeling, life-cycle project
Iledare, O. O. 2011. Advanced Petroleum Economics. Lecture, African design and analysis, policy and strategy research, petroleum-
University of Science and Technology, Abuja, Nigeria. fiscal-systems design, and stochastic analysis. Echendu has
Iledare, O. O. 2014. Upstream Petroleum Economic Analysis: Balancing authored or coauthored more than 10 technical papers. He is
Geologic Prospectivity with Progressive, Stable Fiscal Terms and an active member of SPE, is a technical editor of SPE Econom-
Instruments. The Way Ahead 10 (1): 28–30. SPE-1114-028-TWA. ics & Management, and is currently a member of the SPE Soft
Skills Committee. Echendu is the 2016 SPE Africa Regional Pa-
http://dx.doi.org/10.2118/0114-028-TWA.
per Contest Winner in the PhD degree category. He holds a
Iledare, O. O. and Echendu, J. C. 2013. A Comparative Analysis of the bachelor’s degree in chemical engineering from the University
Impact of Production Sharing Contract Terms on Deepwater E&P Eco- of Lagos and a master’s degree in petroleum engineering
nomics in the Atlantic Gulf of Guinea. Oral presentation given at the with a specialty in petroleum economics from the African Uni-
4th Latin American Energy Economics ELAEE, Montevideo, Uru- versity of Science and Technology, Nigeria.
guay, 8–9 April.
Omowunmi O. Iledare is the Chirota and Emmanuel Egbogah
Johnston, D. 2003. International Exploration Economics, Risk, and Con- Distinguished Professor and director of the Emerald Energy
tract Analysis. Tulsa: PennWell Corporation. Institute at the University of Port Harcourt, Nigeria. He is also
KPMG. 2014. Nigeria’s Oil and Gas Industry Brief. professor emeritus of Petroleum Economics and Policy
Leland, H. E., Norgaard, R. B. and Pearson S. R. 1974. An Economic Research at the Center for Energy Studies at Louisiana State
Analysis of Alternative Outer Continental Shelf Petroleum Leasing University. Previously, Iledare worked as the director of the
Policies. Report prepared for the Office of Energy Research and De- energy information and data division at the Center for Energy
velopment Policy, National Science Foundation, September 1974. Studies. He has more than 25 years of oil and gas experience
Lohrenz, J., Burzlaff, B. H. and Dougherty, E. L. 1981. How Policies spanning both practice and research. Iledare’s research inter-
ests are in international petroleum fiscal economics and geo-
Affect Rates of Recovery From Mineral Sources. SPE J. 21 (6):
politics of oil and gas resources. He was the president of the
645–657. SPE-9553-PA. http://dx.doi.org/10.2118/9553-PA. United States Association for Energy Economics in 2008 and
Mian, M. A. 2010. Designing Efficient Fiscal Systems. Presented at the the 2014 president of the International Association for Energy
SPE Hydrocarbon Economics and Evaluation Symposium, Dallas, 8–9 Economics. Iledare is an active member of SPE and currently
March. SPE-130127-MS. http://dx.doi.org/10.2118/130127-MS. serves as an associate editor of SPE Economics & Manage-
Mercier, D., Mount, P. B. and Willard, A. 1997. Maximizing Field Value With ment. He has facilitated short courses on petroleum econom-
a Royalty Rate That Tracks Oil Price. J Pet Technol 49 (11): 1220–1232. ics and risk analysis at SPE conferences for the SPE Nigeria
SPE-1197-1220-JPT. http://dx.doi.org/10.2118/1197-1220-JPT. Council since 1999. Iledare has also conducted petroleum ec-
Patterson, W. E. 1979. Sliding Scale Royalty and Offshore Lease Sale onomics and policy seminars and workshops on behalf of Shell
Petroleum Development Company, the Petroleum Technol-
Bidding. Presented at the SPE Hydrocarbon Economics and Evaluation
ogy Development Fund, and the Nigerian National Petroleum
Symposium, Dallas, 11–13 February. SPE-7732-MS. http://dx.doi.org/ Corporation in Nigeria. He holds a bachelor’s degree with
10.2118/7732-MS. honors in petroleum engineering from the University of Ibadan,
Petroleum Industry Bill (PIB). 2008. Inter Agency Team 2009 Petroleum Nigeria; a master’s degree in energy resources from the Uni-
Industry Bill draft. versity of Pittsburgh; and a PhD degree in mineral and energy
Petroleum Industry Bill (PIB). 2012. 2012 Petroleum Industry Bill draft. resource economics from West Virginia University.

July 2016 SPE Economics & Management 77

ID: jaganm Time: 15:09 I Path: S:/EM##/Vol00000/160004/Comp/APPFile/SA-EM##160004

You might also like