CORRESP 1 filename1.htm

Page 1

Sasol Limited

1979/003231/06

1 Sturdee Avenue Rosebank 2196 PO Box 5486 Johannesburg 2000 South Africa
Telephone +27 (0)11 441 3111 Facsimile +27 (0)11 788 5092 www.sasol.com

Directors: MSV Gantsho (Chairman) SR Cornell (Joint President & Chief Executive Officer)(American) B Nqwababa (Joint President & Chief Executive
Officer) C Beggs MJ Cuambe (Mozambican) HG Dijkgraaf (Dutch) VN Fakude (Executive) NNA Matyumza IN Mkhize ZM Mkhize MJN Njeke
PJ Robertson (British and American) P Victor (Chief Financial Officer) S Westwell (British)

Company Secretary: VD Kahla

26 July 2016

Mr Brad Skinner
Senior Assistant Chief Accountant
Office of Natural Resources
Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549

Dear Mr Skinner,

Sasol Limited Annual Report on Form 20-F for the Year Ended 30 June 2015
Filed 9 October 2015
File No. 1-31615

We refer to the further comment letter, dated 1 July 2016, from the staff of the Office of Natural
Resources (the “Staff”) relating to the response dated 7 June 2016 (the “Response”) to the Staff’s
comment letter relating to the Form 20-F (the “Filing”) of Sasol Limited (the “Company”) for the
year ended 30 June 2015. Set forth below in detail are the responses to the Staff’s further
comment letter, which have been provided in each case following the text of the comment in the
Staff’s letter.

1.

To further our understanding of your response to our prior comment eight, please provide us

with the cost figures relating to the transportation capacity reservation and asset retirement
costs in Canada and the drilling rig and FPSO contract termination costs, current well activity
costs and asset retirement costs in Gabon.

As part of your response, please clarify each cost as either a future production or development
cost, whether each cost is discretionary or subject to an existing contractual obligation, and tell
us the timing of each cost relative to the producing life of the underlying properties, e.g. the
cost is incurred prior to, during or after cessation of production.

Also provide us with a narrative explaining the nature of the activities related to each cost and
the extent that each cost is necessary for the development and production of the underlying
proved reserves.

Response
To help facilitate the explanation below, the Company would like to advise the Staff that the
Gabon reserves represented 0,1% and Canada 1,6% of Sasol’s Proved Reserves at 30 June
2015.

page 2 of 8

The cost figures relating to the transportation capacity reservation and asset retirement costs
in Canada and the drilling rig and FPSO contract termination costs, current well activity costs
and asset retirement costs in Gabon are presented in the table on the next page. All of these
costs are committed, independent of production volume and not related to any future project;
they will therefore be incurred in production of the proved reserve. These costs will be incurred
regardless if there is any future production. Accordingly, the Company believes they should
be excluded from an analysis of determining if the resources are economically producible as
that term is defined in Rule 4-10(a)(10) of Regulation S-X (“economically producible”). By
excluding these costs, the undiscounted cash flows disclosed in the table on Page G-9 would
be positive. Thus, the company believes the resources would generate revenue that exceeds
the costs of obtaining the resources.

For the purposes of presentation in the table on Page G-9 of the Filing “Asset retirement cost”
and “Well activity in progress” have been classified as future development costs whereas
“Transportation capacity reservation cost” and “Drilling rig and FPSO contract termination
cost” have been classified as future production costs. A short description of the nature of the
four cost categories are provided below.

**CONFIDENTIAL TREATMENT REQUESTED BY SASOL LIMITED, PURSUANT TO RULE
83**

Asset retirement cost – Canada and Gabon
These costs represent the cost to decommission the producing asset, comprising wells,
flowlines and facilities as appropriate, and rehabilitate the environment as required by
regulation and industry best practice. The scope of the decommissioning includes the entire
asset which is required to produce the proved reserve and in Canada increases as additional
wells are drilled to develop the field. The Company does not view these costs as discretionary
and in all cases the cost is incurred after the cessation of production and the Company does
not believe these costs should be included in the analysis to determine if the reserves are
economically producible.

Transportation capacity reservation cost – Canada
This cost is the contracted cost to reserve capacity in third-party pipelines to transport sales
gas to the delivery point at which the market price is set. It is the total cost for sixteen existing
transportation contracts with differing quantities, start dates and end dates; with the longest
running to 2032. A portion of these costs will be incurred during production and a portion after
cessation of production (depending when the field economic life ends, which varies between
the three periods). These contract costs will have to be met irrespective of whether there is
production from the asset and therefore the Company does not believe these costs should be
included in the analysis to determine if the reserves are economically producible.

Drilling rig and FPSO contract termination cost – Gabon
Development and production of our Gabon asset uses contracted drilling and production
facilities. During 2015 this comprised a jack-up drilling rig as well as the long-term leased
floating, production, storage and offloading vessel (FPSO). The drilling rig was contracted until
July 2016 but at the date of the Filing there was insufficient activity to utilise the drilling rig to
the end of the contract; a termination charge would therefore be payable. The relevant
contract termination charge, which is not dependent on future production volume, is therefore
attributed to the proved reserve at the time it would be incurred. Similarly the FPSO is leased
until 2020 with a one-year notice period. According to our modelling at 30 June 2015

page 3 of 8

production of the proved reserve would end around 30 December 2016 and a one-year
termination charge was therefore recognised at that time and the costs is not dependent upon
production and therefore the Company does not believe these costs should be included in the
analysis to determine if the reserves are economically producible.

Well activity in progress – Gabon
Costs incurred in the process of developing our Gabon asset can be recovered against future
production under the terms of the production sharing contract (so-called “cost oil”). At 30 June
2015 there were costs already incurred in the drilling and completion of wells which had not
been billed by the Operator of the asset and therefore had not been recognised in the official
cost oil calculation. To ensure correct calculation of historical and future oil entitlement for the
Company these costs were reflected as future development costs. As the activities giving rise
to the costs had actually been performed the costs were not discretionary and the Company
was committed to pay them under the terms of the Joint Operating Agreement. The costs are
independent of the magnitude of production and will be payable even if there is no future
development, and therefore the Company does not believe these costs should be included in
the analysis to determine if the reserves are economically producible.

2.

For each of the three forecasts used to develop your crude oil price assumptions, tell us the

specific time periods covered by the forecast and the specific forecast prices for each of those
periods. Provide similar information for each of the four forecasts used to develop your natural
gas price assumptions.

Response

Crude Oil
The long-term average crude oil price is presented on a nominal basis in Note 38 to the
Annual Financial Statements.

Crude oil prices, utilised in impairment testing, are derived using forecasts from three external
data analysis and research consultancies (Wood Mackenzie, PIRA Energy and IHS
Incorporated).

The outlooks for each of the consultants are reflected in the graph below:

page 4 of 8

Equal weighting has been given to each of the external consultant’s views in developing the
crude oil price assumptions. Management believes that utilising an average of three forecasts
will provide a higher degree of accuracy than any one assumption. The consultants’ June
2015 price forecasts were used to derive the crude oil assumptions and were considered to
reflect the market and external prices at the reporting date i.e. 30 June 2015.

The price assumptions extracted from the panel of consultants were for the period 2016 to
2025. Due to the nature of Sasol’s operations, major activities and capital expenditure
extending over long periods of time, our budgeting processes extends over 10 years. The
average cash flows are extrapolated over the life of the asset to calculate the terminal value,
as the affected assets are matured and fully developed and the cash flow stream can be
reliably estimated

Natural gas (real)
The long-term average gas price is presented on a real basis, excluding inflation, in Note 38 to
the Annual Financial Statements.

Natural gas prices, utilised in impairment testing, are derived from a combination of forecasts
by external consultants (Wood Mackenzie, PIRA Energy, IHS Incorporated and McDaniel &
Associates Consultants Ltd. (McDaniel)).

The future of the US shale gas industry is extremely uncertain. As a result, there is a very high
variability in any future projections. Accordingly, in developing forecast views, a number of
external consultants’ assumptions are taken into account.

The Canadian shale gas asset is in the process of being developed and accordingly a 10 year
budget would be inappropriate as the terminal value cannot be accurately determined.
Accordingly, the life of field to 2040 is used as the forecast period for impairment testing.

The outlooks are reflected in the graph below:

Explain to us in greater detail the reasons why you believe it is appropriate to incorporate
natural gas price forecasts from McDaniel in developing your natural gas price assumptions.

page 5 of 8

Additionally, explain, in reasonable detail, how you determined the relative weighting between
the four different natural gas price forecasts.

Response

McDaniel provides Sasol with independent, third party assurance in terms of the Canadian
reserves. Based on the significant volatility in the market, management believes that utilising
multiple price forecasts from leading independent consultants, who provide expertise in the
energy industry will provide a higher degree of accuracy than any one assumption.

There is currently no guidance or rules which dictate the optimal weighting of price estimates.
Accordingly, management considered the following factors in calculating the weighting
percentages:

•

Expertise in the Canadian oil and gas market – Canadian gas competes in the US gas
market and is therefore exposed to US supply and demand balances. The adjustment for
AECO therefore incorporates the logistical issues of being a Canadian operator. The panel
of three consultants provide expertise in the Energy industry mainly in the US and to a
lesser extent in Canada and therefore is a credible source for calculating AECO. McDaniel
is a leading independent reserve evaluator and resource assessor in the Canadian oil and
gas industry. Based on McDaniel’s Canadian experience, we believe that the McDaniel
price forecasts are more reflective of the Canadian oil and gas industry. Sasol believes
that by having a price forecast representative from the US and Canadian market would
provide a balanced and reasonable reflection of future pricing.

•

Comparability with peers – Recent research and surveys indicate that Canadian oil and
gas companies generally use price forecasts determined by their reserve engineers.

•

Forecasting period – The Canadian shale gas asset’s full field development plan
currently extends to beyond 2040, however, as indicated in the graph below, the McDaniel
price deck forecasts extends to only 2029 with a standard inflation percentage of 2% being
applied to each year thereafter. McDaniel recommends that a 2% inflation assumption is
applied to prices for periods post 2029.

Accordingly, based on the assessment above, management concluded that using a weighting
of 70/30 (70% for McDaniel and 30% for the others on a combined basis) would result in the
best allocation in the circumstances. The Company believes that this weighting appropriately
reflects the asset’s value.

To illustrate the impact of a different weighting, the Company calculated the recoverable
amount using a weighting of 80/20 and 60/40 and the results thereof indicated that the
impairment would increase or decrease by CAD45 million which represents 1% of profit before
tax. This movement was not considered to be material.

Explain to us, in reasonable detail, how the pricing forecasts described in your response were
used to develop the disclosed long-term average crude oil and natural gas assumptions. As
part of your response, tell us the number of years covered by the cash flow projections
underlying your value-in-use calculations, how the price assumption for each year was
determined and, beginning with the most recent year in your projections, the specific crude oil
and natural gas prices used for each year in your projections.

page 6 of 8

Response

Crude oil (presented in nominal terms)
As noted above, Sasol uses an average of crude oil price forecasts from three external data
analysis and research consultancies (Wood Mackenzie, PIRA Energy and IHS Incorporated)
for impairment testing.

Crude oil is a global market price and accordingly, Sasol assumes that the average of price
forecasts from three external data analysis and research consultancies would be a more fair
reflection of the future price.

Panel of 3

consultants

FY Period

Dated Brent

FY16

68,18

FY17

79,64

FY18

84,63

FY19

88,53

FY20

92,77

FY21

96,85

FY22

101,36

FY23

106,16

FY24

111,08

FY25

116,48

The long-term average oil price was calculated for the 10 year budget period from FY16 to
FY25 as $94,57.

Natural gas (Real)
As noted above, Sasol’s natural gas prices are derived from a combination of forecasts by
external consultants (Wood Mackenzie, PIRA Energy, IHS Incorporated and McDaniel &
Associates Consultants Ltd. (McDaniel)). The McDaniel price deck is assigned a weighting of
70%, whilst the average of the other three external experts are weighted at 30%.

Financial

year

Panel of

three

McDaniel

Weighted

30/70

FY16

2,96

2,90

2,92

FY17

3,40

3,24

3,29

FY18

3,43

3,51

3,49

FY19

3,53

3,63

3,60

FY20

3,83

3,74

3,77

FY21

3,89

3,89

3,89

FY22

4,05

4,13

4,11

FY23

4,25

4,27

4,26

FY24

4,26

4,35

4,32

FY25

4,25

4,35

4,32

FY26

4,25

4,35

4,32

FY27

4,38

4,34

4,35

FY28

4,57

4,34

4,41

page 7 of 8

FY29

4,71

4,33

4,44

FY30

4,75

4,36

4,48

FY31

4,90

4,36

4,52

FY32

4,90

4,36

4,52

FY33

4,91

4,36

4,53

FY34

4,92

4,36

4,53

FY35

4,92

4,36

4,53

FY36

4,94

4,36

4,53

FY37

4,96

4,36

4,54

FY38

4,97

4,36

4,54

FY39

4,98

4,36

4,55

FY40

5,00

4,36

4,55

Given the nature of development of the Canadian shale gas asset and the low gas price, we
expect a prolonged development plan. Accordingly, in estimating the future cash flows for
impairment purposes, the life of reserve was used. The forecasting period therefore extended
from FY16 to FY40. The average gas price, using the weighting of 70% to McDaniel and 30%
to the average of the other three external experts for the period FY16 to FY40 amounted to
$4,25.

However, as noted above, management expects the asset to be fully developed only from
2020 and hence to reflect the long-term gas price in Note 38 of the Annual Financial
Statements, Sasol utilised the pricing forecasts from 2020 to 2040. This yielded an average
long term gas price of $4,40. This was considered appropriate given the high levels of volatility
in short-term prices, as well as the long-term nature of the asset. The pricing assumptions
noted above were applied in full within the calculations in the impairment model.

The Company will clarify the disclosure in all future filings, beginning with the annual report on
Form 20-F for the year ended 30 June 2016, to state that the assumption relates to the period
2020 to 2040. Refer to note 38 in the Annual Financial Statements for other related
assumptions.

Regarding the disclosed long-term average rand/US$ exchange rate, address the following:

•

•
•

Explain to us how the impact of the exchange rate was reflected in the cash flow
projections underlying your value-in-use calculations;
Tell us how the long-term average exchange rate was determined; and,
Beginning with the most recent year in your cash flow projections, tell us the
specific exchange rate used for each year in your projections

Response:

The exchange rate projections provided in the table below are incorporated in the cash flow
models of some the underlying cash generating units to convert US$ revenues and/or costs to
rand. This would primarily apply to South African cash generating units who import and export
in US dollars. The rand/US$ exchange rate assumption is determined by the Company’s in-
house economist. We have an Assumptions Review Meeting quarterly to review key
assumptions (including exchange rate assumptions) against the current economic context.
These assumptions are revised, where appropriate, to ensure alignment with other external
consultancies.

page 8 of 8

Generally, our long term rand/US$ assumption is that the currency will depreciate in line with
the inflation differential between South Africa and the United States (i.e. the principle of
relative purchasing power parity holds). Our assumed long term inflation rate for South Africa
is 5,5% and our assumed long term USA inflation rate is 2%. Deviations from the long term
purchasing power parity depreciation rate can be explained by our estimates on the current
level of over/undervaluation of the rand/US$ exchange rate, assumptions and judgements on
country specific factors that may influence the growth outlook, interest rate and inflation
differentials, commodity price movements, socio-economic risk factors, sovereign credit
ratings and general emerging market risk sentiment (especially in the short- to medium-term).

Financial

year

USD/ZAR

FY16

12,15

FY17

12,45

FY18

12,25

FY19

12,25

FY20

12,60

FY21

13,10

FY22

13,60

FY23

14,15

FY24

14,70

FY25

15,30

The long-term average rand/US$ exchange rate disclosed was calculated using the principles
described above for the period FY16 to FY25. The average exchange rate calculated is at
R13,26.

We acknowledge that:

•
•

•

The Company is responsible for the adequacy and accuracy of the disclosure in the filing;
Staff comments or changes to disclosure in response to Staff comments do not foreclose
the Commission from taking any action with respect to the filing; and
The Company may not assert Staff comments as a defence in any proceeding initiated by
the Commission or any person under the federal securities laws of the United States.

We appreciate the Staff’s review of Filing. Should the Staff have any questions or require any
additional information, please telephone the undersigned at +27-11-441-3505. My email address
is Paul.Victor@sasol.com.

Yours faithfully

/s/ Paul Victor

/s/ Bongani Nqwababa

Paul Victor

Bongani Nqwababa

Incoming Chief Financial Officer

Joint President and Chief Executive Officer