Wagging The Dog

Oil Market Outlook
Wagging the dog

In theory the oil price should be equal to the cost to find and
develop the marginal barrel of oil (the long-dated price), plus
or minus backwardation or contango. The backwardation or
contango level should be set by the oversupply or
undersupply in the current market.
But is this really how the price discovery works in real life?
Or is the long dated price more affected by price changes in
the front of the curve instead of living its own life? Could it
be that the tail is wagging the dog in the oil market?
August - 2017 - DNB Markets - Torbjrn Kjus 1

Oil Market Outlook
Content
1 Executive summary .................................................................... 3

2 Why did we miss the 2Q-2017 price decrease? ......................... 4
3 Wagging the dog ........................................................................ 8
4 Will the OPEC deal fall apart? .................................................. 12
5 Demand growth has recovered from the weak first quarter ...... 14
6 Supply vs Demand ................................................................... 19
7 Global oil inventories are drawing in 2017 ................................ 21
8 We are in phase 2 of the oil price recovery .............................. 24
9 CAPEX cuts how do they affect the oil market? .................... 26
10 Decline rates and call on shale................................................. 30
11 What oil price will be required for shale to grow enough? ........ 34
12 The 2020 bunker spec change will push Brent higher.............. 42
13 Launching a 5-year price deck in real terms............................. 49
14 Oil demand will not peak by 2025 ............................................. 51
15 Peak oil demand does not mean peak oil price ........................ 52
16 Electric vehicles will not make oil demand peak by 2025 ......... 53
17 Low unplanned outages means the risk is now bullish............. 57
18 DNB Markets Brent price forecast & consensus forecast......... 59
19 Historical Brent prices, trading range & time spreads .............. 60
20 Global supply-/demand balance DNB, IEA, OPEC, EIA........ 61
21 DNB Markets OPEC assumption & recent OPEC exports ....... 62
22 Calculated Saudi Arabia oil exports capacity ........................... 63
23 DNB Markets global oil demand assumptions by region .......... 64
24 Oil price score card for 2018 .................................................... 65
August - 2017 - DNB Markets - Torbjrn Kjus 2

Oil Market Outlook
1 Executive summary the lead will now have to run the cuts
until inventories are drawn down
The Brent price did not continue to
enough to flip the whole curve to
increase as expected in Q2. Instead of
backwardation. They have no choice,
the gradual increase by quarter that
now that the platform is on fire.
we had seen since Q1-2016 prices fell.
Our supply-demand balance suggests
The key reason was large, unexpected
0.6 million b/d stock draws in 2017
growth in production from
followed by 0.3 million b/d stock draws
Libya/Nigeria and a market (myth)
in 2018 if OPEC stays disciplined.
belief that US shale growth will destroy
Demand growth has recoverd from a
the market if the Brent price climbs
weak Q1-2017 and with OPEC
above 55 $/b. The US rig count
discipline OECD stocks will draw down
increased quicker than what we had
more than consensus expect.
anticipated and this led to massive
Backwardation is achievable by year
shorting of the oil price by speculative
end for the Brent structure as long as
money. OPEC was also to blame as
Libya/Nigeria does not continue to
the cartel cut exports less than
surprise. This would represent phase 3
production by drawing down its own
of the price recovery and will send
inventory. The market instead would
Brent to 60 $/b.
have wanted to see lower exports from
CAPEX cuts in 2014-2016 will support
OPEC and stock draws in the OECD.
the market from 2020 as a
OECD stock draws will instead be
combination of field decline and
visible in 2H-2017 now that OPEC are
demand growth makes in necessary to
cutting exports more than production
see huge shale growth in 2019 and
and with the help of a large increase in
2020. Enough growth will not be
seasonal demand.
possible with Brent below 60 $/b. IMO
Many players believe that it is possible
implements a 0.5% sulphur cap for the
to calculate the correct level for the oil
global shipping industry in 2020. The
price. This seems to be more
combination of postponed projects and
theoretical than practical. In real life
the IMO spec change will make the
the long-dated Brent price looks to be
Brent price over shoot in that year,
more affected by the price discovery in
before it falls back down on increased
the front of the market instead of living
demand- and supply elasticity.
its own life based on marginal costs. A
We are launching our first take on the
lack of liquidity further out than the
Brent price in real terms for the coming
next 6 months, a lack of natural buyers
5 years. For 2018-2022 we forecast
(airliners practically stopped hedging)
the following Brent prices in $/b: 60,
and new regulations making it difficult
65, 75, 70, and 60. Electric vehicles
for oil traders to warehouse longer
are not a threat to oil prices in this time
term positions are to blame. The tail is
frame. We argue that a combination of
wagging the dog.
a belief of peak demand and the belief
Several analysts believe the OPEC
that US shale will cover all demand
deal will collapse already before year
growth is positive for oil prices in the
end. We disagree. OPEC with Saudi in
medium term.
August - 2017 DNB Markets - Torbjrn Kjus 3

Oil Market Outlook
2 Why did we miss the 2Q- has been the key head winds for the Brent
market from Q2-2017 in addition to the
2017 price decrease? Nigerian lifting of force majure on Forcados
The Brent price development since the start of exports.
2016 was positive quarter by quarter until we
reached Q2-2017. We started Q1-2016 at the
Libyan Oil Production
bottom of the cycle with Brent averaging 35 2.0
1.8
$/b, then rising to 47 $/b in Q2/Q3-2016 and
1.6
moving up to 51 $/b in Q4-2016. When 2017 1.4
Million b/d
1.2
started we thought the oil price would 1.0
0.8
continue to trend higher by the quarter 0.6
through the year. The start was good as Q1- 0.4
0.2
2017 averaged 55 $/b, 4 $/b higher than the 0.0
Jan-01 Jan-03 Jan-05 Jan-07 Jan-09 Jan-11 Jan-13 Jan-15 Jan-17
previous quarter. Then the price increases
Source: IEA
came to an unexpected end in Q2-2017 which
averaged at 51 $/b. Brent even traded below
Other head winds for the Brent price the last
45 $/b at the end of that quarter.
3-4 months has been the surprisingly strong
growth in the US oil rig count. The rig count
ICE Brent Future First Month (USD/b)
60
has increased quicker than what we had
55
anticipated, even though production has not
50
increased quicker than what we had
45
anticipated due to well completions lagging
40
massively on the rig count. Hence for the
35
fundamental supply-demand balance the US
30
oil output has not surprised us but the rig
25
count has.
Jan2016 Apr2016 Jul2016 Oct2016 Jan2017 Apr2017 Jul2017
US Horizontal Oil Rigs Weekly Change

20 (Baker Hughes - 4 week moving average)
So what happened? What went wrong with
15
our forecast of gradually higher oil prices 10
quarter by quarter? 5
-5
The first reason why we missed the price
-10
development into Q2-2017 is that the -15
anticipation phase stopped about 10 $/b lower -20

Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17
than what we had originally thought (see later Total Oil Rigs (LHS) Anticipation
chapter on the phases of price recovery).
The increase in the rig count has hence

On the fundamental side the key negative
represented mainly psychological head winds
factor was the surprisingly strong return of
for oil prices and led to a massive shorting on
output from particularly Libya, but also to
the oil price from financial players. The
some extent from Nigeria. Libya has now
shorting of the oil market has chopped many
increased production from 550 kbd in April to
$/b of the oil price into Q2-2017. Our analysis
currently stand close to 1.1 million b/d. This

Oil Market Outlook
shows that the correlation between oil prices One important issue was probably clumsy
and financial positions are strongest for the communication from OPEC. OPEC should not
short positions during the past two years. have led the market to expect sizeable stock
draws in 1H-2017. Large stok draws could
Brent vs Total Non-Commercials crude exposure only never have materialized in 1H-2017 without
Weekly change in futures exposure MB
Weekly Change Long futures (mb): 26 much larger OPEC cuts. Why? Well, since
Weekly Change Short futures (mb): -30
Weekly Net Change Non-Commercials (mb): 56
OPEC ramped up production to record levels
Weekly Brent Price Change (USD/b): 1.3 in November/December 2016, these barrels
Correlations R-Square
R-Square Last 100 weeks Long positions 0.30 did of course hit the imports market into 2017
R-Square Last 100 weeks Short positions 0.52
R-Square Last 100 weeks Net positions 0.52 due to logistical issues and sailing time.
Statistical value of change in positions USD/b
Value of each 10 mb change in Long position 0.24
Everybody who follows the oil market should
Value of each 10 mb change in Short position 0.49 have been aware of this, but it did not seem
Value of each 10 mb change in Net position 0.23
Difference to last 25-weeks average MB like everybody actually expected this.
Diff to average net exposure last 25 weeks -117
Value of change back to average level (USD/b) 2.7
The other issue was headwinds from the

The statistical value of changes in the long demand side. Every year demand is much
positions vs the change in oil prices is weaker. stronger in the second half than in the fist half.
The fastest/hottest money seems to go in and In fact demand normally drops from the
out from the short side and hence these second half the year before into the first half
positions correlate best with the oil price short the year after.
term. The data shows that financial players
increased their short positions by 200 million Global Oil Demand By Half Years
barrels from March to June and this was one 100
of the key reasons why prices fell from Q1- 98

96
2017 into Q2-2017. The statistical value of
Million b/d
94
adding 200 million short positions by non-
92
commercials was worth a 10 $/b lower oil 90
price. This is excactly what we got. 88
86
2H-2012
1H-2013
1H-2012
2H-2013
1H-2014
2H-2014
1H-2015
2H-2015
1H-2016
2H-2016
1H-2017
2H-2017
SUM: Non-Commercials Crude Futures -

850 Short
800
This happened also in 2017 and the effect

Million barrels
750
was excacerbated by the extremely weak
700
demand numbers from India in Q1-2017,
650 caused by the demonetization in Q4-2016.
600 Indian oil demand actually fell in the beginning
550
of 2017 due to this effect. It was hence
Feb2017 Mar2017 Apr2017 May2017
extremely unfortunate for OPEC that the cuts
started into a period with negative seasonal
But the financial money reacts to news flow demand growth.
and fundamentals and what was it that
unleased this bearishness?

Oil Market Outlook
Diff YOY OPEC Crude Production vs Exports

2.0
Exports cut larger than production cut
1.5
1.0
0.5
Million b/d
0.0
-0.5
-1.0
-1.5
-2.0
Production cut larger than than exports cut
-2.5
Jan-14 Jun-14 Nov-14 Apr-15 Sep-15 Feb-16 Jul-16 Dec-16 May-17 Oct-17
Source: Platts cFlow, IEA
Source: US DOE
Taken together we agree with Raymond

The third head winds for OPEC, which they
James that there have been some myths in
have to blame themselves for, was that until
the oil market that has formed the broad
about May, the drop in production was much
narrative during 1H-2017, and has led many
larger than the drop in exports. Many OPEC
financial players to short the oil price.
countries, Saudi Arabia included, chose to
draw down their crude stocks and hence they
According to Rayond James the most
cut exports less than production.
common ten myths about the oil market are
the following:
OPEC Crude Oil Stocks 1. Weekly DOE inventory data has
(Algeria, Angola, Ecuador, Iran, Iraq, Kuwait, Nigeria, Qatar, Saudi, UAE, Venezuela)
395
looked bearish this year (on the
390
contrary; since February us oil
385
Million barrels
inventories have been drawing)

380
375
370
US Total Petroleum Inventories
1,400
365
Dec-16 Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17
1,350
1,300
Source: JODI
Million barrels
1,250
1,200
This was not what the market wanted to see 1,150

1,100
and we thought the cartel would have been 1,050
sensitive to that fact. The market wants to see 1,000
950
lower exports from OPEC countries that lead J F M A M J J A S O N D
to stock draws in OECD. This has so far not Min / Max Range 5y Average 2016 2017
happened to any meaningful extent, but will

happen in 2H-2017 as demand jumps on 2. US shale production will flood the
average 1.7 million b/d from the first half of market at 35 $/b oil price (need about
the year to the second half of the year, and as 50 $/b wellhead price to generate
OPEC now are cutting exports more than positive cash flow on full cycle)
production. Cargo tracking from Platts
suggests that OPEC exports from the
countries that are part of the cut deal from
about May and onwards have started to
surpass the production cuts.

Oil Market Outlook
Average US Shale Break Even Costs - $/b DOE data (you need to look at crude
(Weighed by volume in Bakken, Eagle Ford, Permian, Niobrara)
75 plus the main refined product stocks)
70 US Crude & 4 Main Products Inventories
(Mogas, Distillates, Jet, Resid)
65
1,050
60 1,000
Million barrels
55 950
900
50
850
45
J F M A M J J A S O N D 800
Min / Max Range 3y Average 2016 2017
750
700
J F M A M J J A S O N D
3. US gasoline demand is weakening Min / Max Range 5y Average 2016 2017
(DOE gasoline exports from Padd 3

must be miscalculated, and driving 8. Rising drilled but uncompleted wells
length is still up suggesting demand is (DUCs) creates an additional overhang
miscalculated) for US supply (should instead look at
US Implied Gasoline Demand DUCs per completed well, padd drilling
(4-week moving average)
10.0 means larger DUC inventory,
9.8
9.6
two/thirds of the costs remain even
Million barrels per day
9.4
though the well is drilled, fracking
9.2
9.0 equipment in short supply)
8.8
8.6 9. Electric vehicles presents an imminent
8.4
8.2 risk for global oil demand (the
8.0
J F M A M J J A S O N D cumulative effect in 2017 is only 80
Min / Max Range 5y Average 2016 2017 kbd displacement and may reach 280
kbd by 2020 if electric car sales
4. Output recovery in Nigeria and Libya quadruple from 2017-2020, Raymond
risks flooding the market (there is no James calculation)
predictability in exports from these two 10. 2018 supply growth exceeding
countries) demand growth is a bearish indicator
5. OPEC production cuts are worthless (you have to look at the flow deficit
and their return will flood the market in from 2017, which means that since
late 2018 (the whole point is to work 2017 is in deficit (supply lower than
off excess inventory and Saudi was demand) you will still have stock draws
producing above its comfort zone in 2018 if non-OPEC supply grows 1.5
before the cuts) million b/d while demand grows 1.4
6. Rising global floating storage suggest million b/d, you just have a lower stock
a deteriorating supply/demand draw than in 2017)
equation (floating storage is highly
volatile and mainly the result of We tend to agree with most of these ten
logistical issues) myths and believe that the market will realise
7. Crude inventories are the only line that global oil fundamentals really are
item worth tracking in the weekly US improving during 2H-2017.

Oil Market Outlook
3 Wagging the dog Volume in ICE Brent Contracts

(Spread out by contract month)
Several oil analysts seem to believe that it is 300
possible to calculate the correct level of the oil 250
Million barrels
price. In theory the oil price should be equal to 200
the cost to find and develop the marginal 150
barrel of oil (the long-dated price), plus or 100
minus backwardation or contango. The 50
backwardation or contango level should be 0

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58
set by the oversupply or undersupply in the Brent Future Contracts
current market. So if oil stocks are building as

a result of oversupply there will be contango
It is also worth mentioning that after the
structure and if oil stocks are drawing down as
financial crisis in 2008-09 the new regulations
a result of unde supply there will be
that came up, Volker rule and Dodd Frank,
backwardation. And the long-dated price is
has led many banks to leave the commodity
anchored at the cost of the marginal barrel.
space altogether. In addition, the oil traders in
the banks that still remain in the commodity
But is this really how the price discovery
space have been exposed to reduced limits to
works in real life? We question that way of
their potential exposure and increased costs
looking at the oil market, particularly after
on the positions they may want to hold.
what we have seen since the financial crisis in
2008-09.
Before these new regulations, it was hence
easier to warehouse oil positions. Or to put it
First of all it is important to know that 85-90%
another way; to keep positions that came in
of all the volume traded in the oil futures
through customer business in your own
market is trading within the first 6 months of
books, without anyone seeing these positions
contracts. The total value in USD of oil traded
other than your own bank. This is not so easy
in the futures market for crude oil in London
anymore after the new regulations.
and New York is about 20 times as large as
the physical oil market (global oil demand),
What this means is that if for example an oil
and 90% of this value only trades contracts
producer is calling a bank to hedge 50 kbd of
that stretch through the next half a year. This
oil production for 2H-2018, the possibilities to
means that the long-dated part of the futures
keep this position in your own book as an oil
curve is very thin in liquidity. There are some
trader has been reduced and become much
trades done in December contracts for the
more expensive. So instead of maybe
coming 2-3 years out but that is about it.
warehousing this position on your own book
(maybe the trader wanted to go long oil), he
has to turn around and sell the position in the
futures market to a larger extent than before
the financial crisis. This means that the hedge
of 50 kbd by the oil producer immediately hits
the forward curve in the futures market. This
was not happening to the same extent before
the new regulations.

Oil Market Outlook
The type of players that trade out on the oil on you. Hence these kinds of positions are not
forward curve has also changed after the very attractive for oil traders in banks.
financial crisis. According to an article in Risk
Magazine from January, the oil price hedging What if you are a private investor that wants
executed by Airliners has been reduced by to invest 100.000 USD in a Brent 2020
90%. This means that we have lost a large contract? Well, the calculations are a bit of the
amount of natural buyers on the oil forward same, although not as extreme as for the oil
curve. What we have gotten instead is a new trader perhaps. But the private investor would
gang of natural sellers on the curve; the shale have to either tie up money on a margin
oil producers. These players are very active account or to trade on a credit line. If you are
hedgers 6-18 months out on the forward able to set up a credit line the bank will have
curve. The Investment Grade producers to price that line according to its cost of
hedge because they want predictability for capital. It is hence much cheaper for a private
their cash flow while many of the High Yield investor to buy an exchange traded fund
producers hedge because their bank requires (ETF) on a stock exchange like any other
them to do so. High Yield players are now equity. The problem is then however that the
behind about 40% of US shale oil production ETF will only hold contracts that are in the
according to Goldman Sachs. front of the market, so you do not get to
expose yourself lets say 3 years out on the
When we then know that the volume on the curve and put the position in the drawer for a
curve is only a fraction of the volume for the while. So for private investors the forward
first 6 months and that the shale players market is basically only open for billionaires
(natural sellers) dominate the trading 6-18 and not for millionaires as we often frase it.
months out on the curve, it is no wonder that
the forward curve struggles to go past the 55- Why is the forward curve open for billionaires
60 $/b range. If it had been easier for normal and not millionaires? The answer is that for a
people to invest in oil out on the curve the billionaire it is so much cheaper to trade on
price discovery out there may have looked the curve. A billionaire will own a lot of
different. different assets like, equities, bonds, property
etc. He will always normally have a quite large
Let us look at a couple of examples. Let us amount of assets that he is not planning to
say that you are an oil trader working in a turn over or sell. He has no alternative use for
bank and you believe that a Brent price of 53- many of his assets. He can use some of these
55 $/b for 2020 is an attractive price with the assets on his margin account and hence he is
risk-reward skewed to the upside. If you were not tying up any expensive cash, like the
to take such a position into your own book millionaire would have to do.
however, you will probably be exposed to an
internal cost of capital of about 10%, in some The problem for the oil forward curve is that
banks it is even higher. So even if the 2020- there are many more millionaires than
price should increase by 30% from lets say billionaires in this world and the millionaires
53 $/b to 69 $/b by 2020 you will have made only trade in the front of the curve. There are
no money on your trade because of the hence not a lot of potential buyers out on the
capital cost requirement you bank is putting forward curve now that we have lost most of
the Airliners.

Oil Market Outlook
Based on the above we do not believe the same as the cost to produce the marginal
forward curve really reflects what the sum of barrel plus or minus backwardation or
players in the oil market believes the oil price contango, then you should have answered
will be 2-4 years down the road. yes to that question.
Just look at the analysts predictions from the Another example is the 2011-2014 period
last Bloomber survey as an example. The when oil prices fluctuated in a 105-125 $/b
consensus forecast for 2020 is 64-65 $/b, yet range. According to Goldman Sachs latest
the forward curve is pricing 10 $/b lower than Top Projects analysis, the range of break
that. even prices 5 years ago for the worlds top
projects spanned from 57-90 $/b. Still the spot
Bloomberg survey Brent $/b 2017 2018 2019 2020 2021 price and the forward curve priced much
Median 54 58 61 65 68
Mean 54 57 60 64 67 higher than that back then. The highest
High 64 75 84 90 90
Low 45 37 39 47 47 (marginal) break even price was 90 $/b but
Forward Market 52 53 53 54 55
still the average price for the 2011-2014
period was 110 $/b. We also saw Brent trade
If you take a look at historical forward curves
as high as 127 $/b. This was of course an
for the past 50 quarters you will see that the
argument for why prices could not stay up
forward curve has had a large tendency to just
there, way above 100 $/b for very long, but it
follow the spot price for Brent.
also shows that for meaningful periods of time
Forwards vs Realised of BRENT CRUDE FUTR Sep17
the oil price can move way above and way
153.81
143.22
below the theoretical correct price.
132.63
122.03
111.44
100.84
The theoretical oil price has now moved to a
90.25
narrower band of 52-59 $/b according to the
79.66
69.06 same Goldman Sachs report (which by the

58.47
47.87 way is a fantastic report that we always read

37.28
26.69
when it is published, normally in May every
07-Mar-06
01-May-11
13-Mar-18
19-Jun-04
24-Nov-07
12-Aug-09
17-Jan-13
06-Oct-14
24-Jun-16
30-Nov-19
18-Aug-21
year). We do however as oil analysts have

ambitions to say something more about the oil
Even in the summer of 2008, when Brent price than just what the normalized level
surpassed 145 $/b the forward curve priced should be and it is also fair to say that for
as if the market would stay up there for the most of the time the oil price is not staying at
coming 5 years. We cannot remember that its theoretical level (as we have discussed
any analysts at the time forecasted that it above).
would be possible for the Brent price to stay at
that elevated level for the next 5 years, yet the The oil price is instead contantly overshooting
forward market priced as if that would happen. and undershooting. Our target is to be able to
forecast also these overshoots and
Did the market believe that the cost to undershoots. We think for example that the
produce the marginal barrel of oil suddenly Brent market will again overshoot in 2020
had increased by 50% from March 2008 to (more on that later).
July 2008? We don not think so. But if you
really believe that the oil price discovery is the

Oil Market Outlook
People also need to be aware that the cost of majors needed oil prices above 110 $/b to
the marginal barrel is not written in stone. The break even and hence the price could not fall
cost to produce oil changes over time and the much below 100 $/b. Some said that yes
key driver to the cost development is the oil maybe costs could drop 5-10% but only
price itself. Hence, even if the break even temporarily since the lower costs would only
costs right now span from 52-59 $/b this does be cyclical. We always believed in a mix of
not mean that we will see the same range 5 cyclical and structural lower costs as the oil
years down the road. It could go up and it price dropped. Crisis makes people learn
could go down. faster and this seems to be the case this time
as well.
Break Even Development Costs Down 30%
90 Digitization of oil installations are for example
80
70
here to stay and efforts like that decreases
$/b Brent Equivalent
60 costs. The theoretical oil price is hence lower

50
40 2014 now than before 2014, and a fair amount of it
30 2017
seems to be structural. But this does not
20
10 mean that the Brent price will trade in a
0
Middle Russia US Brazil West US North Canada Average narrow 52-59 $/b band, even if we now have
East Shale Africa GOM Sea Oil
average Deep
Water
Deep
Water
Sands US shale oil as the new swing producer. We
Source: PIRA Energy - 10% IRR After Tax
believe the market is set up for another
overshoot by 2020 and then the long-dated
The market currently believes that the US price will be dragged along just as we have
shale barrels are the marginal barrels, but seen before. It will not be anchored at 55 $/b.
what if the world needs some more deepwater In the oil market the front prices are the boss,
barrels or Canadian oil sands by 2025? Or not the long-dated prices. The tail is wagging
what if the red queen effect (the increasing the dog
legacy decline) in the shale industry becomes
so large by 2025 that costs for shale are
pushed higher? It will also depend on what is
happening to the demand side of course but
we have argued last year, and repeat the
calculations this year, that oil demand growth
is a given for the next ten years.
When we in 2012 predicted that US shale oil

would be a game changer that would send oil
prices down well below 100 $/b, we met
pushback saying that the break even costs for
the marginal barrels are too high for that to
happen. The argument was that costs are
structurally creeping higher due to
deeper/harsher/more complex and that the
oil majors were only given access to
difficult/complex projects. People said that oil

Oil Market Outlook
4 Will the OPEC deal fall Nigerian Crude Oil Production

(Not including condensate)
2.5
apart? 2.3
We have noticed that several analysts argue 2.1
Million b/d
1.9
that the OPEC deal will fall apart before year
1.7
end and that is one of their key arguments to 1.5
be bearish to oil prices into 2018. The recent 1.3
communication from Ecuador that the country 1.1

will no longer be part of the cut deal because Source: IEA
of its fiscal requirements is of course

increasing the risk that other players in OPEC In November 2014 OPEC surprised the oil
will behave the same way. Iraq is the most market by not cutting production. In retrospect
mentioned country in this setting. We we have seen through interviews with Al-naimi
acknowledge that there is a sizeable risk that who was then the oil minister of Saudi Arabia,
the OPEC deal could fall apart, but we believe that when he came to that meeting he did in
that scenario should be far from the base fact plan to cut production. His original plan
case scenario. was not to flood the market and compete for
market share.
Our base case is that the deal stays in place
until March 2018, if not longer if that is A year before this happened we had a
necessary to draw down inventories to the 5- meeting in Dahran with the strategy
year average. If Libya and Nigeria are able to department of Saudi Aramco and asked them
keep the recently achieved production gains, if the other OPEC members would contribute
there may be a need to cut production more to cut production if necessary to protect the oil
from the rest of the countries under the deal. price. Their answer was; yes of course, but
We see that scenario as the only issue that only if the platform is on fire. Then one year
can prevent inventories from drawing down to later there was a need to cut production to
the stock target by the end of Q1-2018. We protect the price, but what happened? The
have already assumed maybe too high other OPEC members did not want to
production growth from US shale for next contribute with production cuts. Al-naimi has
year, so we are not afraid of that assumption in interviews told the story of what happened
(since we are bullish to the price). We will in the now famous November 2014 meeting.
maybe instead have to revise down our shale He came into the meeting with the intention to
production number, but more on that later. cut production but nobody wanted to
contribute.
Now back to OPEC behaviour. Our base case
is that Libya and Nigeria will not be able to The strategy departmet of Saudi Aramco were
keep the recent production gains through correct. The platform was not on fire in
2018. Production from Libya and Nigeria is November 2014. The oil price was still above
impossible to predict, but history has proven 80 $/b. But two years later, in November
that production from these two are not stable. 2016, the platform was on fire as prices had
dropped below 50 $/b.

Oil Market Outlook
OPEC countries like Nigeria and Angola were

forced to apply for crisis loans from the IMF
and World Bank and the economy in
Venezuela was collapsing. The Saudis have
drawn down their foreign reseves by more
than 30% and has burned about 10 billion
USD per month from their reserves since the
oil price collapsed.
The platform is now on fire and when that

happens the OPEC countries find together
again and they do cooperate. They do this
because they know that the alternative is
worse. They now have to focus everything on
putting out the fire. Right now they cannot
afford to think longer term. If the production
limiting deal falls apart the oil price may again
fall back into the 30s and many of these
countries would then run into a full blown
economic crisis. Worst off would probably be
Venezuela, but Nigeria would not be far
behind. Saudi Arabia still has almost 500
billion USD in foreign reserves, but to have
burned about one third of foreign reserves in
just three years must be painful. The situation
must be nerve wrecking.
We hence believe Saudi Arabia and OPEC

now has no other choice than to continue the
path they started in November 2016. They will
have to continue the cuts until inventories are
down to the 5-year average.
How long time it takes to get down to the 5-

year average will depend first and foremost on
production from Libya, Nigeria and US shale.
Demand growth only needs to be average and
to follow seasonal patterns to get there. With
our base case for the global supply-demand
balance we believe global oil inventories in
days of demand coverage will drop below the
5-year average by year end. If that happens,
we should see a higher oil price than were the
market trades today.

Oil Market Outlook
5 Demand growth has The US started off very weak when it came to
oil demand growth in 2017 but has posted
recovered from the weak much stronger numbers in Q2 and the recent
first quarter weekly data shows very decent demand
growth for the US.
Our base case demand growth for 2017 calls
for a growth of 1.4 million b/d with normal
seasonal swings baked in. This means for Total US Implied Oil Demand
(4-week moving average)
example that global oil demand is estimated 21.5
to be 2.0 million b/d stronger in 2H-2017 than 21.0

20.5
Million barrels per day

in 1H-2017. 20.0
19.5
19.0
World Oil Products Demand 18.5
101 18.0
99 17.5
97
Million b/d
95 Min / Max Range 5y Average 2016 2017
93
91
89 The key weakness for oil demand growth so
87
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
far this year has been coming from India.
5 year range 5 year avg 2015 Indian demand growth even turned negative
Source: IEA, DNB Markets 2016 2017 2018
in Q1-2017 due to the demonetization that the
Indian government executed in Q4-2017.
Demand growth in 2016 was 1.6 million b/d,
Money supply (M1) collapsed as a
down from 2.1 million b/d in 2015. One of the
consequence and led to negative oil demand
interesting features has been to note how
growth. People in India to a large extent use
Europe still seems to have a decent amount
cash when paying for gasoline and diesel so
of price elasticity on the evolution of oil
the hit on oil sales should probably not have
demand. Demand for oil had started what
come as a large surprise.
seemed like a structural trend of weakening
oil demand but then the oil price collapsed
into 2015 and demand growth has been quite
decent in both 2015 and 2016. Year to date oil
demand growth in Europe for 2017 is even
stronger than in 2015 and the growth has
accelerated from Q1-2017 to Q2-2017.
Year on Year OECD Europe Oil Demand (IEA)

1.0
0.5
Source: US DOE
Million b/d
0.0
-0.5
-1.0
-1.5
2003 2004 2005 2006 2008 2009 2010 2011 2013 2014 2015 2016
Source: IEA

Oil Market Outlook
Sales of motor vehicles also collapsed due to Aspects data here instead of the official
the demonetization, but have later returned. numbers since the official numbers are not
The Indian money supply has been growing capturing runs from the teapot refineries) plus
again since the start of 2017 and this has net imports of refined products plus or minus
provided positive effects on Indian demand stock changes for refined products. We then
growth in Q2-2017. We forecast Indian calculated Chinese demand growth as weaker
demand growth for 2018 to be almost as than in 2016, but quite at par with the years
strong as what we saw in 2016, when demand 2013-2015.
grew almost 0.3 million b/d. There will be very
easy year on year comps for India in Q1-2018 Chinese gasoline demand is growing decent
and this is together with slightly stronger on the back of strong car sales where still
global GDP-growth the key reason why we 97% of the vehicles sold in China are gasoline
forecast slightly stronger global oil demand powered.
growth in 2018 compared with 2017.
China Automobile SalesShare By Fuel Type
India has just implemented a federal Goods 100%
90%
and Service tax (GST) in July. The reform is 80%
70%
made to improve GDP-growth and replaces
60%
other state and federal taxes. In the transition 50%
40%
period however there are several analysts 30%
who point out that this new GST could affect 20%
10%
Indian oil demand negatively in 2H-2017. 0%
Jan2006 Jan2008 Jan2010 Jan2012 Jan2014 Jan2016
There is all the more reason then to carry a Gasoline Diesel CNG Battery and Plug in Electric
Source: China Automotive Information
stronger oil demand growth number for India
in 2018 than for 2017.
Year to date gasoline demand growth is about
Chinese oil demand growth has been decent 5%, about twice as strong as last year, but
so far in 2017. IEA assess the Chinese weaker than what we saw in the 2012-2015
demand growth at almost 0.3 million b/d for period which averaged at about 10%.
1H-2017. Based on data from the Chinese
government, the Chinese OGP and using Chinese Gasoline Demand Growth %
(Adjusted foir inventory change)
25%
refinery throughput numbers from Energy
20%
Aspects (the consultancy), we get to a
15%
demand growth number year to date for China 10%
of 0.4 million b/d. 5%
0%
YoY Implied YoY Implied -5%

YoY Implied Chinese Chinese
China Oil YoY Chinese YoY YoY Demand inc. Demand incl. -10%
Data in Gasoline Gasoline Diesel Diesel Inventory Inventory J F M A M J J A S O N D
kbd demand Demand% demand demand% adjustment adjustment %
2017 138 5.1% 16 0.8% 397 3.6%
2016 64 2.5% -134 -3.7% 528 5.0%
2015 242 10.1% 11 0.5% 456 4.5%
2014 230 10.9% 9 0.2% 419 4.3%
2013 231 12.0% 50 1.6% 398 4.3%
2012 164 9.1% 77 2.4% 275 3.0%
The calculation of oil demand is then as

follows: Refinery throughput (we use Energy

Oil Market Outlook
Chinese diesel demand growth is perhaps Global GDP-growth has according to IMF
even more interesting since before 2012 the fallen from 5.4% in 2010 and trended
growth in Chinese oil demand was mainly downwards to 3.1% in 2016. IMF is currently
coming from diesel. Diesel demand in China predicting that global GDP-growth in real
almost stopped growing in 2012 and even terms will improve to 3.5% in 2017 and to
went sizeably negative in 2016 on the back of 3.6% in 2018. The orange line in the graph
a weakening investment cycle. The below shows IMFs estimate for global GDP-
construction sector is behind 40% of Chinese growth.
diesel consumption and lower growth in that
sector has probably been the key reason to
diesel demand weakness since 2012. This
year we have however seen a bit of a revival
in the Chinese diesel consumption as some
demand growth is registered.
Chinese Diesel Demand Growth

(Adjusted for net exports and for stock changes)
0.50
0.30
Million b/d
0.10
Historically the oil intensity vs GDP-growth
-0.10
has been about 0.5 but has trended
-0.30 somewhat lower in recent decades. If we were
-0.50
to use a sensitivity factor of 0.4 vs global
Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17
GDP-growth we should see oil demand
growth of 1.4% (1.35 million b/d) in 2017 and
In the big picture, global economic growth will 1.45% (1.4 million b/d) in 2018. This is very
still be the key for oil demand growth. Electric close to the numbers we are predicting for oil
vehicles will have to wait another 8-10 years demand growth in 2017 and 2018.
before it eats meangfully into oil demand. We
calculate, with aggressive electric vehicle It is also worth noting that our own inhouse
sales which increase 10 times from the DNB Markets global macro economic activity
current level in 8 years, that electric vehicles score has shown a marked improvement into
will have stolen only about 0.5 million b/d of 2017 (see graph below). This bodes well for
what would otherwise have been oil demand the global economic growth forecasts into
in 2025. 2018.

Oil Market Outlook
Med Skimming Urals Margin

6
4
2
0
-2
USD/b
-4
-6
-8
-10
-12
-14
Min / Max Range 3y average 2016 2017
Singapore Skimming Dubai Margin

Refinery margins have generally been very 6
strong so far in 2017, which seems to confirm 4
that the improved macro economic conditions 2
USD/b
also are a positive for the oil market. Below 0
are key margins for US, North West Europe,
-2
Mediteranian and Singapore:
-4
-6
US Gulf Coast LLS FCC Margin J F M A M J J A S O N D
20 Min / Max Range 3y average 2016 2017

18
16
14
12
Another factor that is interesting to note about
USD/b
10 global oil demand is that IEA seems to be

8
6
notoriously too negative in their forecasted oil
4 demand in their first take for the coming year.
2
0
The graph below shows the difference to how
global oil demand ended up in the respective
year vs the summer forecast published half a
NWE Skimming Brent Margin year before the respective year starts. The
8
average error is 1.1 million b/d from 2010-
6
2016.
4
2
USD/b
0 IEA Change In Demand Projections From Initial Take

3.5
-2
3.0
-4
2.5
-6
2.0
-8
Million b/d
1.5
1.0
0.5
0.0
-0.5
-1.0
2010 2011 2012 2013 2014 2015 2016 2017
This means that on average oil demand has

been 1.1 million b/d stronger in the coming

Oil Market Outlook
year compared with what IEA originally

predicted. This is of course a very meaningful
number and if we somehow changed a
number in our global supply-demand balance
for next year by such an amount it would be
enough to turn a conclusion between a bullish
or bearish outlook.

Oil Market Outlook
6 Supply vs Demand For OPEC we assume that all the countries

participating in the production limiting deal
In our appendix you will find a summary of
keep their output at June levels through the
supply vs demand estimates for the global oil
end of 2018. For Libya and Nigeria we
market. You will see how the DNB balance
assume that Libyas average crude production
compares with the balances from IEA, OPEC
will be decrease to 800 kbd from September
and EIA if we use the same DNB OPEC
2017 through 2018 and for Nigeria we
scenario in all the balances. You will find that
assume crude production of 1.5 million b/d for
the DNB balance is quite similar to the IEA
the same period, down from 1.6 million b/d in
balance as with the current OPEC scenario
July. History has proven that it is impossible to
we see a sizeable stock draw for 2017 and a
predict oil production month by month from
smaller one for 2018, despite an expectation
these two countries. At the current forward
of large production growth from the US. The
average assumptions we feel the risk is
OPEC balance and the EIA balance are less
skewed quite equal to the upside and the
constructive for prices.
downside.
Pictures often tell more than words so let us

go through our global supply and demand Libyan Oil Production
2.0
assumptions by showing some graphics to the 1.8
data. 1.6
1.4
Million b/d
1.2
1.0
Non-OPEC production is estimated to grow 0.8
0.6
significantly the next 18 months. We forecast
0.4
a non-OPEC growth of 0.6 million b/d for 2017 0.2
0.0
and 1.4 million b/d in 2018. The key growth is Jan-01 Jan-03 Jan-05 Jan-07 Jan-09 Jan-11 Jan-13 Jan-15 Jan-17
of course coming from US oil production Source: IEA
which we estimate to grow 0.65 million b/d in Nigerian Crude Oil Production
(Not including condensate)
2017 and 1.2 million b/d in 2018. The other
2.5
large growth is coming from Canada (180 kbd 2.3
in 2017 and 220 kbd in 2018), Brazil (175 kbd 2.1
Million b/d
in 2017 and 130 kbd in 2018) and Kasakhstan 1.9

1.7
(180 kbd in 2017 and 110 kbd in 2018).
1.5
Russia is however kept flat. 1.3
1.1
Non-OPEC Production (excl. biofuels) Source: IEA
58.0
57.5
57.0
56.5
Million b/d
56.0
55.5
55.0
54.5
54.0
53.5
53.0
Source: IEA, DNB Markets 2015 2016 2017 YTD

Oil Market Outlook
IEA define condensate production as NGLs DNB Markets World Oil Supply-Demand Balance
for OPEC countries and for non-OPEC 3.5
(Implied global stock change)
countries condensate is classified as crude 2.5
oil. This is why you will often see some 1.5
Million b/d
different numbers for particularly Nigeria 0.5
depending on the source you use. Nigeria -0.5
-1.5
produced 1.59 million b/d of crude in June
-2.5
according to IEA data, but if you include Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
5 year range 5 year avg 2015
condensate the number was 1.97 million b/d. YTD 2017 2016 2017
Source: IEA, DNB Markets 2018
Our assumption for Nigeria is equal to an
average of 1.95 million b/d for total liquids Our supply-demand balance assumes stock
output going forward. 1.5 million b/d of this will draws of 1.0 million b/d in 2H-2017 which
be pure crude oil. equals about 180 million barrels lower oil
stocks from June to end December.
OPEC Crude Production
34.0
33.5 If we use our DNB OPEC scenario on a pure
33.0
IEA balance where we use IEAs forecasts for
32.5
Million b/d
32.0 global oil demand and non-OPEC supply, the

31.5 average stock draw for 2H-2017 is also close
31.0
30.5
to 1.0 million b/d so our forecast for 2H-2017
30.0 is not very different from the IEA case.
Source: IEA, DNB Markets

2015 2016 2017 YTD
IEA World Oil Supply-Demand Balance

2.5
Our assumption of demand growth is already 1.5
described in the prior chapter. We are 0.5
Million b/d
forecasting 1.4 million b/d demand growth for -0.5

2017 and 1.4 million b/d for 2018. See the
-1.5
graph below:
-2.5
5 year range 5 year avg
2017 DNB OPEC scenario 2017
World Oil Products Demand Source: IEA, DNB Markets
101
99
97
Million b/d
95
93
91
89
87
Source: IEA, DNB Markets 2016 2017 2018
When we then put global demand vs supply

against each other we arrive at the following
balance.

Oil Market Outlook
7 Global oil inventories are Current Year US Total Petroleum

Inventories vs 5y Avg
drawing in 2017 270
250
One of the key myths about the oil market in
Million barrels
230
2017, which we have already mentioned, is 210
that you constantly keep reading that oil 190
inventores have not been drawing down this 170
year and that there is still a glut in the global 150
oil market. We wonder how these stories 130

evolve. When we look at the key data we find
that oil inventories are drawing.
The monthly US oil data for May was released
just two days ago and the data showed that
Let us start by looking at the most watched
US crude oil stocks drew down 11 million
market with the timeliest data. This is of
barrels for that month while the five-year
course the US market. The oil stocks in the
average calls for no stock change in May. In
US costitute the largest part of the OECD
April the US crude stocks drew 17 million
stocks.
barrels while the norm calls for a build of 7
million barrels for that month.
When you look at the US oil stocks and
summarize crude oil and the key four refined
Monthly US Crude Stock Change (excl. SPR)
products (gasoline, distillates, jet fuel and (Last 5 years)
25
resid fuel) you will see a very clear picture of 20
Million barrels stock change per month
reduced inventories ever since February this 15
10
year. 5
-5
US Crude & 4 Main Products Inventories -10
(Mogas, Distillates, Jet, Resid)
-15
1,050
-20
1,000 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Million barrels
950 5 year average 2017
900
850
This is bullish data and the weekly data shows
800
750 that the bullish trend has continued in June

700 and July.
US Crude Inventories
550
The total US petroleum stock overhang vs the
500
5-year average is reduced by almost 110
Million barrels
million barrels since the peak in February. 450
400
350
300

Oil Market Outlook
When you look at the OECD data you will see build 1.5 days of coverage; instead there has
that they have been drawing slightly down in a been a 2.6 day draw in stock coverage for this
period where they normally build. The stock period.
draw is not large but nonetheless represents a
drawdown of 20 million barrels since January. OECD Total Oil Industry Stocks - Days of Demand
67
OECD Total Oil Industry Stocks 65
3,200 63
3,100
Days
61
3,000
Million barrels
59
2,900
57
2,800
55
2,700 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2,600 Source: IEA 5 year range 5 year avg 2015 2016 2017
2,500
Source: IEA 5 year range 5 year avg 2015 2016 2017

The above also means that the stock
overhang vs the 5-year moving average has
The latest reported OECD data is now from dropped from 284 million barrels in January to
May. The official June data will not be 167 million barrels in July and we believe we
reported until August 11. If we add the US will see the overhang almost gone by year
stock change for total petroleum stocks end.
reported through June and July (minus 31
million barrels) and assume that the US will OECD Total Oil Industry Stocks
represent half of the OECD, you will see a Differential to the 5-year average stock level
400
clear picture of lower stocks. 300
200
Million barrels
100
OECD Total Oil Industry Stocks 0
3,200 -100
3,100 -200
3,000 5 year range
Million barrels
2015
2,900 2016
Source: IEA,
DNB Markets 2017
2,800 Forecast if 70% of global stock change happens in OECD
2,700
2,600
2,500
If we look at the JODI data to get a gauge of
global oil inventories, the picture is also clear,
particularly when you translate global oil
stocks over to days of global demand
If you translate the OECD stocks to days of coverage. Stocks are drawing down and these
demand coverage, it is quite visible that draws will accelerate into 2H-2017.
OECD stocks are drawing in a period where
they should normally build. OECD stocks in
days of demand coverage are down from 65.5
days in January to 62.9 days in July (if we
assume that US is half of OECD stock
changes). The average for this period is to

Oil Market Outlook
Global Oil Stocks In Days Of Demand Coverage

(Includes SPR but not for China, demand taken from IEA)
64
Days of demand coverage
63
62
61
60
59
58
57
56
5 year min 5 year average 2015
Source: JODI, IEA
2016 Forecast 2017
So where are all these headlines that say the

global oil glut is not over coming from? Supply
has been meaningfully lower than demand
since March according to the IEA database
and this situation will continue through the rest
of 2017.
IEA World Oil Supply-Demand Balance

3.5
2.5
1.5
Million b/d
0.5
-0.5
-1.5
-2.5
We though the definition of a glut was the

same as over supply. The over supply is not
there anymore, and it is unlikely to return in
2017 unless Libya return to Ghadafi-era
production and stay there and Nigeria stops
all oil theft and sabotage in 2H-2017. Neither
of these scenarios are likely to play out in our
opinion. We would advice to stop emphasizing
on headlines that continue to write about an
oil glut. Look instead at the reported data flow.
Dont buy the myth that the glut is still here
and that US shale oil growth has prevented
global oil stocks from drawing down. Stocks
are trending lower. The stock movements are
not lying. Supply has been lower than demand
since March.

Oil Market Outlook
8 We are in phase 2 of the oil The price improvement in 1H-2016 was quite
similar to what happened in 1H-2009. Prices
price recovery went up from about 30 $/b to about 50 $/b
We believe the price recovery process will be despite a market that was still over supplied.
similar to what happened after the financial We believed last autumn that this anticipation
crisis. Then the price recovery could be split phase could take market to 60 $/b, but 50-55
into three distict phases where the price $/b seems to be comparable to the 70-75 $/b
drivers are different from each other. Since that was reached in the summer of 2009. This
the price drivers change from phase to phase is probably because the market believes
this also means that a black box multiple prices above that level will unleash too much
regression model based on many input factors shale oil production from the US.
that correlate with the oil price will not work,
since you have to change the weight of each Time Spread vs Brent Flat Price
factor through time according to how 80
Phase 1 Phase 2 Phase 3
1.0
0.5
important each factor is for the price 70
0.0
formation. Brent 1st month
60
Brent 1 vs3
-0.5
50 -1.0
In the price recovery after the financial crisis, -1.5
40
the first phase took place from January to July -2.0
30
in 2009, when the price improved from -2.5
broadly speaking 40 $/b to 70 $/b. How could 20

Jan2016 Jul2016 Jan2017 Jul2017 Jan2018 Jul2018
-3.0
the price improve that much when the market Brent 1st Month Brent 1 vs 3 (RHS)
was in the middle of the peak over-supply and

oil stocks continued to build through the Phase 1 is over and we are way into phase 2.
period? We call this phase 1 the anticipation Phase 2 is characterized by global oil stock
phase. It has a lot to do with psycology and draws starting to materialize. After the
the feeling in the market that the price had financial crisis this phase lasted from the
fallen too much and was unsustainably low. At summer of 2009 to the summer of 2010.
the same time OPEC had promised to cut Stocks startet to draw down but the price did
production to remove the over supply into very little for those about 12 months. The
2009. Brent price just traded in a range of 70-75 $/b
despite a gradually tighter market where
Time Spread vs Brent Flat Price stocks slowly were drawing lower.
4
140 3
Summer 2009-Summer 2010
2
We are now in a similar phase but the 70-75
120
1 $/b is now broadly 50-55 $/b instead. The
Brent 1st month
Brent 1 vs3
100 0
-1
market will probably not be willing to step
80
-2 much out of that range until it is clear to
60 -3
everybody that stocks are drawing down and
-4
40
-5 will not rebuild into 2018. As long as OPEC
20
Jan2008 Jul2008 Jan2009 Jul2009 Jan2010 Jul2010 Jan2011
-6
sticks to its production limiting deal and
Brent 1st Month Brent 1 vs 3 (RHS) achieve above 80% compliance, stocks will
accelerate the drawdown into 2H-2017 as
demand keeps on growing. Then we are into

Oil Market Outlook
phase 3, where we think it is likely that we will

see backwardation in the front of the Brent
curve unless Libya and Nigeria continue to
surprise to the upside.
During the price recovery after the financial

crisis, phase 3 kicked in late in the autumn of
2010. OECD oil stocks had then drawn down
enough to flip the market structure into
backwardation. Then the 1st month Brent
price increased from 70-75 $/b to above 125
$/b during 1H-2011.
OECD Total Oil Industry Stocks

Differential to the 5-year average stock level
150
100
Million barrels
50
-50
-100
Source: IEA,
DNB Markets 2009 2010 2011
It is interesting to note that since 1995 we

have seen 5 years were the front of the Brent
forward curve has been on average
backwardated and every time the Brent price
has increased. The 17 other years have seen
a contango structure and in those cases it is
50/50 if the Brent price has gone up or down.
We believe phase 3 will send the Brent price
to 60 $/b by the end of Q4 2017, and this is
hence our current price target for that quarter.

Oil Market Outlook
9 CAPEX cuts how do they Total accumulated CAPEX by the Oil Majors
amount to 1.15 billion USD since January
affect the oil market? 2011.
From 2014 to 2016 we have seen a reduction
of investments in the global oil space of Oil Majors Accumulated Capex
almost 50%. According to IEAs World Energy -1,200
(Shell, BP, Total, Statoil, ENI, Exxon, Chevron, Conoco)
Investment 2016 report, global upstream oil

-1,000
(not including natgas) investments fell from
-800
Billion USD
about 550 billion USD in 2014 to about 300
-600
billion USD in 2016 (down 45%).
-400
-200
0
Jan2011 Jan2013 Jan2015 Jan2017
Source: Bloomberg
You may think that this enormous spending

should have generated some production
growth for the oil majors in this time frame but
unfortunately that is not the case. Total
production from the Oil Majors has instead
decreased from 22 million b/d in 2011 to
average 21.4 million b/d the past 4 quarters.
Barclays Capital E&P spending survey from
January this year states that Worldwide E&P
Oil Majors Oil Equivalent Output
spending (including natural gas) fell from 673 (Shell, BP, Total, Statoil, ENI, Exxon, Chevron, Conoco)
24
billion USD in 2014 to 377 billion USD in
2016. 23
Million b/d
22
Oil majors have cut their quarterly spending

21
from a peak of 66 billion USD in Q4-2013 to
20
23 billion USD in Q1-2017, a decrease of
65%. 19
Source: Bloomberg
Oil Majors CAPEX By Quarter

(Shell, BP, Total, Statoil, ENI, Exxon, Chevron, Conoco)
-70 This is then after having invested 1.15 billion
-60 USD. This fact illustrates one of the key
-50 challenges facing the global oil market; most
Billion USD
-40
of the investments that are undertaken are
-30
only contributing to keep decline rates in
-20
check.
-10
0
Source: Bloomberg

Oil Market Outlook
We did see some growth in oil production The IEA reports of a year on year decrease in
from the Oil Majors as a group during the past productionof more than 3% from the group of
3 years but that growth now seems to be countries that are outside of OPEC/Russia/
behind us. For the latest couple of quarters USA/Canada/Brazil/Kasakhstan.
there has been no more year on year oil
production growth reported for this important YoY production change outside of OPEC, Russia,
group of players. USA, Brazil, Canada and Kazakhstan
(12-months mavg)
2.0%
1.0%
Oil Majors Oil Output
(Shell, BP, Total, Statoil, ENI, Exxon, Chevron, Conoco) 0.0%
1.5
-1.0%
1.0
0.5 -2.0%
0.0 -3.0%
Million b/d
-0.5
-4.0%
-1.0 2004 2006 2008 2010 2012 2014 2016 2018
-1.5 Source: IEA
-2.0
-2.5
-3.0 These countries represent about 25% of the
Source: Bloomberg
global oil market. Year on year production
from this group of countries hence fell almost
The mentioned CAPEX cuts from 2014 to 0.9 million b/d at the worst.
2016 are historically large and have already
affected the oil market. The first effect was YoY production change outside of OPEC, Russia,
USA, Brazil, Canada and Kazakhstan
that the oil companies were forced to pull the (12-month mavg)
600
emergency brakes. They had to postpone 400
and reduce activity immediately in order to 200

0
protect their dividend. This meant that not only -200
did finding avd development activities suffer; -400

-600
we also for some regions saw reduced activity -800
in short cycle projects, tie backs and infill -1,000

2004 2006 2008 2010 2012 2014 2016
drilling. Source: IEA
The result of this reduced activity was that the The countries that saw the largest decrease in
number of countries in the world which saw barrels were China and Mexico, but as noted
falling oil production increased from about 45 above there were about 60 countries on this
to 60 countries. list.
Number Of Countries With Drop In Output

(12-months mavg)
60% 65
Number of countries with falliing output
Share of countries with falling output
55% 60
55
50%
50
45%
45
40%
40
35% 35
30% 30
Jun-03 Jun-05 Jun-07 Jul-09 Jun-11 Jun-13 Jun-15 Jun-17
Share (LHS) Number of countries (RHS)

Source: IEA

Oil Market Outlook
Year on Year China Supply (IEA)

We should however mention the following: In
0.60 2013 the growth from these projects were
0.40 estimated by GS at roughly 0.5 million b/d
0.20 (IEA says non-OPEC non-US grew 0.2 million
Million b/d
0.00 b/d that year), for 2014 GS estimated about

-0.20 0.75 million b/d growth (IEA says non-OPEC
-0.40 non-US grew 0.6 million b/d that year), for
-0.60 2015 GS estimated about 1.1 million b/d
2010 2011 2012 2013 2014 2015 2016 2017

growth (IEA says non-OPEC non-US grew 0.4
million b/d that year), for 2016 GS estimated
about 0.85 million b/d growth (IEA says non-
Year on Year Mexican Supply
OPEC non-US fell 0.4 million b/d that year).
0.4
0.2 The key point that we want to make is that

even if you were to correctly track the largest
Million b/d
0.0
-0.2 projects in the world, you could still miss quite

-0.4
significantly the changes in global oil
-0.6
production. 2016 is a very good example.
2003 2005 2007 2009 2011 2013 2015 2017 These 432 projects were supposed to
contribute to a growth of about 0.85 million b/d
for 2016, still net production outside OPEC
The oil market has hence already received and excluding USA fell almost 0.4 million b/d
some support froim CAPEX cuts through that year. So far in 2017 the production
lower production, but we believe the main growth for non-OPEC non-US is averaging
effects of CAPEX cuts remains to be seen. less than 0.3 million b/d but the top 432
The large growth in investments during the projects outside US shale are supposed to
ten years leading up to 2014 is kind of saving grow 1.7 million b/d this year. This does not
the day for the availability of oil so far. Due to look very likely to take place as we are now
the sizeable time-lag from the start of half way into the year.
investments to first oil for all large oil projects,
the world is still benefitting from start-up of We do see in the data that production is
projects were the investment decision was growing in Canada, Brazil and Kasakhstan,
taken before the oil price collapsed. but the growth is so far much smaller than the
list of start-ups suggest and the growth that
Goldman Sachs (GS) writes in their large Top many expected from Russia will now not take
Projects 2017 report that 2017-2019 could place in 2017 and 2018 due to Russia being
see the largest delivery of mega-projects in part of the OPEC/non-OPEC production
history. They estimate that production growth limiting deal.
from the key 432 large projects that they track
(this is not including shale) will grow about 1.7
million b/d in 2017, about 1.3 million b/d in
2018 and about 1.2 million b/d in 2019.

Oil Market Outlook
Year on Year Brazil Supply
0.50
0.40
0.30
Million b/d
0.20
0.10
0.00
-0.10
-0.20
2003 2005 2007 2009 2011 2013 2015 2017
Source: IEA
Year on Year Canadian Supply

0.8
0.6
0.4
Million b/d
0.2
0.0
-0.2
-0.4
2003 2005 2007 2009 2011 2013 2015 2017

Oil Market Outlook
10 Decline rates and call on We have argued several times before that we
did not buy the story that global annual
shale decline rates are as large as 3-4 million b/d.
How much is the current oil production base Many still use those numbers. You still see
declining per year and how much new oil will statements that within the next ten years we
have to be put onto the global oil market by will have to find and develop 4 new Saudi
2020 to cover both demand growth and Arabia just to fight decline.
decline rates? And how much will US shale oil
have to provide of this requirement? These When we calculate the average decline from
are of course key questions for the oil price in the base ten years ago we calculate that the
the coming 3-4 years. base has declined 28 million b/d, which
translates to an annual decline of 2.8 million
The questions are difficult to answer. We b/d on average from the base in 2006.
would not dare to use any bottom up analysis
on decline rate studies to answer such Decline The Past Ten Years - 2.8 Million b/d Per Year
(The calculation is done by extracting all start ups of new projects since year 2006)
questions. There are just too many oil fields in 95
this world and you would have to use a rule of 85
28 million b/d is lost to decline since 2006

thumb on too many of them. We would 75
Million b/d
instead prefer to be approximately correct 65
instead of precisely incorrect on this issue. In 55
a top down approach you may miss the end 45
point by 1 million b/d but the bottom up 35

2006 2008 2010 2012 2014 2016
analysis incorporates a much larger risk than

Source: DNB Markets, Rystad Energy
that. Some players of course hates that way
of thinking. They would like to have 100%
The average yearly decline from the 2006-
control and do the equations bottom up and
base in the ten-year period behind us was
be 100% certain of the answer. They want to
4%. The procedure here is simply to extract
be so certain that even decimals can be used
all the projects that have started up production
when reaching the conclusion.
after year 2006, and then calculate what the
projects that were already producing in 2006
We have noticed that some oil analysts even
were contributing in 2016.
use decimals when they predict next years oil
prices. Maybe they want to give the
According to a recent research report by
impression that it is possible to calculate what
Redburn, most global decline rate studies
the oil price should be? Maybe they think that
estimate global decline rates at about 6% per
if they use decimals it makes the analysis
year and as high as 9% for offshore
more trustworthy? We are not sure. We prefer
production and 11% in deep-water areas.
to use very rund numbers ourselves. Why kid
IEAs study on decline rates from 2008 goes
anyone here? Global oil data are often not
in the same direction as can be seen below.
trustworthy enough for anybody to reach
conclusions with decimals in the answer. That
goes for decline rate calculations as well.

Oil Market Outlook
average decline the coming ten years from

the current production base. We are instead
factoring in half of that decline in our base
case assumption for the coming 5 years.
Net Oil Need Of 11.5 Million b/d By 2020?

(Assuming 2% net decline rate and oil demand growth of 1.2 mbd p.a)
95
11.5
85
75
We have however argued many times that
Million b/d
65
field decline is far from only being a matter of

55
geology. It is as much affected by above

45
ground issues like government take, oil prices

35
and technological developments. 2006 2008 2010 2012 2014 2016 2018
Global output of crude, condensate and NGLs

2020
Observed historical 10 year decline rate 4%

Global oil demand growth 1.2 mbd per year
Source: DNB Markets, Rystad Energy Forward decline rate 2% per year
As the interesting Redburn report points out;

the decline rate studies are always based on This decline still amounts to 7 million b/d in
raw production data where aging production the 4 years 2017-2020, which will be the
facilities in fact explain more of the variance in same as an average yearly decline of 1.75
production than reservoir depletion. The million b/d for that period. This is a quite
report for example mentions that for the UK oil conservative calculation but we believe the
sector, the plant reliability explains 65% of all efficiency improvements coming from less
production losses. Prior studies had shown downtime helps justifiy such an assumption.
that field decline rates were accelerating per
decade, but recently this trend has changed. We also see a tendency to now run short
cycle projects, tie-backs and tie-inns instead
One of the key elements mentioned in the of large complex projects. These are often
report is reduced down-time due to condition phase 2, phase 3 or extensions of already
based maintenance with the help of producing oil fields and helps keep decline
digitisation. The traditional oil field has been from an already producing field in check.
monitored through physical inspection.
Instead the digital oil field can be monitored in
real time through fiber optics. The traditional
oil field has also normally been set up for
time-based maintenance every year after a
pre-defined time-schedule. The new digital oil
field can instead be monitored in real time
measuring heat, pressure and vibration data.
Instead of ad hoc work the new digital
systems means work-flows are continuously
executed. Source: IEA World Energy Investments 2017
This efficiency improvement, we believe, will

for the coming 3-4 years help keep net decline
rates in check so that we will not see the 4%

Oil Market Outlook
We do however believe that the lack of final trump most other factors in the oil market
investment decisions (FID or sanctions if you around 2020.
want) in 2014-2017 will come back and haunt
the market around 2020.
Number Of Projects Sanctioned With CAPEX Above

1 Billion USD
35
30
25
20
15
10
0
2013 2014 2015 2016 YTD 2017
Source: PIRA Energy
Source: Energy Aspects

The lack of new project start-ups, which will
become visible from around 2020-2023 are When we map up the required supply
likely to lead to an accelerated challenge to additions by 2020, we use a net decline rate
keep up production in that period. This will of 2% from the 2016 production base of crude
likely be visible despite continued efficiency and condensate production. This means
improvements and less plant downtime than about 7 million b/d needs to be replaced due
what we historically have seen. There are to field decline in the 4 years 2017-2020. In
simply not enough large projects that are addition the world needs to cover oil demand
being sanctioned anymore as can be seen in growth of about 1.2 million b/d on average for
the graph above. In 2013 more than 33 those 4 years according to IEA. We do not
projects with CAPEX above 1 billion USD was believe that number will be smaller than that.
sanctioned according to data from PIRA It is probably more upside than downside to
Energy. This has since fallen to only 6 that assumption in our opinion. The above
projects in 2016 and so far not many more in suggest that the required oil demand that
2017. This is likely to have a very visible effect needs to be covered by supply in 2020 will be
on the global oil balance from 2020-2025. 101.4 million b/d. The gap that needs to be
filled with fresh started oil projects in 2017-
Energy Aspects calculate that the effects of 2020 is hence above 11 million b/d.
postponed projects since 2014 on production
will amount to about 4-5 million b/d by 2020 When we look at the Rystad database for
and will reach 6 million b/d by 2022. It is really expected start-ups of new projects in 2017-
from 2020 that the effects of these project 2020 we find 4.4 million b/d from non-OPEC
cancellations become very meaningful. The and 1.9 million b/d from OPEC. These are the
effect of these postponements is small in sanctioned new projects that have been given
2017 and 2018 but starts to accelerate in a name. Remember that these projects are
2019. Unless oil companies start to sanction not counting extensions and phase 2, 3, etc
larger projects again the effect is likely to projects. Those kinds of projects on existing

Oil Market Outlook
fields will instead help to keep decline lower

than historical levels.
Our calculation then shows that the call on

shale, which is the only meaningful supply
growth that can move the needle in a 3-4
year perspective, is quite large at above 5
million b/d. In our global supply-demand
balance we assume that US shale production
will grow 0.65 million b/d in 2017 and 1.2
million b/d in 2018. We would then need a
growth in US shale that averaged above 1.6
million b/d in both 2019 and 2020 in order to
cover our calculated requirement of supply.
2016 vs 2020 Global Oil Balance
104
5.1 101.4
102
100
97.0 7.0 1.9
98
4.4
96
kbd
94
92
90
88
86
84
82
80
Global crude, Net decline rate Non-OPEC OPEC projects Required new World liquids
condensate and of 2% (half of projects under under shale oil to demand by
NGLs output in last ten years) development development balance 2017- 2020 (1.2 mbd
2016 (IEA) by 2020 (414) (Rystad (116) (Rystad 2020 pr year-2016
Energy) Energy) was 96.6 mbd)

Oil Market Outlook
11 What oil price will be US Shale Wellhead Price Diff to WTI

Plains South Texas Light Sweet (USD/b)
-8.0
required for shale to grow
-8.5
enough? -9.0
Based on our calculation on the call on shale -9.5
in 2020, total shale oil production in the US

-10.0
would have to more than double from about
-10.5
4.3 million b/d in 2016 to average at 9.4
-11.0
million b/d in 2020. Purely based on break Jan2016 Apr2016 Jul2016 Oct2016 Jan2017 Apr2017 Jul2017
even calculations updated by PIRA Energy in

July 2017 (using 10% cost of capital), it does You hence need to take the communicated
not look like it will be possible to achieve that break even prices for the different companies
kind of growth with a Brent price in the 50-60 with a pinch of salt. Often they just
$/b range. At least not if the oil companies communicate a break even oil price without
want to produce these barrels with an specifying if it is a WTI equivalent price, a
adequate return on capital employed. The 50- Brent equivalent price of if it is the wellhead
60 $/b range simply does not look high price. If a company communicate the
enough to produce economic barrels in the wellhead break even price it will, as described
amount the market will require. above, be quite much lower than the Brent
equivalent price.
It is not totally straight forward to monitor

which price levels that lead to changes in
behaviour for the US shale players. The
second largest play, the Eagle Ford, seems to
be the most sensitive to oil prices in the recent
downturn.
US Oil Rig Count

(Baker Hughes)
90 400
Source: PIRA Energy, break even based on 10% cost of capital
80 350
70
You also have to remember that the shale 60
300
producers are really not achieving the Brent 50

250
price when they are selling their crude. Their 40

200
wellhead price is much lower than that. We 30 150
can for example mention that a shale 20

May-16 Jul-16 Sep-16 Nov-16 Jan-17 Mar-17 May-17 Jul-17
100
producer who wants to deliver his light sweet Eagle Ford North Dakota (Williston basin) Permian (RHS)
barrels into the Plains pipeline system in

South Texas will receive a price that is 9-10 If we focus just on the rig count in the Eagle
$/b lower than the WTI price. If you then add Ford vs the WTI price, it seems the rig count
2-3 $/b to the WTI price you get to the Brent is extremely sensitive when the WTI price
equivalent price. drops below 50 $/b.

Oil Market Outlook
Eagle Ford Oil Rig Count vs WTI Price US Horizontal Oil Rigs Weekly Change
(Baker Hughes) 20 (Baker Hughes - 4 week moving average)
100 60
15
90
55
80 10
50
70
5
45
60
0
50 40
40 -5
35
30
30 -10
20
25 -15
10
0 20 -20
Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17
Eagle Ford Oil Rigs (LHS) WTI price (RHS)
It only took a lag time of 3-4 weeks from the US shale producers have of course hedged a
WTI price fell below 50 $/b and until you saw large part of their output at WTI prices above
a flattening to decreasing rig count for the 50 $/b for 2017, but the rig count nonetheless
Eagle Ford. If we look at the total US seems to be very sensitive to a spot WTI price
horizontal oil rig count we saw the first below 50 $/b.
decrease in the rig count about 4 months after
the WTI price fell below 50 $/b, so the lag-
time is much larger for the total rig count than
that for Eagle Ford.
US Horizontal Oil Rig Count vs WTI Price

(Baker Hughes)
700 60
650
55
600
50
550
500
45 Source: Bloomberg Intelligence
450 40
400
35
350 Calendar 2018 is at the time of writing pricing
30
300
250
25
below 50 $/b and will probably not impose too
200 20 much of a head wind for crude prices as long
Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17
Horizontal Oil Rigs (LHS) WTI price (RHS) as the calendar 2018 is pricing below 55 $/b.
During Q1-2017, the WTI calendar 2018
We do think that it is really the headline mostly priced above 55 $/b but still not much
numbers that affect the oil price formation the hedging seems to have been executed at
most and since the WTI price has been below those price levels for 2018. Most of the
50 $/b for the past 3 months we may have 3 hedging seems to have been related to 2017
months ahead of us now with a flat or production.
decreasing US oil rig count.
We do however see a good argument for a
The 4-week moving average total US oil rig flattening of the WTI curve and possibly a
count has decreased from a weekly addition backwardated market in the front as the
of 11 to about zero since April. A decreasing supply-demand balance tightens in 2H-2017
rig count may hence prove to be a bullish at the same time as US oil producers increase
price catalyst during the summer months. their hedging ratio for 2018 if they can achieve
levels in the 50-60 $/b range. Compared with
one month ago we do in fact already see a

Oil Market Outlook
significant flattening of the WTI curve as can Yield For US High Yield Shale Players vs US 5 Year Interest Rate
(Continental Resources, Chesapeake, Sanchez, Whiting, SM Energy, QEP
Resources, Newfield, Energen)
be seen below. 5.5 41
43
5.0
% Yield - Daily data

45
$/b WTI price

4.5
47
4.0
49
3.5
51
3.0 53
30.04.2017
05.05.2017
10.05.2017
15.05.2017
20.05.2017
25.05.2017
30.05.2017
04.06.2017
09.06.2017
14.06.2017
19.06.2017
24.06.2017
29.06.2017
04.07.2017
09.07.2017
14.07.2017
19.07.2017
24.07.2017
Yield vs the US 5y interest rate WTI
It is hence highly interesting to note how

sensitive the high yield market is to changes
in the WTI price with almost no time-lag. The
The US shale market is very sensitive to
graph above shows the average yield diff to
financing costs. The free cash flow for our
the 5-year US treasury for a 5-year loan
selected group of shale companies stayed
issued by a sample of high yield US shale
negative despite a Brent price that averaged
producers.
53 $/b in Q4-2016/Q1-2017.
It shows that when the WTI price fell below 50

US Shale Companies Free Cash Flow By Quarter
(Anadarko, Devon, Chesapeake, EOG, Apache, Hess, Continental, $/b the yield immediately crept higher. During
Murphy, Pioneer, SM Energy, Whiting, Cimarex, Concho)
2.0 the WTI price decrease from 51 $/b to 43 $/b
0.0
-2.0 the financing costs for these players relative
Million USD
-4.0
-6.0
to the 5-year US treasury increased by almost
-8.0 50%. Since many of these high yield players
-10.0
-12.0
needs a WTI equivalent price above 50 $/b for
-14.0 their CAPEX guiding to meet their free cash
-16.0
Jun2007 Jun2009 Jun2011 Jun2013 Jun2015 Jun2017 flow it is no wonder the high yield market
Source: Bloomberg
starts to close down when the WTI price
decrease towards 40 $/b. This should be quite
According to a recent report by Goldman
a potent argument for why it should not be
Sachs (published May 15th) about 2 million
possible to stay as low as 40-45 $/b Brent
b/d, or equal to 40% of the current US shale
equivalent price for very long.
oil output of 5 million b/d, belongs to
companies that are depending on the high
Now let us return to what we started out with
yield market for financing.
in this chapter. Is it possible to grow US shale
production by 5.1 million b/d by 2020 and
what kind of oil price would then be required?
Will such a ramp up of activity unleash some
bottle necks that will push break even prices
at least temporarily higher?

Oil Market Outlook
We argued in 2012 that the US shale oil US Oil Rig Count

1800 (Baker Hughes)
market was so early in the learning curve that 1600
break even prices would probably decrease 1400
and not increase in the shale industry. As 1200
1000
many readers may remember; at that point in
800
time there were many who argued that since 600
the shale industry was about to run out of 400
sweet spots, the break even prices would 200
0
increase from an already high level. We never Feb-11 Feb-12 Feb-13 Feb-14 Feb-15 Feb-16 Feb-17
agreed with that view. We still believe there is

a learning curve to climb in the shale industry
WoodMac wrote in a recent report that a 10%
and would subscribe to the picture below
increase in the service costs would, as a rule
made by Conoco.
of thumb, increase the shale oil break even
price by about 6%. And we have seen many
examples of higher service costs recently.
The graph below is showing the rig rates per
horsepower segment in the Permian. As can
be seen the rates bottomed out in November
last year.
Source: Bloomberg
Source: Conoco
Costs for drilling are not the only costs that

The ramp up in activity has however been so
have increased. PIRA Energy expects total
large during the past year that it looks as if the
service costs to increase 11% this year with
efficiency improvements are not able to offset
amongst other, 6% higher drilling costs, 20%
the increased service costs. The US oil rig
higher proppant costs and 20% higher
count has increased about 150% since the
pressure pumping costs.
through in May last year and is as such the
best example of the increased activity.

Oil Market Outlook
Goldman Sachs wrote in a report published

May 15th that well costs are down 30% since
2014 but that they should now increase 30%
from the bottom, leading to a net drop in costs
since 2014 of 8% after the inflation has
brought prices higher. The increase in well-
costs does however not mean that it is a like
for like when it comes to unit costs ($/b) since
each well on average produce more than
before. The IP rates (Initial Production) have
Source: BTU Analytics
increased and EURs (Estimated Ultimate
Recovery) per well have increased.
The large difference between these two plays
is that Eastern Eagle Ford had already come
The improved performance per well is
a long way in the high-grading process by
however a very difficult issue to dissect. This
2014. Notice how the red colors were not that
is due to the fact that if you as an operator
large while the green and yellow colors were
choose to only drill your best acreage and
large in Eastern Eagle Ford already in 2014.
abandon the weaker acreage, your efficiency
In the Bakken however you can see that in
measures per well will of course look a lot
2014 the colors were mainly red which means
better. This issue is difficult to segregate on a
the break even price on the wells drilled there
national level. But we can provide some
were quite high. Few green and yellow wells
examples. Data from BTU Analytics show that
were drilled in the Bakken then.
for Eastern Eagle Ford the break even price
has not fallen more than 5-10 $/b.
When the operators in the Bakken then
drastically reduces the number of completed
wells, and almost only complete wells with a
green or yellow color; what happens to the
break even price then? It drops a lot of
course. It is a high grading effect that hits the
metrics that gives us the break even prices.
For Eastern Eagle Ford however most of the
wells drilled were already green or yellow, so
it was not possible to stop drilling only the red
wells (the ones with a high break even price).
Source: BTU Analytics Hence the break even price for Eastern Eagle
Ford only fell marginally compared with the
Break even prices have however fallen a lot Bakken.
more in the Bakken, where they are down
about 20-25 $/b. Because of the efficiency effects that will be
recorded in type curves and EURs due to the
choice of what projects you drill, it makes it
difficult to assume what kind of efficiency
improvements we should use in forward
looking models for US shale oil production.

Oil Market Outlook
We do not have access to enough timely data turned negative. This is due to the fact that
on the monthly number of completed wells for the US shale industry is able/willing to drill
the US to make a model on that, so we are more wells than what they are completing,
instead modelling based on data from EIAs evidenced by the growing numbers of drilled,
monthly Drilling Productivity Report. What you uncompleted wells (DUCs).
get there is monthly data for output, and
calculated productivity per oil rig. DUCs In Eagle Ford, Bakken, Permian, Niobrara
5,500
When the number of rigs started to increase 5,000
4,500
the contribution per average rig in the average
Total DUCs
4,000
shale play started to flatten out and drop.
3,500
Contribution per rig started to decrease after 3,000
September last year. 2,500
2,000
Dec-13 Dec-14 Dec-15 Dec-16 Dec-17
Contribution Per Working Oil Rig In Start Up Month Source: EIA US Drilling Productivity Report
800
Contribution per working oil rig in start up month, b/d
700 If the rig cunt continues to increase it is likely

600
that the DUCs will continue to increase, and if
500
400
the DUCs continue to increase it is likely that
300 the reported efficiency per rig will continue to
200
decrease. If we should see higher oil prices
100
0 which make companies again choosing to drill

Jan-07 Jan-09 Jan-11 Jan-13 Jan-15 Jan-17 Jan-19
Source: DNB Markets, EIA Drilling Productivity Report for Eagle Ford, Bakken, Permian,
Contribution per working oil rig historicals in poorer acreage, this will also decrease the
Utica, Marcellus, Haynesville, Niobrara
Contribution per working oil rig going forward
contribution per rig. We have hence put such

an assumption into our shale model.
We are assuming that as long as there is a
mismatch between capacity to drill new wells
We have assumed a monthly oil rig addition of
and the capacity to complete them, the
20 rigs per month through 2020 in order to
contribution per rig will continue to decrease.
map out what should then happen to US shale
oil production.
Increase In Monthly Productivity Pr Oil Rig
(12-month mavg)
5.0%
Oil Rigs In 7 Shale Regions
4.0%
Monthly improvement for a new oil rig
3.0%
1,400
2.0%
Oil rigs in 7 shale regions
1.0% 1,200
0.0%
1,000
-1.0%
800
-2.0%
600
-3.0%
400
Historical productivity improvement pr rig Modelled productivity improvement pr rig
200
Historical Rig count 7 shale regions Modelled rig count 7 shale regions
There used to be a monthly increase in

productivity per working rig of around 2.5%
per month, but after the rig count started
increasing the contribution fell and has now

Oil Market Outlook
The above shown contribution per rig and volume, but it is still not so that if the Brent
number of rigs is then combined to a model price stays above 55 $/b the volume is
that use a decline rate of 87% for the first 3 limitless. The oil producers will produce
years of a new well (70%, 40% and 30%) and significantly more shale crude if the Brent
thereafter 5% per year. This model has been price is 80 $/b compared with 55 $/b. All the
fairly accurate when it comes to historical key shale oil players have a large range of
production as can be seen in the graph below. different projects and the break even range
even for star players like EOG and Pioneer
US - Oil Production From 7 Shale Regions still spans from the low 20s to the high 90s.
Decline rate pr well: 87% the first three years (70-40-30)
10,000
9,000
Thousand barrels per day
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
Historical US crude production from 7 shale regions
Utica, Marcellus, Haynesville, Niobrara Modelled US crude production from 7 shale regions
The model spits out the following growth rates

in US shale for 2017-2020: Source: Financial Times using Drilling Info data
20 new rigs per month 2016 2017 2018 2019 2020 2017-2020
Average production 5,021 5,706 7,270 8,154 8,606
We do believe that at 60 $/b Brent, which is
Growth kbd -415 684 1,565 884 452 3,584
our forecasted price for 2018, there will be a
lot of new shale oil production coming on
We just argued above that we are looking for
stream. We currently forecast that US oil
a growth of US shale production of 5.1 million production will grow 100 kbd per month in
b/d in 2017-2020. The assumptions above will
2018. This will provide a year on year growth
hence not do the job. We will be short 1.4 of about 1.2 million b/d (the growth has been
million b/d even if the rig count went back quite linear so far in 2017).
above the record levels from 2014. But could
the contribution per rig start to increase again,
One should however be aware that the same
instead of continuing to decrease? It could growth rate for 2019 and 2020 would require
happen of course but that would have to imply
much higher activity or a large improvement in
a drawdown of DUCs we believe. But if DUCs
efficiency. This is due to the accelerating
are drawn down it will probably mean that rigs legacy decline rate that kicks in for the shale
are not increasing like we have built in. All in industry whenever a large chunk of new wells
all we believe a growth of 5.1 million b/d in
are starting up. If activity is high, and many
only 4 years is a stretch, even for the
new wells are completed, then legacy decline
remarkable US shale industry. accelerates. If activity is slowing down and
fewer new wells are completed, then legacy
It is important to remember that the volume
decline slows down.
that will hit the market depending on a specific
oil price is not binary. Yes the cost curve looks
pretty flat for a lot of potential shale oil

Oil Market Outlook
Legacy decline is the volume of new shale oil Modelled Crude Oil Production 7 Shale Regions
Decline rate pr well: 70%, 40%, 30% the first three years (87%), thereafter 5% per year
production that needs to turn to sales every 10.0
month in order to keep production flat from the 9.0
8.0
prior month. 7.0
Million b/d
6.0
5.0
Legacy production decline vs new production
4.0
600
3.0
500 2.0
1.0
400
Thousand b/d
0.0
300 Jan-07 Jan-09 Jan-11 Jan-13 Jan-15 Jan-17 Jan-19 Jan-21
Source: DNB Markets, EIA Drilling Productivity Report for Eagle Ford, Bakken, Permian, Utica, Marcellus, Haynesville, Niobrara
200
100
The graph above hopefully illustrates the
0
Mar-07 Mar-09 Mar-11 Mar-13 Mar-15 Mar-17 challenge. As you can see; the decrease in
Legacy decliine
Source: EIA US Drilling Productivity Report Production from new wells production should no more wells be
completed is much larger by the start of 2019
As you can see in the graph above, the legacy that at the start of 2017. If no more shale wells
decline fell from 345 kbd to 285 kbd as activity were to be completed after January 2019, the
was reduced into 2015. The US shale industry decline in production from US shale would
could then afford to start up 60 kbd less surpass 3 million b/d by the end of that year.
volume every month to keep output flat. But For comparison the same number by the start
when activity was increased from 2H-2016, of 2017 was 1.6 million b/d. The worlds oil
the legacy decline has increased to 369 kbd production is hence likely to be much more
per month. Now the US shale industry needs sensitive to the US oil market activity than
to start up 80 kbd more volume every month ever before. If well completions in the US for
to achieve any net growth in production some reason should suddenly cease, no other
compared with a year back in time. The US countries will be able to make up for those lost
shale industry has however increased the barrels after 2020 and demand would have to
monthly start-up of production by almost 300 be rationed instead (through a higher price).
kbd since 12 months back in time, so the net
production is then growing.
Going forward into 2019 and 2020 these shale

specific decline rates do however mean that it
will be much harder to maintain the growth
rate after 2018. Some call this the red queen-
effect with reference to Lewis Carrols
Through the Looking-Glass where Alice and
the red queen have to run faster and faster
just to stay in the same place.

Oil Market Outlook
12 The 2020 bunker spec be sufficient low sulfur fuel oil available from
2020 the IMO also evaluated a supplemental
change will push Brent marine fuel availability study which concluded
higher that implementing the 0.5% sulphur limit in
2020 looks unachievable.
Excactly at the time when the cancelled
projects in 2015-2016 starts to tighten the
The IMO decision will have a large impact on
global crude oil balance, the largest change in
both the global refining industry and the
global fuel specifications ever implemented
shipping community as we believe most of the
kicks in. This is likely to support the Brent
current high sulfur fuel oil demand will have to
price in our opinion.
be replaced mainly by much higher priced
distillates, supplemented by some volumes of
On October 27th 2016 the IMO (International
lower sulphur fuels. This stands is contrast to
Marine Organization) decided on an early
the CE Delft study who assumed that bunker
implementation of the Marpol Annex VI with
fuels between 0.1% and 0.5% sulphur can
January 1st 2020 as the implementation date
easily be met by a wide range of blends.
for the new sulphur content cap of 0.5%. This
will affect marine fuels for all ships trading
Its modelling suggests these blends will
outside of the sulphur Emission Control Areas
contain various residuals, treated light cycle
(ECAs).
oils, treated light distillates and kerosene etc,
and that the refineries will timely adopt to
This decision of early implementation was a
meet the new demand. From our side we do
large surprise for most market players. The
not see that global refineries will be able to
broad expectation was for implementation in
meet the new demand without ramping up
2025. The fact that implementation now will
their throughput of crude oil. There simply
be less than three years down the road will
does not look to be enough upgrading units
create large price effects for several markets
available to remove all the sulphur that is
in our opinion.
required.
The IMO regulation is the culmination of a
The easiest way to comply is to forego any
series of standards first set in motion almost
upfront capital investments and burn a
two decades ago requiring vessels operating
compliant low sulfur fuel at a much higher
in certain costal ECAs to burn fuel with a
price. But given how low sulphur gasoil is
maximum sulphur content of 0.1% since
already priced about 250 $/mt higher than
January 2015 while permitting the use of 3.5%
high sulphur bunker fuel in 2020, and likely to
sulphur fuel outside of ECAs.
rise further, this would have a major negative
impact on the bottom-line for most shipping
The IMO based their decision on a study lead
companies.
by CE Delft that looked at whether enough
distillates could be produced by 2020 to meet
the increased demand from the maritime
sector. The decision did not consider any of
the economic consequences of such an
increase in demand for higher value products
and while CE Delft concluded that there will

Oil Market Outlook
700
Gasoil vs Fuel Oil 3.5% Rtdm forecast based on probably not a realistic option for an industry
fwd curve and delta correlation
currently struggling with profitability. In
600
addition, the scrubber supply side is limited

500
and therefore a market share above 10% by

400
2020 is not a realistic scenario.
$/mt
300
200
The third alternative for shippers is investing
100
in dual engines capable of burning LNG and
0
Jan'07 Apr'08 Jul'09 Oct'10 Jan'12 Apr'13 Jul'14 Oct'15 Jan'17 Apr'18 Jul'19 Oct'20 Jan'22 liquid fuels. However these engines are
Historic DNB Forecast Forward
expensive and not commonly available and
would require a substantial investment not
It might however be the best option for older only from the shipping side but also in
vessels with limited service life after 2020 and infrastructure. Our understanding is that LNG-
the only option for companies with financial fuellled ships are only viable for newbuilds
constraints as other alternatives require and hence this alternative is not a good option
substantial investments. for the short time left before 2020.
On paper the best option is looks to be In order to fully understand the implications of
investing in scrubbers which would allow for the new IMO sulphur cap one needs to
continued burning of high sulphur bunker understand the basics of a refinery. Below is a
fuels. The total cost of scrubbers would simple refinery unit where the distillate tower
depend on the specific vessel. Factors such produces different products. The higher up
as vessel size, physical design, engine size, you move in the tower the higher the value of
retrofit or newbuilding, etc has to be taken into the product you produce and at the very
consideration. A retrofit would of course be bottom you find residual fuel. This residual
more expensive than a newbuild and some fuel, which normally price cheaper than crude
vessels might have more limited space than oil, is to a large extent consumed as bunker
others and therefore require additional fuel in the shipping industry. From a refinery
modifications, etc. But for some shippers this perspective bunker fuel is considered a waste
option will be the best choice as the huge product as it normally trades at a discount to
discount in bunker fuels relative to distillates its feedstock (crude oil) while gasoil or diesel
results in a short payback period if they can is a high value product always selling at a
afford the upfront investment costs. Recently premium to its feedstock (crude).
we have noted that bunker suppliers and
traders have offered shipping companies to
finance the scrubber if the shipping company
ties itself to a supply deal with the finance
provider.
Despite being an attractive option the current

market share of scrubbers is very limited and
mostly restricted to ferries and cruise ships. A
market share of 10% by 2020 would require
about 30 billion USD in investments and is

Oil Market Outlook
As every refinery would try to maximize its however include price consequenses of this
profit they would seek to maximize the yield of supply shift. How will the major shift in
higher value products and lower the yield of demand affect prices for distillates and bunker
residual fuel. On the right side of the distillate fuels?
tower you can see there are some upgrading
units that would allow a refinery to upgrade Based on evaluations of current refinery
residual fuels into higher value products. capacity including known upgrading units
These upgrading units are expensive and time coming on-stream before 2020 we believe
is about to run out for new such capacity to hit there is sufficient capacity to replace the
the market by January 2020. current bunker demand with a mix of
distillates (MGO, MDO, etc) and some form of
By feeding the residual fuel into the upgrading blended low sulphur fuel, but it will come at
units the complex refineries can increase the high price.
yield of higher value products and decrease
the amount of residual fuel. The distribution of 2020 vs 2016 - All assumed capacity additions
Increased global refinery runs 3.6 mbd (0.9 mbd pr year)
Mogas/Naphtha Diesel / Gasoil
1000 1200
LSFO
400
HSFO
900
products is also dependent on the type of Additional Fluid Catalytic Cracking (FCC)
Additional Hydro Cracking
200
200
100
600 200
-300
-1000
Additional Coking 100 400 -500
crude used as feedstock. Some crude Additional distillate flux saved 100 -100
Additional Vacuume Gasoil Desulfurization (VGO HDS) 50 -50
streams have better yields of distillates while Additional Resid Fuel Oil Desulfurization (Resid HDS) 250 -250
Total supply addition 1500 2400 900 -1300
other crudes yield higher amounts of residual Demand change from the shipping sector 2100 1000 -2600
Demand change outside the shipping sector 2000 800
fuel and thats one of the key reasons why Deficit/Surplus -500 -500 1300
different crude oils price differently.

What will the bunker spec change alone require?
A) A yearly increase in global refinery runs of 0.9 mbd in 2017-2020 - Increase in runs 2016 was only 0.2 mbd
B) A global refinery yield shift towards gasoil/diesel, probably at the expense of gasoline output
C) All the new capacity estimated above must come online
There would still not be enough capacity to "destroy" enough High Sulfur Resid Fuel
High Sulfur Resid Fuel must find it's way onshore into power generation in Africa, Middle East and Asia
The table above shows a breakdown of how

the refineries will have to respond in order to
maximize the output of middle distillates while
at the same time reduce the amount of bunker
fuel.
In order to produce another 1.2 million b/d of

straight run gasoil, refineries will have to run
The new IMO regulation forces the shipping
an additional 3.6 million b/d of crude (1 barrel
community up in the value chain and shippers
of straight run gasoil roughly requires 3
will after 2020 have to bid for higher value
barrels of average global crude as feedstock)
products.
but this will also produce 0.9 million b/d of
undesired high sulphur resid fuel. By running
One of the major arguments against the 2020
all upgrading units we believe that refineries
deadline is that the refinery industry will not be
will be able to upgrade a major part of current
able to deliver in time. The IMO study
resid fuel supply but will be left with close to
however concluded that the refining side will
1.3 million b/d of unwanted high sulphur resid
respond adequately and that all current
fuel.
bunker demand can be replaced by gasoil (or
equal distillate). The IMO study did not

Oil Market Outlook
This leftover resid fuel will have to be sold at 1,400

Gasoil forecast based on fwd curve and delta
correlation
further discount and most likely find its way to
1,200
power generators is Africa, Middle East or

1,000
Asia where sulphur regulations are much

800
more relaxed.
$/mt
600
400
In addition to the 1.2 million b/d straight run
gasoil supply, refiners also get another 1.2 200
million b/d via the different upgrading units as 0

Jan'07 Apr'08 Jul'09 Oct'10 Jan'12 Apr'13 Jul'14 Oct'15 Jan'17 Apr'18 Jul'19 Oct'20 Jan'22
well as a total of 0.9 million b/d of low sulphur Historic DNB Forecast Forward
resid fuel.
The forward curve for resid fuel (the red line in
The historical correlation between both fuel oil the graph below) is pricing much lower than
and gasoil to Dated Brent is very high and by what the historical relationship between crude
incorporating the futures curve of both oil and resid fuel suggest. That is fair in our
products we have modelled forecast prices opinion since the history of which the
based on our own Brent forecast. regression is based upon has not seen any
specification shifts like the one that will take
Dated Brent vs Gasoil place in 2020.
0.20
0.15
Fuel Oil 3.5% Rtdm forecast based on fwd curve
0.10 800
y = 0.8527x + 7E-05
and delta correlation
R = 0.9127 0.05
700
0.00
-0.40 -0.30 -0.20 -0.10 0.00 0.10 0.20 0.30 600
-0.05
-0.10 500
-0.15
$/mt
400
-0.20
300
-0.25
200
-0.30
100
Dated Brent vs 3.5% Rotterdam 0

Jan'07 Apr'08 Jul'09 Oct'10 Jan'12 Apr'13 Jul'14 Oct'15 Jan'17 Apr'18 Jul'19 Oct'20 Jan'22
0.30
Historic Forward
0.20
y = 1.074x + 0.0027 0.10

R = 0.8416
-0.40 -0.30 -0.20 -0.10

0.00
0.00 0.10 0.20 0.30 If we look back to 2008 we had a similar
-0.10
-0.20
situation where increased demand for diesel
-0.30 (distillate) helped push oil prices to an all time
-0.40 high of close to 150 $/b.
-0.50
-0.60
Back in 2008 Brent reached an all-time high of
147.5 $/b and today most analysts seem to
The below charts show our forecasted price agree that the price rally was driven mainly by
curves for gasoil and fuel oil compared to their a diesel-squeeze. In 1H-2008 there were an
forward curves. Based on our own Brent unprecedented strong demand growth for
forecast we believe that the current gasoil diesel from emerging markets, mainly China
curve should be 180-190 $/mt higher in 2020 but also countries like South Africa and Chile.
than what is currently priced in.

Oil Market Outlook
This unleased a gearing effect on crude oil

Global Refinery Throughput
demand in 1H-2008. As mentioned earlier; for 76.0 89
75.5
every 1 barrel of straight run diesel output the 88
75.0
refinery needs 3 barrels of crude if all
Million b/d
74.5
Million b/d
87
upgrading units are fully utilized. In 2008 all 74.0
the upgrading units in global oil refining 73.5 86
(cokers, crackers and desulphurization units) 73.0

85
72.5
were fully utilized already in March that year.
72.0 84
The Brent price was then about 100 $/b and Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
the margin to produce diesel (the diesel crack Source: IEA Global refinery runs Global oil demand
spread) was already a record high 25 $/b, so

all the refineries in the world had a maximum This process of running more crude through
incentive to fully utilize their upgrading units. the distillate tower provides the refiner with
more straightrun gasoil/diesel. But the
Gasoil Cracks Led Crude In 2008 increase in refinery throughput resulted in way
2008 price spike led by the diesel squeeze
40 140
too much output of resid fuel oil and there was
35 130
no demand growth for that output. Increased
refinery throughput to cover diesel demand
$/barrel crack spread
30 120
$/barrel Brent price
25 110 growth in 2008 was hence the cause behind

20 100 the price collapse of bunker fuel crack
15 90 spreads. The weakness in resid fuel prices
10
Jan'08 Feb'08 Mar'08 Apr'08 May'08 Jun'08 Jul'08 Aug'08 Sep'08
80 became an opposite mirror of the strength in
Historic Gasoil crack (LHS) Historic Brent price (RHS)
diesel prices.
Gasoil crack & Resid Fuel crack

But demand for diesel continued to grow also Becomes a mirror when all upgrading units are fully utilized
35
after March that year, so what do a refiner 25
have to do then to produce even more diesel? 15
The answer is that he has to run more crude 5

$/barrel
-5
oil through his distillate tower, since he cannot -15
buy straight run fuel oil and put into the -25
-35
upgrading units anymore, because those units
-45
Jan'07 Apr'07 Jul'07 Oct'07 Jan'08 Apr'08 Jul'08 Oct'08 Jan'09 Apr'09 Jul'09 Oct'09
had no more capacity left. Notice in the graph
below how global oil demand (grey line) fell Historic Gasoil crack Historic Resid Fuel Rtdml crack
through the whole of 2008 (the global 2008-09

economic recession had started) but that We now run the risk of facing a similar
crude demand (the green line) was shooting situation in 2020 if the increase in gasoil
up from March to July. This is the gearing demand is so large that the refineries have to
effect on crude demand we are talking about significantly ramp up their runs in order to
and which sent the Brent price from 100 $/b to cope with the incremental gasoil demand that
almost 150 $/b from March to July that year. is likely to materialize from the IMO bunker
spec change.

Oil Market Outlook
Some argued during the price spike in 2008 85%. A global increase in refinery utilization of
that it had to be a speculative bubble when 1% would currently add almost 1 million b/d to
the Brent price took off from March to July. global crude demand.
We did however not see any increase in net
length by speculative money in the period Global Refinery Utilization
when this took place. It was in fact the 88%
(1% increased utilization is worth about 1 mbd extra crude demand)
opposite as you can see in the graph below. 86%

84%
82%
Non-Commercials WTI Futures 80%
300 78%
250 76%
200 74%
150 72%
Million barrels
100 70%
50 1980 1985 1990 1995 2000 2005 2010 2015
0 Source: BP stats
-50
-100
-150 What kind of crudes will be in demand if this
-200
Jan2008 Mar2008 May2008 Jul2008 Sep2008 distillate squeeze materializes? Just like we
WTI Futures Long WTI Futures Short WTI Futures Net saw in 2008 it will not be heavy-sour crudes.
Some argue that there is not enough CDU

capacity globally to see a ramp up in global
crude demand in 2020. We would dismiss that
argument because global refineries swing
seasonally by many million barrels per day so
just postponing maintenance because gasoil
margins are spiking will add massively to
crude demand if many refiners do this at the
same time.
Global Refinery Throughput The graph above shows how Murban, a

82
lighter more distillate rich crude quality, priced
81
80 gradually more expensive vs Dubai, a heavier
79
and less sweet crude, during the diesel
Million b/d
78
77
76
squeeze in 2008.
75
74
73
72
It is also worth noting that global refinery

utilization has not been particularly high in
recent years. According to the latest set of BP
stats the global refinery utilization in 2016 was
below 83%. In 2004 and 2005 it was above

Oil Market Outlook
The market will in such a situation prefer US Lower 48 Crude production by gravity -
Percent
crude grades which are distillate rich. This
100%
means crudes that have a hight natural 90%
80%
content of distillate. These will be crude 70%
60%
50%
qualities that you do not need to run through 40%
30%
an upgrading unit in order to receive a decent 20%
10%
output of gasoil/diesel. Brent is such crude oil, 0%
Oct-2015
Nov-2015
Dec-2015
Oct-2016
Nov-2016
Dec-2016
Jan-2015
Apr-2015
Aug-2015
Sep-2015
Jun-2015
Jan-2016
Apr-2016
Jun-2016
Aug-2016
Sep-2016
Jan-2017
May-2015
May-2016
Feb-2015
Mar-2015
Feb-2016
Mar-2016
Feb-2017
Jul-2015
Jul-2016
but what about the US shale industry? Will the
shale barrels be able to contribute much in Light API > 35 Intermediate Heavy API<25
2020 if we see our expected distillate

squeeze? Unfortunately the US shale crude &
Since much of the US shale crude is not very
condensate will not be able to contribute to
suitable for covering the expected increase in
the same extent as Brent. Much of the US
distillate demand in 2020, we expect the price
shale crude is very light, and much of it can
spreads to blow out between Brent and US
be classified as condensate and not crude. No
shale wellhead prices in that year.
common definition of condensate exists, but
many would classify a crude oil with an API
number above 50 as condensate.
As you can see in the graph above the US

shale oil is mainly contributing to cover
gasoline and petchem demand. It is not great
for satisfying distillate demand. According to
the US Department of Energy as much as
70% of the US domestically produced crude is
classified as light (API above 35).

Oil Market Outlook
13 Launching a 5-year price since petroleum subsidies have been

removed or reduced in so many developing
deck in real terms countries. The expected high price in 2020
As you have probably realized by now, we hence leads to weak oil demand growth in
hold a bullish view on the Brent price for 2020. 2021-2022. If all the countries that have
We have never published a formal Brent price removed or reduced petroleum subsidies
forecast for the 2019-2022-period before but during the past two years start to put them
we are launching that in this report. back in place we may change this
assumption.
We are revising down our 2017 Brent price
from 60 $/b to 53 $/b. Our 2018 Brent price is When oil prices rose above 100 dollars per
revised down from 70 $/b to 60 $/b. We barrel and stayed above that level for almost 4
believe a price of 60 $/b will be reached by years, petroleum demand was protected by
year-end 2017 as inventories will draw down subsidies or price control mechanisms in
quicker than what consensus currently countries like, China, India, Indonesia,
believe, but with our updated higher Thailand, Malaysia, Saudi Arabia, UAE and
production base case for Libya and Nigeria Kuwait. Now these countries have different
we now only see a small stock draw for 2018, price policies in place.
not large enough to continue to push the
Brent price higher than 60 $/b for a while.
Then come 2019 and it will start to look more

contructive for oil prices again as the legacy
decline and red queen effect kicks in for US
shale oil, making it difficult to keep the growth
rate from 2018, at least without unleasing
inflationary effects on the cost side. At the
same time the global oil market starts to run
out of large project startups. We hence
believe the Brent price will improve in 2019
and we forecast it at 65 $/b. Then the fun The table above shows just some changes to subsidies the
past two years
starts in 2020 when the bunker spec
regulations are implemented in the same year
If global oil prices overshoot into 2020, the oil
as the global oil market really starts to feel the
consumers will have to bear a much larger
pinch of the low number of large project
part of this burden instead of the state
sanctioning in 2015-2017. We forecast a
budgets. We believe this translates into more
Brent price overshoot to 75 $/b for 2020, but
price elastic global oil demand growth. There
the risk is probably to the upside for that
exist no data so far to execute an analysis on
number.
how the consumers in China, India, Indonesia,
Saudi Arabia, etc will respond to higher end
Then the price again falls back into our
user prices with respect to their consumption
normalized price range of 50-70 $/b (real
of refined oil products when prices are rising
terms) after 2020. We believe oil demand
fast. Theory however suggests we should see
growth will be more price elastic going forward

Oil Market Outlook
a weaker growth in consumption if that Historical

Real $/b DNB Markets Brent Price Forecast In Real Terms
2000 39.8 Global bunker spec change coming
150 at the same time as the effects of
happened. 2001
2002
33.7
33.4
shelved projects from the 2015-16
CAPEX cuts becomes visible
2003 37.1
130
2004 48.3 The Brent price is set to again
2005 67.9 overshoot
2006 78.7 Supply respons and
2007 84.1 110 demand elasticity to
We do not believe electric vehicles will steal a 2008

2009
109.8
70.1
the price overshoot
sends the price down
again
2010 88.4 90
$/barrel
meaningful amount of oil demand growth from 2011
2012
2013
118.3
116.7
112.0
70
2014 100.8
the oil market in this 5-year perspective so 2015
2016
54.3
45.1
Price forecast 50
that is not an argument for weaker demand Q3-2017

Q4-2017
Real $/b
53
60 30 OPEC "cheating" in combination
Q1-2018 60 with large US shale growth and
growth in 2021-22 from our side. We will Q2-2018
2017
60
55 10
OPEC cuts in combination with solid
demand growth sends the price to 60 $/b
still some ramp up from old investment
decisions keeps prices from rising further
2018 60 Jan-2000 Jan-2004 Jan-2008 Jan-2012 Jan-2016 Jan-2020
make a short repetition of why electric 2019
2020
65
75 Normalized range
Historical Brent price real terms
Normalized range (real terms)
DNB Forecast real terms
2021 70
2022 60 FWD curve real terms (2% inflation)
vehicles are not important for the oil market in
this time frame later in the report.
We believe a Brent price in the region of 50-
The high price expected for 2020 will also 70 $/b would be necessary to make US shale
secure that the US shale industry fires on all oil grow enough production to meet global oil
cylinders into 2021, and coupled with weaker demand growth in a 5-10 year perspective.
demand growth we believe this is an Such a price level should also secure enough
argument to expect lower prices again in investments in the rest of the market so that
2021-2022. Our forecast for those two years global decline rates are kept in check for that
is hence 70 $/b and 60 $/b respectively. time frame. Our normalized price of 50-70 $/b
is not meant to represent an oil price view that
We have earlier used a normalized price can be used for prices after 2025. We believe
range of 60-80 $/b in nominal terms. This it is close to impossible to argue strongly for a
would suggest falling oil prices over time in view on the Brent price that far ahead.
real terms. Going forward we will instead keep
all forecasted prices in real terms and use a
normalized range of 50-70 $/b. A Brent price
of 80 $/b in 2022 would be equal to 72 $/b in
real terms using an inflation factor of 2%.
We believe it makes sense to think in real

terms when we have a forecast that stretch 5
years out in time. We will show historical
prices in real terms, deflated by US CPI, and
we will show our own forecast made in real
terms compared with a forward curve that is
deflated by 2% yearly inflation.

Oil Market Outlook
14 Oil demand will not peak by Total Global Primary Energy Demand
(Energy intensity vs GDP seen at 0.3 and average GDP growth assumed at 2.4%)
18,000
0.8% growth
2025 16,000
Million tonnes oil equivalents

14,000
2.6% growth
Everyone who has analyzed energy markets 12,000
10,000
knows that energy demand is mainly driven by 8,000
economic growth. And global economic 6,000
4,000
growth is heavily affected by mega trends like 2,000
population growth, urbanization and how 0

1965 1975 1985 1995 2005 2015 2025 2035
Total primary energy demand Forward looking primary energy demand growth
many people who step in and out of the Source: BP stats, DNB Markets
middle class.
If we then assume that oil will loose market
Even if trend line economic growth should fall share going forward
from 3.4% to 2.4%...
Oil Market Share In World Primary Energy Mix
Global GDP Growth Seen Lower Than Historicals 49%
6.0%
Historical average: 3.4% Forward looking: 2.4% 44%
5.0%
-0.3%
39%
4.0%
Global GDP growth
-1.5%
3.0% 34%
-2.0%
2.0% 29%
1.0%
24%
1965 1980 1995 2010 2025 2040
0.0%
1979 1989 1999 2009 2019 2029 2039 Source: BP stats, DNB Markets
Global GDP growth Forward GDP growth

Source: IMF, DNB Markets
The world will still see an average oil demand

and the global energy efficiency should growth of 0.5 million b/d for another ten years.
double from 0.6 to 0.3 per GDP unit
World Oil Demand
Energy Intensity vs GDP Growth Seen Lower 5,000

0.5 mbd growth
Energy unit demand growth vs GDP unit growth
1.5 4,500
Million tonnes oil equivalents
Historical average: 0.6 Forward looking: 0.3

4,000
1.0
3,500
0.5
3,000
0.0 2,500
-0.5 2,000
1,500
-1.0 1965 1980 1995 2010 2025 2040
1979 1989 1999 2009 2019 2029 2039
Source: BP stats, DNB Markets
Historical Energy intensity vs GDP grpwth Forward energy intensity vs GDP growth
Source: IMF, BP stats, DNB Markets
We will still see a yearly increase in energy

demand of 0.8% per year.

Oil Market Outlook
15 Peak oil demand does not Most of the investments that are undertaken
in the global oil market are just helping to
mean peak oil price maintain the production level where it already
We do not necessarily see peak oil demand is. If investments hence fall too much, we
as a good argument for lower oil prices. The would need a very strong negative demand
reason is that the more oil becomes the new response to make up for that, and the only
tobacco, which means that people see it as way to get such a response would be through
unethical to invest in the sector; the better it is a high oil price.
for the oil price. Even if the oil market
becomes smaller and that is negative for the Last years price behavior in the coal market
large oil companies, it woul lead to lower can probably serve as a good illustration of
investments, and as such the drop in supply how the price of a commodity can suddenly
could be larger than the drop in demand. double even if that commodity is seeing a
global reduction in demand.
We would claim that the stronger the story of
stranded assets becomes, and the more Steam Coal Price (CIF ARA)
politicans scare away investments from the oil 80
75
market, the more bullish it is for the oil market 70
itself with respect to the price effects. At least 65

USD/tonne
60
in a 5-10 year perspective we think this will be 55
the case, because peak demand is not likely 50
45
in that time frame. Recent statements from 40
the CEO of Royal Dutch Shell can serve as a 35
30
good example: May-15 Aug-15 Nov-15 Feb-16 May-16 Aug-16 Nov-16
Chief Executive Officer Ben Van Beurden An even better example is the global tobacco
says the next thing he will buy is a car that industry. UK demand for tobacco has
doesnt need either oil or gas to run. The decreased every year for at least the past 20
whole move to electrify the economy, electrify years. Yet the price of tobacco in the whole
mobility in places like northwest Europe, in the sale market has tripled in the UK since year
US, even in China, is a good thing, Van 2000.
Beurden said in an interview on Bloomberg
TV. We need to be at a much higher degree
of electric vehicle penetration-- or hydrogen
vehicles or gas vehicles -- if we want to stay
within the 2-degrees Celsius outcome.
Another example is the recent news that UK

will ban the sales of fossil fuel vehicles by
2040. 2040 is quite far out in time but news
like that is nonetheless suitable to scare away
investments in large oil projects.

Oil Market Outlook
16 Electric vehicles will not Global Fleet Of Electrical Vehicles

(Stock target for USA, China, Germany, France, Japan, India by 2025 is
12,3 million vehicles, that is less than half of the graph below)
make oil demand peak by 30
Million PHEV/BEV Fleet

25
2025 20
People like for example Tony Seba argue that 15
electrification of the transport sector will lead 10
5
to peak oil demand already by 2020. We
-
wonder if he has really done the math here or 2015 2018 2021 2024
if it is just a statement to get attention. It is not

enough to show the picture of all the horses in
If we assume that global vehicle sales of fossil
New York City in year 1900 and how that
fuel driven vehicles fall back from last years
picture changed just ten years later to
83 million to only average at 79 million the
convince us about the speed of change that is
next ten years (very unlikely when looking at
in front of us.
the trends, but let us use it for the sake of the
argument), and we use the average global
We believe the changes are coming and the
scrap rate of 4% for the current fleet, the
future definetly looks to be electric (many
global LDV fleet of fossil fuel driven vehicles
recent statements to mention from car
will increase from 1 billion in 2015 to almost
manufactorers), but it is the speed we
1.4 billion in 2025. So with a 17 fold increase
question. If global fuel demand for light duty
in the sales of electric cars from 2015-2025
vehicles starts falling around 2025 that would
and no more increases in the sales of fossil
be a very quick transition. It will take some
fuel vehicles in the same period, the number
more years until heavy duty trucking can be
of fossil fuel driven vehicles on the road still
electrified. To talk about t this issue like it has
increases by 340 million in this time frame
already affected demand for gasoline and
while electric cars increases by 28 million
diesel to any meaningful extent is just fake
vehicles. Even with this aggressive sales
news.
growth for electric vehicles the next ten years,
the size of the fossil fuel driven fleet of light
The global fleet of electric vehicles reached 2
duty vehicles grows 12 times faster.
million in 2016. This is spot on the assumption
we used in our calculation for electric vehicle
If we at the same time assume that the fossil
penetration one year ago, so no surprises
fuel driven vehicles become 2% more efficient
there. In our calculations we assume that
every year, we calculate that by 2025 the
global electric vehicle sales grow from
penetration of electric vehicles (plug-in
450.000 in 2015 to 7.650.000 in 2025. This
hybrids and pure battery vehicles) will have
would increase the global fleet of electric
stolen about 0.5 million b/d of what would
vehicles from 2 million in 2016 to 29 million by
otherwise have been fossil fuel demand
2025, which by the way is twice as high as the
(mainly gasoline).
total target for electric vehicle penetration set
by US, China, Germany, France, Japan and
India.

Oil Market Outlook
Displacement Of Fuel By Electrical Vehicles Norway Electric Vehicles Sales Market Share
(Includes plug-in and Battery electric)
(Assuming average global driving length for LDVs) 45%
0.6
40%
0.5 35%
Million b/d
30%
0.4
25%
0.3 20%
15%
0.2
10%
0.1 5%
0%
0.0 Jan2012 Jan2013 Jan2014 Jan2015 Jan2016 Jan2017
2015 2018 2021 2024
Source: Bloombertg Intelligence
We still however calculate a growth in fossil Norway only holds 0.2% of the global market
fuel demand in the global light duty fleet until share for new cars sold and the trend looks to
2025 with these assumptions, and by 2030 be down so we can only hope to be a signpost
the demand for fossil fuels in the global car for the markets that really matter and hope
fleet is still higher than in was in 2015, even that the rest of the world will follow in our
though the peak will be seen around 2025. footsteps.
Global Fuel Demand From LDV's Norway Global Car Sales Market Share
0.3%
27
26
Million b/d
25 0.2%
24
23
22 0.1%
21
20
0.0%
19 Jan2012 Jan2013 Jan2014 Jan2015 Jan2016 Jan2017
2015 2018 2021 2024 2027 2030 2033 2036 2039
The growth in the fossil fuel driven fleet has Unfortunately subsidies seem to be a very
mainly to do with population growth and a important feature to get sales growth for
growing middle class in emerging markets, electric vehicles and nobody subsidises more
particularly in Asia. than Norway for this segment. When
Netherland reduced subsidies the sales of
People in Norway seem to be misled by the alternative vehicles dropped from 42.500 in
speed of when we can expect global oil 2015 and if we annualize the Q1-2017 sales it
demand to suffer from electrification of the will end up at 20.000 for this year. Denmark
transportation sector. This is of course due to has also reduced the subsidies for electric
the fact that almost 40% of all the new cars vehicles and annualized sales of alternative
sold in Norway are now electric (plug-in plus fuel vehicles in Denmark in Q1-2017 was only
pure battery). The thing is however that this is 1.700 vehicles compared with about 5.000 in
not the case for the markets that matter for 2015.
the oil market.

Oil Market Outlook
In China, the largest market in the world for Electric vehicles are winning market share in
electric cars, the market share for electric the new cars sold but is only at 1.8% vs 1.6%
vehicles is still increasing but not as fast as in 2016. This is hence so far hardly a
before the government reduced subsidies for revolution when you look at the hard facts.
electric vehicles in January this year.
China Automobile Sales Market Share By Fuel Type
100%
China YoY Increase in Market Share For
90%
200% Electric Vehicles
80%
70%
150%
60%
50%
100%
40%
30%
50% 20%
10%
0% 0%
-50% Gasoline Diesel CNG Battery and Plug in Electric
Jan2016 Apr2016 Jul2016 Oct2016 Jan2017 Apr2017 Source: China Automotive Information
Source: China Automotive Information
When we look at the second largest market

The key global car market is in Asia. Total car for electric vehicles in the world, the US
sales in Asia are twice as large as Europe and market, we do not see a revolution taking
the US put together. place there either. Total car sales are strong
at about 1.4 million units sold per month so far
Motor Vehicle Sales this year. For the US we now have sales for
4,500 the first half of the year, and annualized the
Thousand units Sold per month
4,000
3,500
sales are on track to come in at about 17
3,000 million units.
2,500
2,000
1,500 USA Total Automobile Sales
1,000 1,900
Thousand units Sold per month
500
1,700
0
Jan2005 Jan2007 Jan2009 Jan2011 Jan2013 Jan2015 Jan2017 1,500
USA Europe Asia China
Source: Bloomberg Intelligence 1,300
1,100
Chinese car sales alone are larger than that of 900
both Europe and the US. Total sales of 700
electric vehicles in China have reached 500

Jan2005 Jan2007 Jan2009 Jan2011 Jan2013 Jan2015 Jan2017
162.000 units so far this year. In the same Source: Bloombertg Intelligence
period total sales of pure gasoline vehicles

have been 9.1 million vehicles. The data so Sales of electric vehicles in the US for 1H-
far hence definetly suggest that the global car 2017 were 87.000 units, so annualized almost
fleet for fossil fuel driven vehicles still increase 175.000 units. This suggests that the market
a lot faster than the fleet of electric vehicles. share of electric vehicles in the US should
come in at about 1% in 2017. The market
In fact the market share for gasoline powered share in 2016 was 0.9%. This is hence quite
cars in China has averaged at 97.2% so far in far off from the revolution in electric vehicle
2017, while the average in 2016 was 96.8%. sales that we see in Norway.

Oil Market Outlook
The US market share for Light Trucks, SUVs will hence not be a meaningful size for the
and pick up trucks which are almost only run global supply-demand balance in the oil
on gasoline, continue to increase and has market for the 5-year view that we are
reached 64% of all the new cars sold in the providing in this report. Come 2030 and the
US. situation could be very much changed but that
far out is not the scope for this report.
USA Light Truck Sales Market Share
70% Instead one could argue, like we did in
65% chapter 14, that the more people believe in
60% peak oil demand from electric vehicle
55%
penetration, the better it is for the oil price.
50%
45%
Oil prices start falling

40%
When comparing the market share for Light

Trucks (which are almost all gasoline driven),
with the market share for electric vehicles in
the US, it is easy to conclude that people in
Norway live in a different world than what is
going on in the markets that really matter for
oil demand.
USA Motor Vehicle Sales Market Share

70%
60%
50%
40%
30%
20%
10%
0%
Share Light Trucks Share Plug-in Hybrid & BEV

It is always important to put things into

perpective and we hope that the perspective
we have provided in this report have
enlightened the reader to see that since the
sales of fossil fuel driven vehicles will still be
much larger than the sales of electric vehicles
also the next 5-6 years, global oil demand will
not be meaningfully affected in that time frame
by the electrification of the transport sector. It

Oil Market Outlook
17 Low unplanned outages returning from that country in the coming 6-12
months. Nigeria could bring in 100 kbd more
means the risk is now bullish but The Niger Delta Revolutionary
Many would probably agree that geopolitical Crusaders are the latest group to threaten to
risk is still high and has been high for most of resume attacks on oil infrastructure as
the year so far. The difference from the start negotiations with the federal government have
of the year and now is that right now many of so far failed to deliver economic benefits in
the unplanned outages have returned to the the country. We see no reason to expect
market. The prime examples are production stable production from Nigeria going forward.
from both Libya and Nigeria.
Colombia is struggling with militant attacks
Since we knew at the start of the year that and public opposition to the oil sector. Despite
there was limited downside risk from those ongoing peace talks, attacks on oil
two countries as they already were producing infrastructure have continued. The 0.22 million
way below capacity, we could not score b/d Cao Limn-Coveas pipeline has been
geopolitical risk bullish with a high weight offline since may 30th as a result of repeated
then. But now we can do that since both Libya bombings. The 85 kbd Transandino pipeline
and Nigeria have returned with so many was also shut by vandalism on July 15th.
barrels since April. Several municipalities in oil-producing regions
are holding referendums on whether to allow
Current global unplanned outages are now at production to continue because of
the lowest level since early 2011. environmental and economic concerns. A total
of 0.12 million b/d of production could be at
risk because of these referendums. The
4.0
Global Unplanned Supply Outages problems in Colombia serve to illustrate that
Venezuela
3.5 UAE there is still high risk to oil supplies around the
Nigeria
3.0 Neutral Zone
Libya
world.
2.5 Kuwait
Million b/d
Iraq
2.0 Iran
Gabon The largest potential unplanned supply
1.5 Ecuador
1.0
Angola
Yemen
disruption could be coming from Venezuela of
0.5
US
Thailand
course. Venezuela held a controversial
Syria
0.0 South Sudan election to appoint a constituent assembly on
Jan-14 Jul-14 Jan-15 Jul-15 Jan-16 Jul-16 Jan-17 Jul-17 Peru
Source: Energy Aspects
Sunday the 30th of July despite domestic
opposition and the threat of US sanctions. US
This means that from here on the risk looks sanctions could divert Venezuelan oil flows.
clearly skewed to the upside. Venezuelan oil would then probably have to
travel to Asia. The US administration
In July the largest unplanned outages were in announced sanctions against President
Colombia (100 kbd), Syria (150 kbd), Yemen Maduro earlier this week but have imposed no
(140 kbd), the Neutral Zone (480 kbd), oil-related sanctions so far. Various measures
Venezuela (210 kbd) and Nigeria (100 kbd). have been under consideration, including
cutting off US exports of diluants to the
Libya is probably now at its potential for 2H- country. Venezuela needs this light oil to
2017 so we are not afraid of more barrels blend with its own heavy oil in order to make

Oil Market Outlook
the crude more tradable. US officials stress

that sanctions against the oil sector could be
imposed in the future. Output is already
declining in Venezuela and has fallen from
almost 2.5 million b/d two years ago to 2.0
million b/d in June this year with the latest
IEA-numbers.
Venezuela Oil Production

3.2
3.0
2.8
2.6
Million b/d
2.4
2.2
2.0
1.8
1.6
Source: IEA
Rising US-Venezuelan tensions can also

complicate the upstream and investment
outlook for the country.
Peak unplanned outages last year were

above 3 million b/d. If unplanned outages
should increase back to that level again from
the current 1.35 million b/d it would be very
bullish to the global oil supply-demand
balance which is currently already in a deficit
(global supply is lower than global demand
despite the low current unplanned outages).

Oil Market Outlook
18 DNB Markets Brent price forecast & consensus forecast
Historical
Real $/b DNB Markets Brent Price Forecast In Real Terms
2000 39.8 Global bunker spec change coming
150 at the same time as the effects of
2001 33.7
shelved projects from the 2015-16
2002 33.4 CAPEX cuts becomes visible
2003 37.1
130
2004 48.3 The Brent price is set to again
2005 67.9 overshoot
2006 78.7 Supply respons and
2007 84.1 110 demand elasticity to
2008 109.8 the price overshoot
sends the price down
2009 70.1
again
2010 88.4 90
$/barrel
2011 118.3
2012 116.7
2013 112.0
70
2014 100.8
2015 54.3
2016 45.1
Price forecast 50
Real $/b
Q3-2017 53
Q4-2017 60 30 OPEC "cheating" in combination
Q1-2018 60 with large US shale growth and
OPEC cuts in combination with solid still some ramp up from old investment
Q2-2018 60 demand growth sends the price to 60 $/b decisions keeps prices from rising further
2017 55 10
2018 60 Jan-2000 Jan-2004 Jan-2008 Jan-2012 Jan-2016 Jan-2020
2019 65
2020 75 Normalized range Normalized range (real terms)
2021 70 Historical Brent price real terms DNB Forecast real terms
2022 60 FWD curve real terms (2% inflation)
Bloomberg consensus forecast:
Bloomberg survey Brent $/b 2017 2018 2019 2020 2021

Median 54 58 61 65 68
Mean 54 57 60 64 67
High 64 75 84 90 90
Low 45 37 39 47 47
Forward Market 52 53 53 54 55

Oil Market Outlook
19 Historical Brent prices, trading range & time spreads

Platts ICE Brent ICE Brent ICE Brent ICE Brent ICE Brent ICE Brent Contango/Backwardation
Brent Dated Average Max Min Trading range Trading range % Price change Green if backw, red if contango
1995 17.0 16.9 19.2 15.6 3.6 21% 1.1 0.07
1996 20.7 20.3 24.9 16.2 8.6 43% 3.4 0.35
1997 19.1 19.3 24.8 16.5 8.3 43% -1.0 -0.23
1998 12.7 13.3 16.6 9.6 7.0 52% -6.0 -0.62
1999 18.0 18.0 26.1 10.1 16.0 89% 4.7 -0.06
2000 28.5 28.5 34.6 21.3 13.3 47% 10.5 -0.04
2001 24.4 24.9 29.9 17.7 12.2 49% -3.7 -0.42
2002 25.0 25.0 30.2 18.4 11.8 47% 0.2 -0.01
2003 28.8 28.5 34.1 23.3 10.8 38% 3.4 0.35
2004 38.3 38.0 51.6 28.8 22.7 60% 9.6 0.22
2005 54.5 55.3 67.7 40.5 27.2 49% 17.2 -0.73
2006 65.1 66.1 78.3 57.9 20.4 31% 10.8 -0.95
2007 72.4 72.7 95.8 51.7 44.1 61% 6.6 -0.27
2008 97.3 98.5 146.1 36.6 109.5 111% 25.9 -1.27
2009 61.7 62.7 79.7 39.6 40.1 64% -35.8 -1.00
2010 79.5 80.3 94.8 69.6 25.2 31% 17.7 -0.84
2011 111.3 110.9 126.7 93.3 33.3 30% 30.6 0.34
2012 111.7 111.7 126.2 89.2 37.0 33% 0.8 -0.01
2013 108.7 108.7 118.9 97.7 21.2 20% -3.0 -0.05
2014 98.9 99.4 115.1 57.3 57.7 58% -9.3 -0.50
2015 52.4 53.6 67.8 36.1 31.7 59% -45.9 -1.21
2016 43.7 45.1 56.8 27.9 28.9 64% -8.5 -1.40
2017 51.3 52.2 57.1 44.8 12.3 24% 7.1 -0.95
Q1-2016 33.9 35.2 41.8 27.9 13.9 40% -9.5 -1.27

Q2-2016 45.6 47.0 52.5 37.7 14.8 32% 11.8 -1.44
Q3-2016 45.9 47.0 50.9 41.8 9.1 19% 0.0 -1.13
Q4-2016 49.3 51.1 56.8 44.4 12.4 24% 4.1 -1.74
Q1-2017 53.7 54.6 57.1 50.6 6.5 12% 3.5 -0.92
Q2-2017 49.6 50.8 56.2 44.8 11.4 22% -3.8 -1.15
Q3-2017 48.7 49.4 52.7 46.7 5.9 12% -1.4 -0.72

Oil Market Outlook
20 Global supply-/demand balance DNB, IEA, OPEC, EIA

DNB Markets World Oil Supply-Demand Balance: 2010 Change 2011 Change 2012 Change 2013 Change 2014 Change 2015 Change 2016 Change 2017 Change 2018
OECD Demand 47.0 -0.6 46.4 -0.5 46.0 0.1 46.1 -0.4 45.7 0.7 46.4 0.4 46.8 0.2 47.1 0.0 47.0
Non-OECD Demand 41.3 1.3 42.6 1.7 44.3 1.4 45.6 1.5 47.2 1.4 48.6 1.1 49.7 1.2 50.9 1.4 52.3
Total Demand 88.3 0.7 89.0 1.2 90.2 1.5 91.7 1.1 92.9 2.1 95.0 1.6 96.6 1.4 97.9 1.4 99.3
Non-OPEC Supply 50.2 0.0 50.2 0.6 50.8 1.4 52.2 2.3 54.5 1.4 55.9 -0.8 55.0 0.6 55.7 1.4 57.1
OPEC NGL's and non-conventional oil 5.5 0.4 5.9 0.4 6.2 0.0 6.2 0.2 6.4 0.2 6.6 0.2 6.8 0.1 6.9 0.1 7.0
Global Biofuels 1.8 0.0 1.8 0.0 1.9 0.2 2.1 0.2 2.2 0.1 2.3 0.1 2.3 0.0 2.4 0.1 2.5
Total Non-OPEC supply 57.5 0.4 57.9 0.9 58.9 1.6 60.5 2.6 63.1 1.7 64.8 -0.6 64.2 0.8 65.0 1.6 66.6
Call on OPEC crude (and stocks) 30.8 0.2 31.1 0.3 31.4 -0.1 31.2 -1.4 29.8 0.5 30.3 2.1 32.4 0.6 33.0 -0.2 32.8
OPEC Crude Oil Supply 29.7 0.6 30.2 1.5 31.7 -0.9 30.8 -0.1 30.7 1.1 31.8 1.0 32.8 -0.4 32.4 0.1 32.5
Implied World Oil Stock Change -1.2 -0.9 0.4 -0.4 0.9 1.6 0.4 -0.6 -0.3
IEA World Oil Supply-Demand Balance (July 2017): 2010 Change 2011 Change 2012 Change 2013 Change 2014 Change 2015 Change 2016 Change 2017 Change 2018
OECD Demand 47.0 -0.6 46.4 -0.5 46.0 0.1 46.1 -0.4 45.7 0.7 46.4 0.4 46.8 0.2 47.0 0.0 47.0
Non-OECD Demand 41.3 1.3 42.6 1.7 44.3 1.4 45.6 1.5 47.2 1.4 48.6 1.1 49.7 1.2 50.9 1.4 52.3
Total Demand 88.3 0.7 89.0 1.2 90.2 1.5 91.7 1.1 92.9 2.1 95.0 1.6 96.6 1.4 98.0 1.4 99.4
Non-OPEC Supply 50.2 0.0 50.2 0.6 50.8 1.4 52.2 2.3 54.5 1.4 55.9 -0.8 55.0 0.6 55.7 1.3 57.0
OPEC NGL's and non-conventional oil 5.5 0.4 5.9 0.4 6.2 0.0 6.2 0.2 6.4 0.2 6.6 0.2 6.8 0.1 6.9 0.1 7.0
Global Biofuels 1.8 0.0 1.8 0.0 1.9 0.2 2.1 0.2 2.2 0.1 2.3 0.1 2.3 0.1 2.4 0.1 2.5
Total Non-OPEC supply 57.5 0.4 57.9 0.9 58.9 1.6 60.5 2.6 63.1 1.7 64.8 -0.6 64.2 0.8 65.0 1.5 66.5
OPEC Crude Oil Supply (DNB OPEC scenario) 29.7 0.6 30.2 1.5 31.7 -0.9 30.8 -0.1 30.7 1.1 31.8 1.0 32.8 -0.4 32.4 0.1 32.5
Implied World Oil Stock Change -1.2 -0.9 0.4 -0.4 0.9 1.6 0.4 -0.6 -0.4
OPEC World Oil Supply-Demand Balance (July 2017): 2010 Change 2011 Change 2012 Change 2013 Change 2014 Change 2015 Change 2016 Change 2017 Change 2018
OECD Demand 47.0 -0.6 46.4 -0.5 45.9 0.2 46.1 -0.3 45.8 0.6 46.4 0.5 46.9 0.2 47.1 0.2 47.3
Non-OECD Demand 40.3 1.5 41.8 1.4 43.2 1.8 45.0 1.2 46.2 1.1 47.3 0.9 48.2 1.1 49.3 1.0 50.3
Total Demand 87.3 0.9 88.2 0.9 89.1 2.0 91.1 0.9 92.0 1.7 93.7 1.4 95.1 1.3 96.4 1.2 97.6
Non-OPEC Supply (Incl all Biofuel) 52.4 0.0 52.4 -0.5 51.9 2.3 54.2 2.0 56.2 1.5 57.7 -0.7 57.0 0.8 57.8 1.2 59.0
OPEC NGL's and non-conventional oil 5.0 0.4 5.4 0.3 5.7 -0.1 5.6 0.3 5.9 0.1 6.0 0.1 6.1 0.2 6.3 0.2 6.5
Total Non-OPEC supply 57.4 0.4 57.8 -0.2 57.6 2.2 59.8 2.3 62.1 1.6 63.7 -0.6 63.1 1.0 64.1 1.4 65.5
Implied World Oil Stock Change -0.2 -0.2 0.2 -0.8 0.4 1.5 0.5 0.1 0.4
EIA World Oil Supply-Demand balance (July 2017): 2010 Change 2011 Change 2012 Change 2013 Change 2014 Change 2015 Change 2016 Change 2017 Change 2018
OECD Demand 46.1 -0.3 45.8 0.1 45.9 0.2 46.1 -0.3 45.8 0.7 46.4 0.4 46.9 0.3 47.2 0.4 47.5
Non-OECD Demand 41.0 1.5 42.5 0.8 43.3 1.2 44.4 2.3 46.7 1.0 47.7 2.4 50.1 1.2 51.2 1.2 52.5
Total Demand 87.1 1.2 88.3 0.9 89.2 1.3 90.5 2.0 92.4 1.6 94.1 2.9 96.9 1.5 98.4 1.6 100.0
Non-OPEC Supply (Incl all Biofuel) 51.8 0.2 52.0 0.7 52.7 1.5 54.1 2.8 56.9 0.6 57.5 0.5 57.9 1.0 58.9 1.2 60.1
OPEC NGL's and non-conventional oil 5.5 -0.3 5.3 0.5 5.8 0.4 6.1 0.1 6.3 0.3 6.6 0.0 6.6 0.4 6.9 0.2 7.1
Total Non-OPEC supply 57.3 -0.1 57.2 1.2 58.4 1.8 60.2 2.9 63.2 0.9 64.0 0.5 64.5 1.3 65.8 1.4 67.2
Call on OPEC crude (and stocks) 29.8 1.3 31.1 -0.3 30.8 -0.5 30.2 -1.0 29.3 0.8 30.0 2.4 32.4 0.1 32.6 0.2 32.8
Implied World Oil Stock Change -0.1 -0.8 1.0 0.6 1.4 1.7 0.1 -0.2 -0.3

Oil Market Outlook
21 DNB Markets OPEC assumption & recent OPEC exports
Scenario Oct/2016 Nov/2016 Dec/2016 Jan/2017 Feb/2017 Mar/2017 Apr/2017 May/2017 Jun/2017 Jul/2017 Aug/2017 Sep/2017 Oct/2017 Nov/2017 Dec/2017
Algeria 1130 1120 1120 1,050 1,050 1,050 1,060 1,060 1,060 1,060 1,060 1,060 1,060 1,060 1,060
Angola/Cabinda 1510 1690 1640 1,630 1,650 1,640 1,660 1,610 1,670 1,670 1,670 1,670 1,670 1,670 1,670
Ecuador 540 550 540 525 525 520 530 530 530 530 530 530 530 530 530
Equatorial Guinea 140 130 127 126 130 135 122 130 120 120 120 120 120 120 120
Gabon 220 230 210 200 200 195 200 200 200 200 200 200 200 200 200
Iran 3850 3800 3780 3,750 3,810 3,790 3,750 3,780 3,790 3,790 3,790 3,790 3,790 3,790 3,790
Iraq 4590 4650 4680 4,480 4,440 4,440 4,460 4,480 4,500 4,500 4,500 4,500 4,500 4,500 4,500
Kuwait 2930 2830 2810 2,710 2,710 2,700 2,710 2,720 2,720 2,720 2,720 2,720 2,720 2,720 2,720
Libya 510 580 620 690 670 610 550 740 820 1,000 900 800 800 800 800
Neutral Zone 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Nigeria 1450 1530 1390 1,430 1,450 1,300 1,380 1,530 1,590 1,600 1,500 1,500 1,500 1,500 1,500
Qatar 630 650 630 610 590 610 620 630 620 620 620 620 620 620 620
Saudi Arabia 10560 10640 10450 9,800 9,980 9,930 9,960 9,920 10,050 10,050 10,050 10,050 10,050 10,050 10,050
United Arab Emirates 3120 3130 3140 2,990 2,930 2,910 2,930 2,930 2,930 2,930 2,930 2,930 2,930 2,930 2,930
Abudhabi 3120 3130 3140 2,990 2,930 2,910 2,930 2,930 2,930 2,930 2,930 2,930 2,930 2,930 2,930
Dubai 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Sharjah/Ras Al Khaimah 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Venezuela 2150 2120 2100 2,050 2,050 2,030 2,020 2,000 2,000 2,000 2,000 2,000 2,000 2,000 2,000
OPEC 33,330 33,650 33,237 32,041 32,185 31,860 31,952 32,260 32,600 32,790 32,590 32,490 32,490 32,490 32,490
OPEC exports counted by Platts cFlow:

Extracted August 2
Confidence level Crude exports Dec/2016 Jan/2017 Feb/2017 Mar/2017 Apr/2017 May/2017 Jun/2017 Jul/2017
Moderate Algeria 608 528 515 652 667 670 531 524
High Angola/Cabinda 1,672 1,519 1,699 1,495 1,628 1,545 1,669 1,440
High Ecuador 444 350 449 369 393 400 494 465
Moderate Gabon 193 184 203 166 222 229 186 254
High Iran 2,474 2,543 2,724 2,500 2,217 2,511 2,640 2,649
High Iraq 3,545 3,323 3,096 3,269 3,091 3,604 2,976 3,107
High Kuwait 2,445 1,986 2,074 2,120 2,041 2,259 2,137 1,807
Moderate Qatar 659 443 716 679 557 584 623 529
High Saudi Arabia 7,566 6,806 7,373 7,503 7,031 6,494 6,878 6,870
Moderate United Arab Emirates 3,101 3,144 3,015 3,112 2,896 3,025 2,858 2,966
Moderate Venezuela 1,735 1,548 1,798 1,669 1,587 1,645 1,440 1,597
Sum of participating countries 24,442 22,374 23,662 23,534 22,330 22,966 22,432 22,208
3-month moving average OPEC 11 24,581 23,977 23,493 23,190 23,175 22,943 22,576 22,535
Change since October 2016 363 -241 -725 -1,028 -1,043 -1,275 -1,642 -1,683
Year on year OPEC 11 2,843 1,470 860 50 66 -506 -949 -914
Moderate Libya 587 757 654 524 563 687 640 640
High Nigeria 1,646 1,896 1,850 1,698 1,539 1,839 1,808 1,808
Sum Libya/Nigeria: 2,233 2,653 2,504 2,222 2,102 2,526 2,448 2,448
OPEC including Libya/Nigeria 26,675 25,027 26,166 25,756 24,432 25,492 24,880 24,656
3-month moving average total OPEC 26,940 26,462 25,956 25,650 25,451 25,227 24,935 25,009
Year on Year total OPEC exports 2,539 1,348 786 64 50 -401 -698 -482
Year on Year total OPEC production (IEA) 1,104 -514 -49 -204 -510 -30 -182
Diff exports vs production -1,435 -1,862 -835 -268 -560 371 516

Oil Market Outlook
22 Calculated Saudi Arabia oil exports capacity

Saudi Arabia crude production capacity
Max Saudi Arabia crude production (June-2008 output) (Including neutral zone) 9,500
2008-2016 field decline 3.0 % -2,278
Startup new crude projects 2008-2017:
Khursaniyah 500
AFK Phase 2 200
Khurais Exp Phase 1 600
Nuayyim Expansion 100
Shaybah phase 2 250
Khurais expansion phase 2 600
Neutral zone expansion 30
Manifa phase 1 500
Manifa phase 2 400
Shaybah expansion 250
Added spare capacity from June 2008 (see note below)*: 1,200
Sum gross new projects Saudi Arabia: 4,630
Saudi Arabia capacity 2017 (capacity summer 2008 less field decline 2008-2017 plus additions to capacity 2008-2017): 11,852
Saudi Araba domestic demand in 2008 2,272
Saudi Arabia domestic demand in 2016 3,190
Increase in demand 2008-2017: 918
Increase in exports capacity 2008-2017: 2,352
Calculated exports capacity increase 2009-2017: 1,434
Reported increase in crude & product exports from Saudi Arabia 2009-2017 2,300
* This was not real spare in 2008 due to the refining b ottleneck b ut should now b e included as the b ottle neck in refining is not there anymore

Oil Market Outlook
23 DNB Markets global oil demand assumptions by region

DnB Demand (YoY-changes) 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
North America (Canada, Mexico) 200 -100 -125 134 71 32 -33 -132 -36 26 -30 0
US -20 -1,243 -723 409 -227 -457 472 147 429 106 123 0
Europe -265 -147 -724 -3 -480 -425 -234 -135 305 215 180 72
Australia, New Zealand, Japan, Korea 19 -308 -384 163 48 374 -82 -249 -7 69 -50 -82
Total OECD -66 -1,798 -1,957 704 -588 -475 123 -370 691 416 222 -10
Europe/Africa Med & FSU -10 221 -111 65 271 148 -57 197 140 225 138 166
Middle East AG/Asia Pacific/East Africa 1,111 521 1,031 1,775 838 1,223 1,244 1,147 1,316 1,016 965 1,166
Middle East AG excl. Iran and Saudi 48 230 134 87 70 178 90 124 9 98 76 57
Iran 51 44 73 -197 -52 64 203 -6 -52 -42 27 39
Saudi Arabia 87 152 196 205 115 192 -17 227 106 -91 -10 102
Asia Pacific/East Africa excl. China and India 319 -101 319 537 164 221 327 241 338 352 370 399
China 381 124 268 1,034 373 392 599 407 762 412 339 300
India 227 72 43 108 169 175 42 153 153 286 164 269
West Africa 65 82 -16 78 40 40 76 -6 41 23 59 51
Latin America (excl. Mexico) 150 351 17 446 124 273 113 182 -62 -123 -8 37
Total Non-OECD 1,316 1,175 922 2,364 1,272 1,684 1,375 1,519 1,435 1,140 1,155 1,420
North America 180 -1,343 -848 544 -156 -425 439 15 393 132 92 0
Europe/Africa Med & FSU -274 74 -835 62 -209 -276 -291 61 445 439 319 238
Middle East AG/Asia Pacific/East Africa 1,130 213 647 1,938 885 1,597 1,162 897 1,309 1,084 915 1,084
Middle East AG 185 426 402 96 133 435 276 346 63 -35 92 198
Asia Pacific/East Africa 945 -213 245 1,843 753 1,163 886 552 1,246 1,119 823 886
West Africa 65 82 -16 78 40 40 76 -6 41 23 59 51
Latin America (excl. Mexico) 150 351 17 446 124 273 113 182 -62 -123 -8 37
Total World 1,250 -623 -1,035 3,068 685 1,209 1,498 1,149 2,126 1,556 1,377 1,410

Oil Market Outlook
24 Oil price score card for 2018
2017 Oil Price Scorecard Comments Oil Price Weight
We believe the Brent price will increase to 60 $/b before year end. It will
however struggle to take any further upward leaps in 2018 as the "Call on
OPEC" is decreasing due to large expected growth in non-OPEC supply,
Overall Outlook mainly coming from the US, Brazil and Canada. Demand growth could 60 $/b
however surprise to the upside and geopolitical risk is a bullish factor. If
OPEC discipline deteriorates and the current deal falls apart, prices will be
lower than what we currently predict.
Fundamentals
The global balance suggest that supply will be lower than demand also in 2018 if OPEC stays
Global Fundamental Balance disciplined, but the "Call on OPEC" is reduced from the average 2017 level.
NEUTRAL MEDIUM
Refinery margins have surprised to the upside in 2017 and may stay fairly strong also in 2018
since we believe crude prices will stay capped at 60 $/b. Our demand growth estimate of 1.4
Crude vs Product Balance (Margins) million b/d is conservative as global macro economic indicators look strong. Any upside surprise to
NEUTRAL MEDIUM
demand growth should benefit refinery margins.
Stock levels will have drawn down significantly by the start of 2018, but will struggle to draw much
OECD Stock levels more next year.
NEUTRAL MEDIUM
OPEC spare capacity is still low despite recent volontary production cuts. We calculate Saudi
OPEC Spare Capacity Arabia's total production capacity to be about 11.8 million b/d vs current output of about 10 million NEUTRAL MEDIUM
b/d.
US oil production is estimated to grow 1.2 million b/d in 2018, twice as fast as in 2017. We also
expect zero oil demand growth in the US for 2018 compared with 2017. The US oil balance is
US Oil Statistics - Fundamentals hence bearish for 2018 and is maybe the most important factor that makes us sceptical to further
BEARISH HIGH
price increases until we reach 2019.
Global oil demand growth is estimated to be at the average level for oil intensity vs global GDP
growth for next year (0.4). Macro economic indicators have however been solid recently and there
Global Demand Growth may be further upside to global economic growth than what is currently predicted. IMF predicts
NEUTRAL MEDIUM
global GDP growth of 3.6% in 2018, up slightly from the 3.5% predicted for 2017.
We believe OPEC will stay disciplined into 2018 and in fact cut production more if necessary to
protect oil prices. We believe Saudi Arabia would like to see oil prices at 60 $/b or higher when the
OPEC Supply planned listing of Saudi Aramco is to take place during next year. We are not a subscriber to the
BULLISH HIGH
thesis that the OPEC deal will collapse and fall apart.
Non-OPEC supply will grow significantly in 2018. We are predicting a growth of 1.4 million b/d, but
you can add another 0.2 million b/d to that if OPEC NGLs and global biofuels are included. Most of
Non-OPEC Supply the growth will be coming from US shale, but Canada/Brazil/Kasakhstan are estimated to grow 0.5
BEARISH HIGH
million b/d and makes up the other significant chunk of the supply growth story.
Political Risk
Now that Nigeria and Libya have increased production by about 0.7 million b/d since April, the risk
has turned to the downside for those two countries when assessing the future output. For
Venezuela the risk is very much on the downside for production as the country's economy is about
to collapse and there is in fact a rsik for civil war-like conditions. The "palace coup" in Saudi Arabia
Venezuela, Iraq, Iran, Saudi, Nigeria, has increased the risk for unrest there to. Much will depend on how the current head of the Saudi
National Guard, the son of the former king Abdullah, will handle the situation if king Salman tries to
BULLISH HIGH
Russia, Israel, MENA, Brazil, etc
put the control over the National Guard under the army which is controlled by his son the newly
appointed crown price Mohammad bin-Salman. Global unplanned outages are at the lowest since
early 2011 and this makes it easy to score political risk to supply as a bullish factor with high
weight.
Other Factors
It is hard to predict this factor for 2018. If OPEC succeeds to bring the structure of the forward
market into backwardation, and if the USD continue to weaken, the financial money flow can
Financial Money Flow contribute to higher oil prices in 2018. But if OPEC discipline deteriorates and the USD starts to
NEUTRAL MEDIUM
strengthen the financial money flow can provide headwinds.

Oil Market Outlook
Important/Disclaimer
This note (the Note) must be seen as marketing material and not as an investment recommendation within the meaning of the
Norwegian Securities Trading Act of 2007 paragraph 3-10 and the Norwegian Securities Trading Regulation 2007/06/29 no. 876. The
Note has been prepared by DNB Markets, a division of DNB Bank ASA, a Norwegian bank organized under the laws of the Kingdom of
Norway (the Bank), for information purposes only. The Note shall not be used for any unlawful or unauthorized purposes. The Bank, its
affiliates, and any third-party providers, as well as their directors, officers, shareholders, employees or agents (individually, each a DNB
Party; collectively, DNB Parties) do not guarantee the accuracy, completeness, timeliness or availability of the Note. DNB Parties are
not responsible for any errors or omissions, regardless of the cause, nor for the results obtained from the use of the Note, nor for the
security or maintenance of any data input by the user. The Note is provided on an as is basis. DNB PARTIES DISCLAIM ANY AND
ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR
FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE
NOTES FUNCTIONING WILL BE UNINTERRUPTED OR THAT THE NOTE WILL OPERATE WITH ANY SOFTWARE OR
HARDWARE CONFIGURATION. In no event shall DNB Parties be liable to any party for any direct, indirect, incidental, exemplary,
compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost
income or lost profits and opportunity costs) in connection with any use of the Note, even if advised of the possibility of such damages.
Any opinions expressed herein reflect the Banks judgment at the time the Note was prepared and DNB Parties assume no obligation to
update the Note in any form or format. The Note should not be relied on and is not a substitute for the skill, judgment and experience of
the user, its management, employees, advisors and/or clients when making investment and other business decisions. No DNB Party is
acting as fiduciary or investment advisor in connection with the dissemination of the Note. While the Note is based on information
obtained from public sources that the Bank believes to be reliable, no DNB Party has performed an audit of, nor accepts any duty of due
diligence or independent verification of, any information it receives. Confidentiality rules and internal rules restrict the exchange of
information between different parts of the Bank and this may prevent employees of DNB Markets who are preparing the Note from
utilizing or being aware of information available in DNB Markets/the Bank that may be relevant to the recipients of the Note.
The Note is not an offer to buy or sell any security or other financial instrument or to participate in any investment strategy. Distribution
of material like the Note is in certain jurisdictions restricted by law. Persons in possession of the Note should seek further guidance
regarding such restrictions before distributing the Note.
The Note is for clients only, and not for publication, and has been prepared for information purposes only by DNB Markets - a division of
DNB Bank ASA registered in Norway with registration number NO 984 851 006 (the Register of Business Enterprises) under
supervision of the Financial Supervisory Authority of Norway (Finanstilsynet), the Monetary Authority of Singapore, and on a limited
basis by the Financial Conduct Authority and the Prudential Regulation Authority of the UK, and the Financial Supervisory Authority of
Sweden. Details about the extent of our regulation by local authorities outside Norway are available from us on request. Information
about DNB Markets can be found at www.dnb.no/markets.
Additional information for clients in Singapore
The Note has been distributed by the Singapore Branch of DNB Bank ASA. It is intended for general circulation and does not take into
account the specific investment objectives, financial situation or particular needs of any particular person. You should seek advice from
a financial adviser regarding the suitability of any product referred to in the Note, taking into account your specific financial objectives,
financial situation or particular needs before making a commitment to purchase any such product.
You have received a copy of the Note because you have been classified either as an accredited investor, an expert investor or as an
institutional investor, as these terms have been defined under Singapores Financial Advisers Act (Cap. 110) (FAA) and/or the
Financial Advisers Regulations (FAR). The Singapore Branch of DNB Bank ASA is a financial adviser exempt from licensing under the
FAA but is otherwise subject to the legal requirements of the FAA and of the FAR. By virtue of your status as an accredited investor or
as an expert investor, the Singapore Branch of DNB Bank ASA is, in respect of certain of its dealings with you or services rendered to
you, exempt from having to comply with certain regulatory requirements of the FAA and FAR, including without limitation, sections 25,
27 and 36 of the FAA. Section 25 of the FAA requires a financial adviser to disclose material information concerning designated
investment products which are recommended by the financial adviser to you as the client. Section 27 of the FAA requires a financial
adviser to have a reasonable basis for making investment recommendations to you as the client. Section 36 of the FAA requires a
financial adviser to include, within any circular or written communications in which he makes recommendations concerning securities, a
statement of the nature of any interest which the financial adviser (and any person connected or associated with the financial adviser)
might have in the securities.
Please contact the Singapore Branch of DNB Bank ASA at +65 6212 6144 in respect of any matters arising from, or in connection with,
the Note.
The Note is intended for and is to be circulated only to persons who are classified as an accredited investor, an expert investor or an
institutional investor. If you are not an accredited investor, an expert investor or an institutional investor, please contact the Singapore
Branch of DNB Bank ASA at +65 6212 6144.
We, the DNB group, our associates, officers and/or employees may have interests in any products referred to in the Note by acting in
various roles including as distributor, holder of principal positions, adviser or lender. We, the DNB group, our associates, officers and/or
employees may receive fees, brokerage or commissions for acting in those capacities. In addition, we, the DNB group, our associates,
officers and/or employees may buy or sell products as principal or agent and may effect transactions which are not consistent with the
information set out in the Note.
Additional Information, including for Recipients in the United States:
The Note does not constitute an offer to sell or buy a security and does not include information, opinions, or
recommendations with respect to securities of an issuer or an analysis of a security or an issuer.

Wagging The Dog

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wagging The Dog

Uploaded by

Copyright:

Available Formats

Oil Market Outlook

Wagging the dog

August - 2017 - DNB Markets - Torbjrn Kjus 1

1 Executive summary .................................................................... 3

August - 2017 - DNB Markets - Torbjrn Kjus 2

August - 2017 DNB Markets - Torbjrn Kjus 3

US Horizontal Oil Rigs Weekly Change

our forecast of gradually higher oil prices 10

anticipation phase stopped about 10 $/b lower -20

chapter on the phases of price recovery).

The increase in the rig count has hence

August - 2017 DNB Markets - Torbjrn Kjus 4

The other issue was headwinds from the

of the key reasons why prices fell from Q1- 98

SUM: Non-Commercials Crude Futures -

This happened also in 2017 and the effect

August - 2017 DNB Markets - Torbjrn Kjus 5

Diff YOY OPEC Crude Production vs Exports

Source: Platts cFlow, IEA

Taken together we agree with Raymond

inventories have been drawing)

This was not what the market wanted to see 1,150

happened to any meaningful extent, but will

August - 2017 DNB Markets - Torbjrn Kjus 6

3. US gasoline demand is weakening Min / Max Range 5y Average 2016 2017

(DOE gasoline exports from Padd 3

August - 2017 DNB Markets - Torbjrn Kjus 7

3 Wagging the dog Volume in ICE Brent Contracts

possible to calculate the correct level of the oil 250

the cost to find and develop the marginal 150

barrel of oil (the long-dated price), plus or 100

minus backwardation or contango. The 50

backwardation or contango level should be 0

current market. So if oil stocks are building as

August - 2017 DNB Markets - Torbjrn Kjus 8

August - 2017 DNB Markets - Torbjrn Kjus 9

69.06 same Goldman Sachs report (which by the

47.87 way is a fantastic report that we always read

year). We do however as oil analysts have

August - 2017 DNB Markets - Torbjrn Kjus 10

60 costs. The theoretical oil price is hence lower

When we in 2012 predicted that US shale oil

August - 2017 DNB Markets - Torbjrn Kjus 11

4 Will the OPEC deal fall Nigerian Crude Oil Production

We have noticed that several analysts argue 2.1

communication from Ecuador that the country 1.1

of its fiscal requirements is of course

August - 2017 DNB Markets - Torbjrn Kjus 12

OPEC countries like Nigeria and Angola were

The platform is now on fire and when that

We hence believe Saudi Arabia and OPEC

How long time it takes to get down to the 5-

August - 2017 DNB Markets - Torbjrn Kjus 13

to be 2.0 million b/d stronger in 2H-2017 than 21.0

Million barrels per day

95 Min / Max Range 5y Average 2016 2017

Year on Year OECD Europe Oil Demand (IEA)

August - 2017 DNB Markets - Torbjrn Kjus 14

of 0.4 million b/d. 5%

YoY Implied YoY Implied -5%

The calculation of oil demand is then as

August - 2017 DNB Markets - Torbjrn Kjus 15

Chinese Diesel Demand Growth