CRUDE TANKER SECTOR Consensus upgrades on the horizon

Transcription

1 EQUITY RESEARCH CRUDE TANKER SECTOR Consensus upgrades on the horizon We maintain our positive view on the sector, and have raised our crude tanker forecasts on strong tanker fundamentals. We also expect a re-valuation of the crude tanker stocks from a broad and significant consensus upgrade as well as the ongoing shareholder rotation from momentum and growth investors to the larger income and long-only investors focusing on cash flow and yield. Our 216 VLCC forecast of USD55k/day is c3% above that implied by consensus. Tanker Shipping vs OSEBX (12m) jun aug okt des feb apr jun Tanker Shipping OSEBX (Rebased) Source: Factset We expect crude tanker tonne-mile demand to grow by 4%, up from -2% in 215e, with volume growth of 2%, up from 1% in 215e. We have a positive view on the sector, underpinned by changing trading patterns, with Asian tonne-mile absorbing the US shortfall, OPEC releasing its spare capacity, growth in and reallocation of refining capacity, together with our view that US crude exports will continue to accelerate. Forecasting average VLCC rates of USD62k/day in 215 and USD55k/day in 216. We have raised our 215e VLCC forecast to USD62k/day (USD43k/day) and 216e to USD55k/day (USD4k/day), while leaving 217e broadly unchanged at USD33k/day. For Suezmax, we have raised our 215e forecast to USD4k/day (USD31k/day), 216e to USD37k/day (USD3k/day), and again kept 217e broadly unchanged at USD26k/day. 2% upside to a 5-year-old VLCC on our new forecasts. On the back of our updated spot rate forecasts, we reiterate our positive view of prices of vessels on-water, with an average 17% upside for crude tankers. For VLCCs, we forecast 14% upside for resales, rising to 2% for 5-year-old vessels and 33% for 1-year-old ships. We forecast fleet growth of 1.9% for 215, 5.6% for 216 and 3.8% for 217. Based solely on the current order book, i.e. excluding our contracting estimate, 217e fleet growth is 2.4%. We assume scrapping of 1.1% in 215, 1.4% in 216 and 1.8% in 217. Potential trigger for the stocks from a broad, material consensus upgrade. We expect consensus to be raised once analysts start to look at their estimates ahead of Q2 and Q3 and the strong fundamentals entering 216e. In percentage terms, our revised estimates are some way (double-digit) above consensus for Q2, 215 and 216. Our 216 VLCC rate forecast of USD55k/day is c3% above the USD43k/day implied by consensus. Target prices raised. We reiterate our BUY recommendations on all the companies we cover in the sector bar Frontline, on which we have a HOLD recommendation given the potential merger with Frontline 212. We have raised our target prices on all the companies covered in the sector and our top picks are DHT and Euronav for their size, liquidity and attractive P/NAVs. Prospects of a tie-up with Frontline and the creation of a USD1.8bn market cap entity also make Frontline 212 a top pick. Company Cur Rec Target Price P/E 15e P/E 16e P/E 17e DHT Holdings USD BUY Euronav USD BUY Frontline NOK HOLD Frontline 212 NOK BUY Tanker Investments NOK BUY Teekay Tankers USD BUY Tsakos Energy Navigation USD BUY ANALYSTS Nicolay Dyvik nicolay.dyvik@dnb.no Øyvind Berle PhD oyvind.berle@dnb.no Petter Haugen petter.haugen@dnb.no Please see last pages for important information

2 CRUDE TANKER SECTOR Contents Tanker peer group 3 Estimate revisions 6 Crude tanker supply 8 Crude tanker demand 21 Crude tanker market balance 32 Crude tanker vessel price model 35 Company overviews DHT 42 Euronav 59 Frontline 72 Frontline Tanker Investments 17 Teekay Tankers 124 Tsakos Energy Navigation 141 2

3 % EBITDA (%) Tanker peer group Top picks DHT, Euronav and Frontline 212 Liquidity drives liquidity We reiterate our BUY recommendations on all the companies we cover in the tanker space bar Frontline, on which we have a HOLD recommendation ahead of a potential merger with Frontline 212. We have raised our target prices on all the companies in our coverage in the sector, and our top picks are DHT and Euronav based on their size and liquidity, and attractive price/nav. We also like Frontline for its pending tie-up with Frontline 212 a merger would create a USD1.8bn market cap company. Recommendations and changes to target prices Current Previous Company recommendation recommendation 216 EBITDA estimate revisions (%) Current target Previous target DHT BUY BUY USD12.5 USD1 Euronav BUY BUY USD2.4 USD16.8 Frontline HOLD HOLD NOK22.8 NOK 19 Frontline 212 BUY BUY NOK55 NOK 58 Tanker Investments BUY BUY NOK143 NOK118 Teekay Tankers BUY BUY USD9.1 USD7.6 Tsakos Energy Navigation BUY BUY USD12.2 USD1.9 6 % 57% 5 % 44% 4 % 3 % 2 % 15% 22% 22% 23% 23% 1 % % TNP TNK TIL FRNT FRO DHT EURN Change 216e EBITDA (%) Potential upside to target prices (%) 5 % 45 % 4 % 35 % 3 % 25 % 2 % 15 % 1 % 5 % % 48% 36% 38% 33% 23% 19% 5% FRO TNP TNK EURN FRNT TIL DHT Upside to Target 3

4 Yield (%) Yield (%) Price/NAV Price/one year forward NAV % 1-month share price performances (%) 3-month share price performances (%) 5 % 45 % 45% 35 % 3 % 3% 4 % 35 % 3 % 25 % 2 % 15 % 1 % 13% 13% 2% 23% 25% 36% 25 % 2 % 15 % 1 % 5 % % 2% 2% 3% 4% 9% 5 % % TIL FRO DHT EURN TNP TNK FRNT -5 % -1 % -6% FRO TNK TIL DHT TNP EURN FRNT 3 months change (%) 1 mos (%) Price/current NAV Price/one year forward NAV 1.6x 1.4x 1.2x 1.x x.9x.8x.7x.6x x.5x.6x.4x.2x.4x.3x.2x.1x.x TIL DHT EURN FRO FRNT TNK.x TIL TNP DHT EURN FRNT TNK Price/NAV Price/one year forward NAV Dividend yield based on 1% 216e payout Dividend yield based on 8% 216e EPS payout 25 % 23 % 2 % 19% 2 % 15 % 15 % 15 % 17 % 18 % 2 % 18 % 16 % 14 % 12 % 12% 13% 14% 14% 16% 1 % 1 % 8 % 6 % 5 % 4 % 2 % % DHT EURN TNP FRNT TNK TIL % DHT EURN TNP FRNT TNK TIL 216 full payout 216 dividend yield (%) 4

5 Fleet Multiples Leverage Sensitivities Valuation Misc. Crude tanker deviation table Frontline Frontline 212 Teekay Tankers Tsakos Energy Tanker Investments Euronav DHT Average Median Ticker FRO FRNT TNK TNP TIL EURN us DHT Currency used vs USD (Local) Recommodation HOLD BUY BUY BUY BUY BUY BUY Price (as of ) (Local) Target (Local) Deviation from target (%) EPS (USD/share) Performance (%) EBITDA (USDm) month month year NIBD (USDm) , Market cap (USDm) 447 1, , EV (USDm) 516 1,953 1,355 1, ,171 1, Current NAV (Local/share) NAV one year ahead (Local/share) NAV low case (Local/share) NAV high case (Local/share) Price/current NAV ratio Yield (%, base current price) ChgNAV per +1% values (Local/share) FALSE EPS sensitivity*** NIBD/EBITDA ratio Contract coverage (%) EV/EBITDA P/E VLCC Number Suexmax Number Aframax Number LR - product Number MR - product Number Handysize - product Number 8 8 Capesize Number Other Number Total Number

6 1.1% 1.4% 5 % 5 % 5 % 5 % 5 % 5 % 6 % 2 % 1.7% 2 % 1.8% 1.8% 7 % 2.% USD/day 22,124 22,124 25, 24, 23, 22, 3 % 3 % 2 % 3 % 3, 5 % 6 % 35, 6 % 7 % 21,922 21,922 USD/day 31,88 31,88 35, 33, USD/day 27, 26, 43, 4, 31, 3, 62, 55, 37, 4, Estimate revisions VLCC spot rate forecasts Suezmax spot rate forecasts 65, 45, 6, 55, 4, 5, 45, 35, 4, 3, 35, 3, 25, 25, 2, 2, 15, e 216e 217e 15, e 216e 217e VLCC (USD/day), old VLCC (USD/day), new Suezmax (USD/day), old Suezmax (USD/day), new Aframax spot rate forecasts Crude tanker deliveries 4, 8 % 35, 7 % 6 % 3, 5 % 4 % 25, 3 % 2, 2 % 1 % 15, e 216e 217e % e 216e 217e Aframax (USD/day), old Aframax (USD/day), new Deliveries, old Deliveries, new Crude tanker scrapping forecasts Crude tanker new orders forecasts 2.5% 1 % 9 % 2.% 8 % 7 % 1.5% 6 % 5 % 1.% 4 % 3 %.5% 2 % 1 %.% e 216e 217e % e 216e 217e Scrapping, old Scrapping, new New ordering, old New ordering, new 6

7 85 % 86 % 72 % 71 % 88 % 88 % 75 % 74 % 74 % 75 % 77 % 91 % 93 % 93 % 94 % 82 % YOY increase in transported volumes.7%.6%.9% 1.% 1.2% 2.1% 2.% YOY increase in ton-mile demand -1.3%.8% 1.3% 1.9% 2.1% 1.9% 2.6% 4.4% 3.9% -2.9 % -2.9 % -2.3 %.5 %.6 %.7 % 1.9 % 2.9 % 5% 5% 5% 1.7 % 5% 2.9 % 5% 6% 6% 3.8 % 8.1 % 8.2 % 4.6 % 5.6 % 12.1 % Fleet growth forecasts Speed-adjusted supply growth forecasts 6 % 17 1% % 5 % 9% 128% % 4 % 7% 76% % 3 % 5% 2 % 24% % 3% 1 % -32%% 1% % e 216e 217e -8%% 214e 215e 216e 217e Conventional supply growth, old Conventional supply growth, new Speed-adj supply New ordering, growth, old New Speed-adj ordering, supply newgrowth, new Volumes transported (incl. pipeline) Tonne-mile demand forecasts 3.% 5.% 2.5% 2.% 4.% 3.% 2.% 1.5% 1.% 1.%.% -1.%.5% -2.%.% e 216e 217e -3.% e 216e 217e Volume growth, old Volume growth, new Demand growth, old Demand growth, new Utilisation of the crude tanker fleet (speed-adjusted) forecasts Conventional (not speed-adjusted) utilisation of fleet forecasts 96 % 84 % 94 % 82 % 92 % 9 % 88 % 86 % 84 % 82 % 8 % 8 % 78 % 76 % 74 % 72 % 7 % 68 % 66 % 78 % e 216e 217e 64 % e 216e 217e Utilisation, old Utilisation, new Conventional utilisation, old Conventional utilisation, new 7

8 e 216e 217e Fleet growth (%) Deliveries (m dwt) Crude tanker supply Increasing fleet growth 215e fleet growth of 1.9%, 216e at 5.6% and 217e at 3.8% We forecast fleet growth of 1.9% for 215, 5.6% for 216 and 3.8% for 217. Looking exclusively at the existing order book without our contracting forecast, 217e fleet growth is 2.4%. We forecast 1.1%, 1.4% and 1.8% annual scrapping for We expect the crude tanker fleet to increase from 343m dwt at the beginning of 215 to 383m dwt at end-217. The current fleet comprises 58% VLCCs, 21% Suezmax, 2% Aframax and 2% Panamax tankers. Positive crude tanker supply factors: Limited supply growth next 12 months Floating and forced storage absorbing capacity Less savings from increasing the speed the additional knot keeping fleet growth low Preference for having vessels on the water in a strong rate environment leading to fewer new orders Lower oil price environment reducing savings from eco ships Less interest from capital markets to support newbuild orders The 1996-built double-hull fleet will be 2 years old in 216 Negative crude tanker supply factors: Continued pick up in new orders leading to oversupply Ample yard capacity for 217 deliveries at top 1 yards Release of floating storage back in the crude tanker fleet Risk of a continued increase in vessel speed Low scrapping given strong rates and modern fleet with the single-hull phasing out Crude tanker fleet growth (%) Deliveries 8.% 7.% 6.% 5.% 4.% 3.% 2.% 6.8% 6.4% 6.3% 4.8% 4.1% 4.2% 4.1% 3.1% 1.6% 1.4% 1.9% 5.6% 3.8% %.6%.% Expected deliveries from new ordering (m dwt) Expected deliveries from current OB (m dwt) Historical deliveries (m dwt) 8

9 Jan-3 Oct-3 Jul-4 Apr-5 Jan-6 Oct-6 Jul-7 Apr-8 Jan-9 Oct-9 Jul-1 Apr-11 Jan-12 Oct-12 Jul-13 Apr-14 Jan-15 Oct-15 Jul-16 Apr-17 Jan-3 Sep-3 May-4 Jan-5 Sep-5 May-6 Jan-7 Sep-7 May-8 Jan-9 Sep-9 May-1 Jan-11 Sep-11 May-12 Jan-13 Sep-13 May-14 Jan-15 Sep-15 May-16 Jan-17 Sep-17 Monthly fleet growth Annual fleet growth YTD15 ROY YTD15 ROY m dwt m dwt e 216e 217e m dwt Cumulative share of fleet (%) m dwt Crude fleet by year built 14 1 % 1 % 85 % 12 9 % 8 % 1 7 % 8 48 % 6 % 5 % 6 4 % 4 2 % 3 % 2 % 2 % % % % 4 % 1 % % Panamax Aframax Suezmax VLCC Cumulative share of fleet Fleet profile (year-end) Delivery year VLCC Suezmax Aframax Panamax Source: Clarksons, Source: Clarksons, Crude tanker fleet and order book by vessel size Crude tanker fleet and order book by shipyard country Panamax Aframax Suezmax VLCC Other South Korea Japan China 1 1 Source: Clarksons, Fleet by year of delivery Gross orderbook Source: Clarksons, Fleet by year of delivery Gross orderbook Monthly fleet growth 2.% 1.5% 1.%.5%.% -.5% -1.% -1.5% -2.% Annualised, smoothed monthly growth versus YOY growth 12% 1% 8% 6% 4% 2% % -2% -4% -6% MOM 3mths MA YOY 3mths MA annualised Source: Clarksons, Source: Clarksons, 9

10 Jan-7 Aug-7 Mar-8 Oct-8 May-9 Dec-9 Jul-1 Feb-11 Sep-11 Apr-12 Nov-12 Jun-13 Jan-14 Aug-14 Mar-15 Oct-15 May-16 Dec-16 Jul-17 Jan-7 Aug-7 Mar-8 Oct-8 May-9 Dec-9 Jul-1 Feb-11 Sep-11 Apr-12 Nov-12 Jun-13 Jan-14 Aug-14 Mar-15 Oct-15 May-16 Dec-16 Jul-17 Fleet growth (YOY, %) Fleet growth (YOY, %) Jan-7 Aug-7 Mar-8 Oct-8 May-9 Dec-9 Jul-1 Feb-11 Sep-11 Apr-12 Nov-12 Jun-13 Jan-14 Aug-14 Mar-15 Oct-15 May-16 Dec-16 Jul-17 Jan-7 Aug-7 Mar-8 Oct-8 May-9 Dec-9 Jul-1 Feb-11 Sep-11 Apr-12 Nov-12 Jun-13 Jan-14 Aug-14 Mar-15 Oct-15 May-16 Dec-16 Jul-17 Fleet growth (YOY, %) Fleet growth (YOY, %) Jan-7 Aug-7 Mar-8 Oct-8 May-9 Dec-9 Jul-1 Feb-11 Sep-11 Apr-12 Nov-12 Jun-13 Jan-14 Aug-14 Mar-15 Oct-15 May-16 Dec-16 Jul YTD15 ROY Fleet growth (YOY, %) Share of total fleet and order book (%) Monthly fleet growth YOY by vessel size Crude tanker fleet and order book by shipyard country 15% 1% 5% % -5% -1% 9 % 8 % 7 % 6 % 5 % 4 % 3 % 2 % 1 % % -15% Fleet by year of delivery Gross orderbook VLCC Suezmax Aframax Panamax Source: Clarksons, China share Japan share South Korea share Source: Clarksons, Monthly fleet growth VLCC 12% 1% 8% 6% 4% 2% % -2% Monthly fleet growth Suezmax 12% 1% 8% 6% 4% 2% % -2% VLCC Source: Clarksons, Monthly fleet growth Aframax 8% Suezmax Source: Clarksons, Monthly fleet growth Panamax 1% 6% 5% 4% 2% % % -5% -2% -1% -4% -15% Aframax Panamax Source: Clarksons, Source: Clarksons, Floating storage set to tie-up 6.8% of fleet capacity We calculate that floating storage will tie-up 6.8% of fleet capacity in 215, up from 4.9% in 214, based on a combination of floating and forced storage. With forced storage we mean additional time spent to discharge as the cargo needs to find a home when the market is 1

11 VLCC speed (knot) TCE (USD/day) 61,12 64,43 69,815 7,987 7,875 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15 Apr-15 May-15 Jun-15 Floating storage (m barrels) Storage as a share of fleet (%) flooded with oil and port congestion is an issue. Around 176m barrels are tied up in floating storage, an increase of 71m barrels since Q4 and corresponds to 36 VLCC equivalents. We expect floating storage to decline into 216 to 6.% of fleet capacity, and then to 5.% by 217, releasing capacity into the tanker market. Floating storage (m barrels) Floating storage as a % of fleet capacity % 7.% 6.% 5.% 4.% 3.% 2.% 1.% 6.8% 6.% 4.9% 5.% 4.7% 4.6% 4.% 3.8% 3.3%.6%.7%.% m barrels Average ballast speed has increased by one knot We calculate that the average ballast speed has increased by a knot since Q3 214 to about 13 knots, which corresponds with the Euronav and Frontline 212 fleet speeds. We believe this been absorbed by the market. We calculate that an increase in ballast speed from 1 knots to 13 knots would raise the timecharter equivalent rate by cusd6k/day, while an increase from 13 knots to 15 knots would only increase the timecharter rate by cusd1k/day. Hence, owners do not have the same incentive to increase the speed any further as the flip side would be additional supply coming into the market by faster speeds, ratcheting up supply growth. We highlight that all else being equal, raising speeds should bolster owner profits in a high rate (cusd1k/day) but lower bunker price environment. VLCC speeds, laden and ballast (knots) 14.5 Faster speed increases timecharter earnings 85, 14 8, , 7, 65, , 11 55, 5, VLCC Speed laden VLCC Speed ballast Ballast speed (knot) 11

12 m dwt m dwt Ballast speed (knot) Ballast speed (knot) Euronav ballast speed (knots) Frontline 212 ballast speed (knots) EURN FRNT Crude tanker order book At the beginning of June, the gross order book was 49m dwt, or 14% of the fleet at the start of 215. We estimate 3.m dwt (or 6.5%) of the order book is not delivered (see chart below to the right) and have adjusted our delivery assumptions accordingly. VLCCs account for 65% of the order book by capacity, Suezmaxes 22%, Aframaxes 12% and Panamaxes 1%. The order book/fleet ratio is 17% for VLCCs, 17% for Suezmaxes, 9% for Aframaxes and 9% for Panamaxes. The crude tanker order book is split between China (33%), Japan (9%) and South Korea (49%), while other building countries account for the remaining 9%. Comparing this with deliveries in , we note that China s market share has declined from 37% and Japan s from 12%, while South Korea s share has increased from 48%. Crude tanker order book at start of year, by delivery year Our expected order book adjusted for future delivery To be delivered current year Y+2 and onwards next year (Y+1) 43 Order book reported by Clarkson...of which delivery earlier than first forecast month Estimated cancellations of current future order book DNB adjusted orderbook for future delivery Source: Clarksons, Source: Clarksons, 12

13 m dwt m dwt m dwt m dwt m dwt m dwt Crude tanker order book by vessel size 3 Crude tanker order book by builder ROY ROY VLCC Suezmax Aframax Panamax China Japan South Korea Other Source: Clarksons, Source: Clarksons, Total order book, delivery, and contracting year Total order book, contracting, and delivery year ROY Delivery year YTD < < YTD15 Ordering year ROY15 Source: Clarksons, Source: Clarksons, Total order book, build country, and contracting year Total order book, contracting year, and shipyard country China Japan South Korea Others Builder location YTD < < YTD15 Ordering year Others South Korea Japan China Source: Clarksons, Source: Clarksons, 13

14 m dwt m dwt m dwt m dwt m dwt m dwt VLCC order book, delivery, and contracting year VLCC, contracting, and delivery year ROY Delivery year YTD < < YTD15 Ordering year ROY15 Source: Clarksons, Source: Clarksons, Suezmax order book, delivery, and contracting year Suezmax order book, contracting, and delivery year YTD < ROY15. ROY Delivery year. < YTD15 Ordering year Source: Clarksons, Source: Clarksons, Aframax order book, delivery, and contracting year 3. Aframax order book, contracting, and delivery year YTD < ROY15. ROY Delivery year. < YTD15 Ordering year Source: Clarksons, Source: Clarksons, 14

15 Jan-3 Sep-3 May-4 Jan-5 Sep-5 May-6 Jan-7 Sep-7 May-8 Jan-9 Sep-9 May-1 Jan-11 Sep-11 May-12 Jan-13 Sep-13 May-14 YOY change in MA(3) monthly deliveries e 216e 217e Monthly deliveries (m dwt) Deliveries (m dwt) m dwt m dwt Panamax order book, delivery, and contracting year.25 Panamax order book, contracting, and delivery year.35.2 YTD < ROY15. ROY Delivery year. < YTD15 Ordering year Source: Clarksons, Source: Clarksons, Deliveries 214 deliveries equated to 2.8% of the fleet and we forecast 2.9% for 215, 7.% for 216 and 5.6% in 217, including contracting. Excluding our additional contracting forecast, we expect 4.2% deliveries in 217. As the crude tanker market is predominantly composed of fewer larger vessels than say, dry bulk, volatility in deliveries becomes significant from month to month, as illustrated in the chart below showing 3-month average delivery growth. The time from ordering to expected delivery based on new orders in 215 YTD is 1.9 years (chart to the left below). As shown in the chart below to the right, deliveries in 215 would on average have spent 4.1 years in the order book, an all-time high (note this is before cancellations). We expect this to fall to 2.6 years in 216, and to 2.5 years in 217. Deliveries by month Deliveries by year % 15% 1% 5% % -5% -1% VLCC Suezmax Aframax Panamax YOY (3mth MA) Source: Clarksons, Source: Clarksons, Expected deliveries from new ordering (m dwt) Expected deliveries from current OB (m dwt) Historical deliveries (m dwt) 15

16 m dwt Share not delivered (%) Expected cancellations in current orderbook (m dwt) YTD Years from ordering to delivery Years from ordering to delivery Time from ordering to delivery by ordering year, all sizes Time from ordering to delivery by delivery year, all sizes Year of ordering Year of delivery Source: Clarksons, Source: Clarksons, Cancellations At the beginning of June, the gross order book was 49m dwt, or 14% of the fleet at the statt of 215. We assume 3.m dwt (6.5%) of the order book is not delivered, which we primarily believe will come in 215 and from the VLCC and Aframax vessel segments, with Suezmax close behind. Share of order book not delivered (%) Expected future cancellations % 5 % 4 % 3 % % 1 % % To be delivered current year Actual delivered e 216e 217e Share not delivered (%) >> VLCC Suezmax Aframax Panamax Source: Clarksons, Source: Clarksons, Contracting 18.3m dwt was contracted in 214, or 5.4% of the fleet as it stood at the start of that year. We expect 3.9m dwt to be contracted in 215 (based on annualised contracting YTD), reflecting 7.3% of the fleet at the start of 215, with 5.7% for 216 and 4.6% for 217. Our new ordering model is based on ordering and earnings data from the previous 12 years, excluding the current year. On a qualitative level, the reasoning is simple as it assumes that higher TC earnings lead to increased new ordering we use a simple regression model to quantify this relationship (see charts below). Using the aforementioned historical data and this methodology, we find a fairly good fit for our model with an R-squared of 42%. 16

17 Actual contracting as % of fleet Contracting (% of fleet as start of year) YTD15 2 % 3 % 5 % 4 % 5 % 6 % 5 % 9 % 9 % 8 % 1 % 14 % 15 % 22 % May-11 Jul-11 Sep-11 Nov-11 Jan-12 Mar-12 May-12 Jul-12 Sep-12 Nov-12 Jan-13 Mar-13 May-13 Jul-13 Sep-13 Nov-13 Jan-14 Mar-14 May-14 Jul-14 Sep-14 Nov-14 Jan-15 Mar-15 May Monthly contracting (m dwt) YTD15 Oct-1 Aug-2 Jun-3 Apr-4 Feb-5 Dec-5 Oct-6 Aug-7 Jun-8 Apr-9 Feb-1 Dec-1 Oct-11 Aug-12 Jun-13 Apr-14 Feb-15 New contracting (m dwt) VLCC (USD/day) Contracting (m dwt/month) VLCC spot rates (USD/day) Annual contracting Monthly contracting and VLCC earnings 7 12, , , 8, 6, 4, 2, Panamax Aframax Suezmax VLCC VLCC (USD/day) , 15, 1, 5, Source: Clarksons, (Annualised) contracting by year Source: Clarksons, << Crude tanker contracting VLCC (USD/day) >> Past four years of contracting by vessel size VLCC Suezmax Aframax Panamax Source: Clarksons, Our crude tanker contracting model Historical contracting as a percentage of the fleet 25 % 25 % 2 % 15 % y = 1E-6x R² = % 15 % 1 % 1 % YTD15 5 % 5 % % % 2, 4, 6, 8, 1, 12, VLCC (USD/day) Actual contracting as % of fleet Model contracting as % of fleet Source: Clarksons, Yard overview Aggregated, the top 1 yards have 79% of the total order book. YTD, Hyundai has received most orders (2.6m dwt), followed by Daewoo (2.2m dwt) and Dalian (1.2m dwt). 17

18 YTD15 Scrap value VLCC (USD m) VLCC 1 year TC rate (USD/day) Scrapping as % of fleet Also, from this overview it seems apparent that all major yards still have ample available capacity for 217 deliveries, but we now model for the 216 order book to be full. Top ten shipyards ranked by share of current order book Order book by year of delivery (m dwt) Order book by year of ordering (m dwt) Yard group Country Total ROY < YTD15 Hyundai Samho HI South Korea Daewoo South Korea Dalian Shipbuilding China P.R Shanghai Waigaoqiao China P.R Hyundai HI (Gunsan) South Korea Jinhai Heavy Ind China P.R Hyundai HI South Korea JMU Ariake Shipyard Japan New Times SB China P.R Daewoo-Mangalia Romania Total top Other yards Top 1 share of total 79 % 84 % 76 % 84 % 54 % 35 % 49 % 92 % 82 % 82 % Source: Clarksons, Scrapping Crude tankers scrapping has been fairly modest in recent years, ranging between 1.8% and 2.5% annually since 21; we forecast 1.1% for 215, 1.4% in 216 and 1.8% in 217. The average scrapping age of crude tankers in recent years rose again for all segments in 214 and into 215, except for Panamax vessels, which have seen a small decline in % of the fleet will be due for a fourth special survey (2 years) in 216e, 1.3% in 217e and 2.4% in 218e. We believe scrapping 15-year-old vessels will be limited as we are optimistic on vessel values and do not believe all charterers will be equally strict on age as we expect the market to become fairly firm although we do believe some badly maintained 15- year-old vessels will be scrapped. Our scrapping model is based on data from the previous eight years. The main point is that higher TC earnings reduce scrapping, while higher scrap prices incentivise scrapping. We use scrap price divided by earnings (12-month TC rates for VLCCs) as the explanatory variable. Using the aforementioned historical data and this methodology, we find a fit for our model with an R-squared of 39%. Scrap prices (USDm) and 1-year TC rates, both VLCCs 25 7, Our tanker scrapping model 2.5 % 2 6, 5, 2. % , 3, 2, 1, 1.5 % 1. %.5 % YTD15 y = 2E-5x +.54 R² =.393 << Scrap price VLCC (USD m) VLCC 1 year TC (USD/day) >>. % Scrap price / earnings Source: Clarksons, (assumes lightweight of 17k tonne for VLGCs) Source: Clarksons, 18

19 e 216e Average age when scrapped Scrapping (m dwt) < < m dwt vessels scrapped # vessels scrapped YTD15 May-11 Aug-11 Nov-11 Feb-12 May-12 Aug-12 Nov-12 Feb-13 May-13 Aug-13 Nov-13 Feb-14 May-14 Aug-14 Nov-14 Feb-15 May Scrapping (m dwt) Scrapping (% of fleet) Montly scrapping (m dwt) Historical scrapping, by size and total as percentage of fleet 25 1% 9% 2 8% 7% 15 6% 5% 1 4% 3% 5 2% 1% % Past four years of scrapping by vessel size VLCC Aframax Total scrapping (% fleet) >> Source: Clarksons, Suezmax Panamax Source: Clarksons, VLCC Suezmax Aframax Panamax Past three years of scrapping, by capacity Past three years of scrapping, by number of vessels 12 1 % 8 1 % % 8 % 7 % 6 % 5 % 4 % 3 % 2 % 1 % Panamax Aframax Suezmax VLCC % 8 % 7 % 6 % 5 % 4 % 3 % 2 % 1 % Panamax Aframax Suezmax VLCC % Cumulative share - % Cumulative share Age when scrapped (years) Age when scrapped (years) Source: Clarksons, Source: Clarksons, Average age of scrapped vessels Scrapping % -1 % -2 % -3 % -4 % -5 % -6 % -7 % -8 % VLCC Suezmax Aframax Panamax Historical scrapping (m dwt) Expected scrapping (m dwt) % of fleet start of year >> Source: Clarksons, Source: Clarksons, 19

20 % of fleet Renewal survey schedule for the current fleet 12.% 1.% 8.% 6.% 6.6% 5.4% 4.% 2.%.% 3.4% 5.5% 2.4% 3.4% 1.9% 1.3%.1%.2%.6%.5% 216e 217e 218e 219e % of fleet doing the 3th renewal survey % of fleet doing the 4th renewal survey % of fleet doing the 5th renewal survey Source: Clarksons, 2

21 Volume growth (%) Tonne mile (%) Crude tanker demand Tonne-mile set to outperform volume growth 216e and 217e tonne-mile growth forecast of 4% We forecast e crude tanker tonne-mile to grow by 4%, up from -2% in 215e, with volume growth of 2%, up from 1% in 215e. Fundamentally, stronger refinery production growth and a reallocation of refinery capacity fuels increased crude trade and tonne-mile demand. We expect Chinese tonne-mile demand to grow by 6% over , India by 5% and other Asia by 7%, outgrowing the negative tonne-mile growth of 12% we expect to the US in the same period. With exports, already materialising, we expect the reduction in US imports from the Middle East to slow compared with 215. Additional OPEC volumes, in particular from the Middle East, more or less directly translate into increased demand for seagoing transportation; i.e. a continuation of its volumemaximisation strategy should provide support for crude tanker rates. Five reasons why we are positive on crude tanker demand: Changing trading patterns with Asian tonne-mile absorbing the US shortfall OPEC releasing spare capacity as a strategic response to non-opec supply Likely higher oil demand growth from lower prices Growth in refining capacity at 1.3mbpd, up from 1.mbpd and reallocation of refining capacity US exports of crude leading to inefficiencies and increased tonne-mile demand Four risk factors to follow closely in terms of crude tanker demand: Lower OPEC production Reduced Chinese and Indian inventory building and demand US sourcing less crude oil from the Middle East, reducing tonne-mile demand US imports from Canada and Keystone pipeline Crude tanker volume growth (%) Crude tanker tonne-mile growth 5 % 4 % 3 % 2 % 1 % % -1 % -2 % -3 % -4 % 4 % 2 % 2 % 3 % -1 % -4 % 1 % 1 % % % 2 % 1 % 2 % 1 % 5 % 4 % 3 % 2 % 1 % % -1 % -2 % -3 % -4 % -5 % -6 % 4 % 4 % 4 % 4 % 2 % 2 % 2 % 2 % 1 % % -1 % -1 % -1 % -5 % Changing trade patterns Asia should offset negative tonne-mile demand from US We expect Chinese tonne-mile demand to grow by 6% over , India by 5% and other Asia by 7%, outgrowing the negative tonne-mile growth of 12% we expect to the US in the same period. 21

22 kbpd Laden tonne-miles (bn) Our Asian tonne-mile growth forecast is based on increased sourcing from the Atlantic Basin, with Latin American and West African exports following lower US imports from the same regions. The negative tonne-mile growth in the US has been particularly evident in 215 with the cut back in imports from the Middle East; however, based on our assumptions of US exports of crude oil (which are already happening), we expect stable Middle East imports (see separate section on US crude exports). We forecast annual Chinese imports of 4kbpd from e, up from 285kbpd from , Other Asia-Pacific of 33kbpd, up from 39kmpd during the past three years, and we expect US imports to decline by 2kbpd, from 495kbpd. Change in crude tanker laden tonne-mile demand by region (bn) 4, 3, 2, 1, -1, -2, -3, e 216e 217e China US India Europe Other Asia Pacific Others Change in crude tanker volume growth by region (kbpd) 1,5 1, , -1, e 216e 217e China US India Europe Other Asia Pacific Others 22

23 Chinese import demand, YOY US import demand, YOY Annual average additions (kbpd) Annual average additions (kbpd) Annual average additions (kbpd) Annual average additions (kbpd) Historical average annual change by importer Forecast average annual change e by importer Source: BP, Bloomberg, Source: BP, Bloomberg, Historical average annual change by exporter Forecast average annual change e by exporter Source: BP, Bloomberg, Source: BP, Bloomberg, Chinese import demand growth in tonne-mile and volume (%) US import demand growth in tonne-mile and volume (%) 1 % 9 % 8 % 7 % 6 % 5 % 4 % 3 % 6 % 9 % 7 % 4 % 3 % 9 % 8 % 5 % 3 % 6 % 5 % 8 % 7 % % -5 % -1 % -15 % -3 % -5 % -5-5 % % -5 % -9 % -14 % -14 % -1 % % -7 % -5 % -8 % 2 % 1 % 1 % -2 % -2 % % e 216e 217e -25 % e 216e 217e Tonne-mile Volumes Tonne-mile Volumes 23

24 Oct-9 Dec-91 Feb-93 Apr-94 Jun-95 Aug-96 Oct-97 Dec-98 Feb- Apr-1 Jun-2 Aug-3 Oct-4 Dec-5 Feb-7 Apr-8 Jun-9 Aug-1 Oct-11 Dec-12 Feb-14 Oct-9 Dec-91 Feb-93 Apr-94 Jun-95 Aug-96 Oct-97 Dec-98 Feb- Apr-1 Jun-2 Aug-3 Oct-4 Dec-5 Feb-7 Apr-8 Jun-9 Aug-1 Oct-11 Dec-12 Feb-14 mbpd mbpd European import demand, YOY Global import demand, YOY Indian import demand, YOY Other Asia Pacific import demand, YOY Indian import demand growth in tonne-mile and volume (%) Other Asia import demand growth in tonne-mile and volume (%) 25 % 2 % 15 % 1 % 5 % % -5 % 21 % 14 % 14 % 8 % 7 % 7 % 7 % 7 % 7 % 4 % 4 % 3 % 1 % -2 % e 216e 217e 16 % 14 % 12 % 1 % 8 % 6 % 4 % 2 % % -2 % -4 % -6 % 14 % 1 % 7 % 5 % 4 % 5 % 3 % 3 % 2 % 1 % % -3 % 11 % 6 % e 216e 217e Tonne-mile Volumes Tonne-mile Volumes European import demand growth in tonne-mile and volume (%) Global import demand growth in tonne-mile and volume (%) 14 % 12 % 1 % 8 % 6 % 4 % 2 % % -2 % -4 % -6 % 11 % 5 % 1 % 2 % 4 % 2 % 1 % 1 % % % -1 % -1 %-1 % -4 % e 216e 217e 5 % 4 % 3 % 2 % 1 % % -1 % -2 % 4 % 4 % 2 % 2 % 2 % 2 % 2 % 1 % 1 % 1 % 1 % % -1 % -2 % e 216e 217e Tonne-mile Volumes Tonne-mile Volumes US crude imports from Nigeria (mbpd) US crude imports from Angola (mbpd) , Bloomberg, Bloomberg 24

25 Jan- Oct- Jul-1 Apr-2 Jan-3 Oct-3 Jul-4 Apr-5 Jan-6 Oct-6 Jul-7 Apr-8 Jan-9 Oct-9 Jul-1 Apr-11 Jan-12 Oct-12 Jul-13 Apr-14 Jan-15 Jan-4 Aug-4 Mar-5 Oct-5 May-6 Dec-6 Jul-7 Feb-8 Sep-8 Apr-9 Nov-9 Jun-1 Jan-11 Aug-11 Mar-12 Oct-12 May-13 Dec-13 Jul-14 Feb-15 mbpd mbpd Oct-9 Dec-91 Feb-93 Apr-94 Jun-95 Aug-96 Oct-97 Dec-98 Feb- Apr-1 Jun-2 Aug-3 Oct-4 Dec-5 Feb-7 Apr-8 Jun-9 Aug-1 Oct-11 Dec-12 Feb-14 mbpd mbpd Oct-9 Dec-91 Feb-93 Apr-94 Jun-95 Aug-96 Oct-97 Dec-98 Feb- Apr-1 Jun-2 Aug-3 Oct-4 Dec-5 Feb-7 Apr-8 Jun-9 Aug-1 Oct-11 Dec-12 Feb-14 Oct-9 Dec-91 Feb-93 Apr-94 Jun-95 Aug-96 Oct-97 Dec-98 Feb- Apr-1 Jun-2 Aug-3 Oct-4 Dec-5 Feb-7 Apr-8 Jun-9 Aug-1 Oct-11 Dec-12 Feb-14 mbpd mbpd US crude imports from Venezuela (mbpd) US crude imports from Canada through pipeline (mbpd) , Bloomberg, Bloomberg US crude imports from Saudi Arabia (mbpd) US crude imports from Persian Gulf (mbpd) mbpd, Bloomberg, Bloomberg China crude oil imports (mbpd) China crude imports Saudi Arabia , Bloomberg, Bloomberg 25

26 May-14 Jun-14 Jul-14 Aug-14 Sep-14 Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15 Apr-15 May-15 Jan-5 Aug-5 Mar-6 Oct-6 May-7 Dec-7 Jul-8 Feb-9 Sep-9 Apr-1 Nov-1 Jun-11 Jan-12 Aug-12 Mar-13 Oct-13 May-14 Dec-14 OPEC production (m bpd) VLCC rates (USD/day) m bpd Jun-8 Oct-8 Feb-9 Jun-9 Oct-9 Feb-1 Jun-1 Oct-1 Feb-11 Jun-11 Oct-11 Feb-12 Jun-12 Oct-12 Feb-13 Jun-13 Oct-13 Feb-14 Jun-14 Oct-14 Feb-15 Jan-7 Jun-7 Nov-7 Apr-8 Sep-8 Feb-9 Jul-9 Dec-9 May-1 Oct-1 Mar-11 Aug-11 Jan-12 Jun-12 Nov-12 Apr-13 Sep-13 Feb-14 Jul-14 Dec-14 mbpd mbpd China crude imports Colombia (mbpd) China crude imports Brazil (mbpd) , Bloomberg, Bloomberg OPEC releasing spare capacity Saudia Arabian crude oil production hits all-time high in May As illustrated in the chart below, OPEC increased production by 1.63m bpd in May 215 versus May 214 (from 29.9), of which Saudi Arabia accounted for about 35% (.58m bpd of the 1.63m bpd increase). Saudia Arabian crude oil production reached 1.3mbpd in May, a 6% rise YOY. Although VLCC rates do not correlate closely with production from month to month, we believe it is at the core of the strong crude tanker market. In particular, the increase in Saudi crude exports, now (March) at their highest level (7.9m bpd, which is 1.2m bpd higher than August 214 and.6m bpd higher than November 214 when OPEC held its last meeting) since November 25, is likely to have had a positive influence on VLCC rates. Additional OPEC volumes, in particular from the Middle East, more or less directly translate into increased demand for seaborne transportation; i.e. a continuation of its volume-maximisation strategy should provide support for crude tanker rates, which have shown strength in 215, averaging about USD6k/day YTD. OPEC total oil production vs. VLCC rates (past 12 months) Saudi Arabia crude exports and implied export share , 7, 6, 5, 4, 3, 2, 1, % 95% 9% 85% 8% 75% 7% 65% 6% 55% 5% OPEC production VLCC rates Saudi crude exports Export share of production Source: Poten&Partners, Bloomberg Source: JODI, Bloomberg 26

27 Dec-2 Jul-3 Feb-4 Sep-4 Apr-5 Nov-5 Jun-6 Jan-7 Aug-7 Mar-8 Oct-8 May-9 Dec-9 Jul-1 Feb-11 Sep-11 Apr-12 Nov-12 Jun-13 Jan-14 Aug-14 Mar-15 Dec-2 Aug-3 Apr-4 Dec-4 Aug-5 Apr-6 Dec-6 Aug-7 Apr-8 Dec-8 Aug-9 Apr-1 Dec-1 Aug-11 Apr-12 Dec-12 Aug-13 Apr-14 Dec-14 m bpd Saudi oil production (m bpd) Oct-9 Jan-92 Apr-93 Jul-94 Oct-95 Jan-97 Apr-98 Jul-99 Oct- Jan-2 Apr-3 Jul-4 Oct-5 Jan-7 Apr-8 Jul-9 Oct-1 Jan-12 Apr-13 Jul-14 mbpd mbpd OPEC crude oil production, Saudi Arabia (mbpd) OPEC crude oil production, Saudi Arabia (mbpd) , Bloomberg, Bloomberg OPEC is incentivised to release spare capacity As we argued in earlier research, we find the economic rationale for OPEC to increase, not reduce, production compelling. In 214 non-opec countries increased supply by 2.17m bpd, of which the US accounted for about 1.4m bpd. As US unconventional production seems to be both: 1) price-sensitive around the current oil price; and 2) capable of rapidly adjusting output, we view it as highly likely that it has replaced OPEC as the marginal supplier in the global oil market. While this should not be news, forecasters and analysts appear to be overlooking the possibility that OPEC could deploy its spare capacity at some stage. Given the low cash-cost of OPEC crude production and the loss of price control to US shale, OPEC holds a dominant strategy in terms of maximising output, which should limit further investments in unconventional production and hence stunt future growth prospects, which inevitably would eat into OPEC s market share if OPEC were to reduce production to support prices. Also, demand elasticity should mitigate the price decline, given time. OPEC and Saudi Arabia s spare capacity Saudi Arabia s estimated production capacity and actual production implying current spare capacity of 2.25m bpd OPEC spare capacity Saudi spare capacity Saudi oil production Saudi production capacity Source: Bloomberg Source: Bloomberg 27

28 Annual oil demand growth (m bpd) OPEC spare capacity by member (m bpd) OPEC spare capacity by member (m bpd) Source: Bloomberg Estimates for OPEC spare capacity have significant uncertainty attached to them. However, our point is not whether it is currently 4m bpd or 6m bpd (the latter being Bloomberg s current estimate), but whether it will be deployed or not. A 1m bpd increase in OPEC crude exports would equate to close to a 3% increase in global traded crude volumes; i.e. 4m bpd is about 1 11%. Such an increase would likely have a tonne-mile multiplier of about 1.3, as the average distance travelled by Middle East exports is c3% longer than the same figure for the total market. This is probably not possible in today s tanker market simply because there is not enough transport capacity to accommodate it (unless all vessels travel at full-speed). Also, the simple fact that OPEC has increased production by 1.2m bpd since the November meeting could be interpreted as support for spare capacity already being released. Likely higher oil demand from lower prices IEA oil demand forecasts on the conservative side First, we use the IEA s estimates for oil demand growth, which were recently reduced despite the lower oil price. Hence, we believe this has more upside than downside potential. In the reports published in 213 and 214, the IEA estimated oil demand growth of m bpd for 216; this was lowered to 1.2m bpd in its 215 report. IEA demand growth estimates report 214-report 215-report Source: DNB Market 28

29 Annual change in refinery capacity (kbpd) e 216e 217e 218e Refinery capacity additions (k bpd) Reallocation of refining capacity Growth in refining capacity at 1.3mbpd, up from 1.mbpd Despite the IEA lowering its estimates in its latest report published in February, it still expects refinery additions to increase annually by an average of 1.25m bpd in (average growth was 1.m bpd in ). The growth in refining capacity for 215 is to a large extent represented by US and Middle East and the fact that Middle East would like to process more crude for product exports, means that China has to reply more of its imports from the Atlantic basin which is positive for tonne-mile demand. Historical and expected refinery capacity additions IEA still expects refinery additions to increase annually by an average of 1.25m bpd in Source: BP, IEA As before, the additions are skewed compared to the currently installed capacity: China and the Middle East, which today have about 23% of global refining capacity, are accountable for 55% of the growth expected in Expected refinery capacity additions by region 2, 1, , , ,289 Africa Middle East Latin America Other Asia China Non-OECD Europe FSU OECD Pacific China and the Middle East are accountable for 55% of the growth expected in OECD Europe Source: IEA -1, OECD North America The IEA s recent estimate reductions for the future refinery additions were the largest for China, while the expectations for the Middle East expansions have been raised in recent reports. This should be good news for the product tanker market, as expansions in the Middle East are mostly used to increase exports, but much less so for China, which mainly intends to cover its own demand for petroleum products. 29

30 Change in global oil production (m bpd) Change in global oil production (k bpd) US share of production growth 14% 42% 48% 62% 22% Added CDU capacity in the period (k bpd) Estimated expansions for by year of publication Africa Middle East Latin America Other Asia China Publishing year of IEA report Non-OECD Europe FSU OECD Pacific OECD Europe OECD North America Source: IEA US crude exports Growth in US shale revolution is a true revolution; from 21 to 214 US growth (3.8m bpd) was twice as large as the rest of the world (1.8m bpd). US crude production has not led to a similar decline in imports. We believe our US crude export, first included in our estimates in February 214, case is still relevant, mainly because production is still expected to increase while US refineries do not seem capable of processing much more light tight oil in the nearterm future. Change in global oil production (mbpd) Change in global oil production (mbpd) US share of production growth (%) % 2% 15% 1% 5% Change other countries Change other countries % Change US production Change US production US share of production growth 3

31 Jan-1983 Aug-1984 Mar-1986 Oct-1987 May-1989 Dec-199 Jul-1992 Feb-1994 Sep-1995 Apr-1997 Nov-1998 Jun-2 Jan-22 Aug-23 Mar-25 Oct-26 May-28 Dec-29 Jul-211 Feb-213 Sep-214 U.S. Percent Utilization of Refinery Operable Capacity k bpd Jan-27 Jun-27 Nov-27 Apr-28 Sep-28 Feb-29 Jul-29 Dec-29 May-21 Oct-21 Mar-211 Aug-211 Jan-212 Jun-212 Nov-212 Apr-213 Sep-213 Feb-214 Jul-214 Dec-214 Total oil (incl NGL and non-shale) production, EIA STEO-report Jan 215, mbpd Jan-13 Apr-13 Jul-13 Oct-13 Jan-14 Apr-14 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16 WTI price US crude export (k bpd) Oct-9 Dec-91 Feb-93 Apr-94 Jun-95 Aug-96 Oct-97 Dec-98 Feb- Apr-1 Jun-2 Aug-3 Oct-4 Dec-5 Feb-7 Apr-8 Jun-9 Aug-1 Oct-11 Dec-12 Feb-14 US oil production (k bpd) mbpd 11,297 11,33 US crude production (kbpd) US crude imports (kbpd) 12, , 1. 1, 8. 9, 8, 7, , EIA oil production and WTI spot price forecast (short-term energy outlook, Jan 215); still expect production increase and US crude export is already happening, Nov 214 showed an all-time-high at 52k bpd EIA, Jan 215 Historical spot price (dollars per barrel) STEO price forecast (dollars per barrel) US refinery utilisation starting out 215 at high level Domestic LTO production has replaced imports 9, 8, 7, 6, 5, 4, 3, 2, 1, 7. jan feb mar apr may jun jul aug sep oct nov dec y avg Heavy Light 31

32 Crude tanker market balance Our crude tanker model Variables exogenous to the model are shown in purple in the figure below, while endogenous variables or assumptions we made are shown in dark green. For future periods we use our own trade forecasts as discussed above. Our crude tanker model Trade between each of the following regions US & Canada Latin America Africa Europe FSU Volumes crossed with a distance matrix and adjusted for ballast and deadfreight estimated from fixture data Crude tanker transport demand Middle East China Asia Pacific Steel prices Scrapping Crude tanker fleet Spot rates New orders Deliveries Current order book Cancellations Crude tanker transport capacity Utilisation of fleet Bunker prices Vessel speed Storage Port operations On the supply side our modelling starts and ends with spot rates: based on historical rates and new ordering, we use a simple regression model to forecast new ordering. Future scrapping is modelled as a function (parameterised using another simple regression) based on the ratio of steel prices and rates (intuition is that higher rates give less scrapping, while, since steel prices are typically heavily correlated with scrap prices, higher steel prices should result in increased scrapping hence the ratio of scrap prices to earnings correlates positively with scrapping). Deliveries are estimated using the current order book taken from Clarksons, adjusted for our assumed cancellations; we assume that all vessels ordered before 28 that are still in the order book will be cancelled, and also all orders at Rongsheng. Deliveries from our estimated new ordering are added with a 28-month time lag from the date of ordering. To arrive at an estimate of transport capacity, we model the optimal vessel speed for given bunker prices (which are derived from our oil analyst s oil price forecast) and estimated spot rates (modelled with a 3-month time lag). Also note that we base the time used for normal port operations on the vessel speed, as a slower speed at sea implies that all vessels use a smaller share of their time in port than when sailing at normal/design speed at sea. 32

33 VLCC rates (USD/day) VLCC rates (USD/day) Our notation on transport capacity is then tonne-mile/month, which we also adjust for storage and the market discrimination of single-hull tonnage. We index demand equivalent to a fleet utilisation of 95% in 23. The historical demand data is based on data provided by BP on a region-by-region basis by year along with various sources for Japan, China, the US and Singapore. We also use historical fixture data to adjust for dead freight and ballast. We have now opted not to adjust for a changing ballast pattern as this makes a better fit between historical rates and fleet utilisation. In the charts below we show the historical relationship between spot rates and fleet utilisation, which makes the basis for our forecast spot rates. That said, we acknowledge that this is only one factor in the spot rate formation, the other being what we call the fundamental arbitrage, which determines the ability and willingness to pay for crude oil transportation services when fleet utilisation is sufficiently high. However, in the current crude oil market transportation costs are an insignificant part of the total value of goods delivered. Hence, higher transportation costs are not significantly demand destructive (as they may be in, for instance, the product tanker and dry bulk markets). In effect, the potential upside to spot rates for large crude carriers is extremely high as proven by history with VLCC rates occasionally peaking above USD3,/day. Historical relationship between spot rates and conventional (not speed-adjusted) fleet utilisation 12, Historical relationship between spot rates and actual (speed-adjusted) fleet utilisation 12, 1, 1, 8, 6, R² = , 6, R² =.177 4, 4, 2, 2, 7 % 75 % 8 % 85 % 9 % 95 % 1 % Conventional (not speed-adjusted) fleet utilisation (%) 86 % 88 % 9 % 92 % 94 % 96 % Acutal (speed-adjusted) fleet utilisation (%) Source: Various, Source: Various, 33

34 Our crude tanker supply, demand, balance and rate forecasts, base case SUPPLY e 216e 217e Fleet, start of year (m dwt) Gross OB with delivery current year as start of year Not delivered* Historical deliveries (m dwt) % not delivered of historical OB for current year* -32 % -43 % -45 % -54 % -25 % -29 % 6 % 22 % 18 % 25 % 48 % 35 % 13 % Order book Assumed not delivered** Deliveries from current order book (m dwt) % not delivered of remaining OB 23 % 3 % 1 % New orders placed, historical New orders placed, future Total ordering (m dwt) Ordering as % of fleet start of year 15.2% 1.1% 6.6% 23.6% 9.9% 16.7% 4.5% 11.3% 2.% 2.9% 5.% 5.4% 7.3% 5.7% 4.6% Delivery of new orders placed, future Total deliveries (m dwt) Deliveries as % of fleet start of year 1.3% 8.1% 8.8% 5.8% 6.8% 7.1% 1.9% 9.3% 1.3% 7.9% 4.5% 2.8% 2.9% 7.% 5.6% Historical scrapping (m dwt) Forecasted scrapping (m dwt) Total scrapping (m dwt) Scrapping as % of fleet start of year 7.2% 2.6% 1.1%.4%.4%.8% 1.9% 2.5% 2.3% 2.5% 2.4% 1.8% 1.1% 1.4% 1.8% Misc./Net additions Fleet, end of year (m dwt) YoY growth 4.1% 6.4% 4.2% 4.1% 1.6% 6.3% 3.1% 6.8% 4.8% 1.4%.6% 1.9% 5.6% 3.8% Fleet average normal speed, laden (knot) Fleet potential transport capacity (trn ton-mile/year) Time used for port operations (%) 27 % 27 % 27 % 27 % 27 % 27 % 27 % 26 % 24 % 23 % 22 % 22 % 26 % 25 % 23 % Port operations (trn ton-mile/year) Storage, share of fleet (%).6%.6%.6%.6%.6%.7% 3.3% 4.7% 4.% 3.8% 4.6% 4.9% 6.8% 6.% 5.% Storage (trn ton-mile/year) Single hull share of fleet (%) 53 % 45 % 36 % 28 % 22 % 18 % 14 % 8 % 2 % % % % % % % Single hull discrimination as share of single hull fleet (%) 4 % 61 % 7 % 92 % 1 % 1 % 1 % 1 % 1 % 1 % Single hull discrimination (trn ton-mile/year) Bunker price (USD/ton) Reported speed reduction (%)*** -3 % -7 % -15 % -2 % -23 % -25 % Optimal speed in open sea, laden (knot) Optimal speed in open sea, ballast (knot) Optimal speed, blended by size and L/B (knot) Speed reduction from normal to optimal speed (%) % 2 % -2 % -2 % -9 % -4 % -17 % -19 % -33 % -32 % -31 % -22 % -2 % -6 % -15 % Share of market able to utilize optimal speed, historical (%) 42 % 21 % 34 % 44 % 6 % 71 % 93 % 1 % Share of market able to utilize optimal speed, future (%) 95 % 1 % 1 % Share of market able to utilize optimal speed, combined (%) 42 % 19 % 34 % 44 % 6 % 71 % 93 % 89 % 1 % 1 % Economic speed (trn ton-mile) Capacity taken out in slower speed % -2 % -5 % -11 % -14 % -16 % -18 % -14 % -14 % -13 % Net transport capacity (trn ton-mile/year) YoY growth 3 % 5 % 5 % 5 % -8 % -3 % 3 % 7 % 4 % -1 % -3 % -2 % 8 % 12 % DEMAND e 216e 217e Transported volume (kbpd) 35,418 36,797 37,423 38,186 39,157 38,948 37,553 37,59 37,866 38,323 38,145 38,93 39,331 4,372 4,875 YoY growth 4 % 2 % 2 % 3 % -1 % -4 % % 1 % 1 % % 2 % 1 % 3 % 1 % Transport demand, barrel-mile per year (1^9) 53,85 55,133 55,361 56,526 57,176 56,379 53,639 55,57 56,863 58,93 57,691 58,736 57,811 6,223 62,365 YoY growth 4 % % 2 % 1 % -1 % -5 % 4 % 2 % 2 % -1 % 2 % -2 % 4 % 4 % Transport demand, adjusted for deadfreight 6,496 62,835 63,622 64,939 66,46 65,782 63,186 65,661 67,572 68,986 68,651 69,9 68,798 71,669 74,218 YoY growth 4 % 1 % 2 % 2 % -1 % -4 % 4 % 3 % 2 % % 2 % -2 % 4 % 4 % Transport demand, adjusted for deadfreight and ballast 95,626 98,811 1,115 12,1 14,151 13,14 99,4 12,867 15,695 18,8 17,882 19,965 18, , ,752 YoY growth 3 % 1 % 2 % 2 % -1 % -4 % 4 % 3 % 2 % % 2 % -1 % 4 % 4 % Indexed demand (trn ton-mile/year, 95% in 23) YoY growth 3 % 1 % 2 % 2 % -1 % -4 % 4 % 3 % 2 % % 1.9% -1.3% 4.4% 3.9% Utilisation (adjusted for speed) (%) 95 % 95 % 92 % 89 % 87 % 94 % 93 % 94 % 9 % 88 % 89 % 93 % 94 % 91 % 85 % Conventional utilisation (not speed adjusted) (%)**** 95 % 95 % 92 % 89 % 87 % 94 % 88 % 88 % 79 % 75 % 74 % 75 % 74 % 72 % 71 % Spot rates (USD/day) VLCC 53,75 93,69 56,878 58,721 56,743 96,184 25,659 35,695 13,941 18,153 17,9 31,88 62, 55, 33, Suezmax 34,855 62,945 46,398 47,374 35,647 59,97 22,4 23,493 1,51 13,59 13,663 21,922 4, 37, 26, Aframax 29,297 36,33 34,49 32,532 27,614 39,152 9,67 14,962 5,61 1,333 14,965 22,124 35, 3, 22, Panamax***** 38,53 1,445 12,285 8,57 11,629 11,15 18,717 28, 27, 21, 34

35 Crude tanker vessel price model On the back of our updated spot rate forecast and views of a strong market next 18 months, we reiterate our positive view for asset prices of vessels on water with average 17% upside for crude tankers and 2% for product tankers, with more upside to older and larger vessels. However, we do not forecast any upside to newbuilds prices due to the highly competitive yard situation with lack of order backlog, falling steel prices and limited investor interest for funding new orders. For VLCCs, we forecast 14% upside for resales, increasing to 2% for 5-year-old vessels and 33% for 1-year-old ships. For Suez- and Aframax, we forecast 11 13% for resales, 8 11% for 5-year-old vessels and 2 23% for 1-year-old ships. For LR2s, we see 6 7% upside for resales and 5-year-old vessels, and 15% upside for those in the 1-year-old bracket, with less upside for LR1s and MRs. We believe the crude tanker market is still firming up. With increased confidence, we expect spot rates to drive second-hand prices. Our methodology is essentially based on a regression analysis with the differential between spot rates and cash breakeven rates as the explanatory variable. We have made adjustments from the model output on the back of the value of fuel efficiency. asset value forecasts: base case (USDm) Current Newbuild Resale 5 year old 1 year old VLCC Suezmax Aframax LR LR MR Handymax Base case Newbuild Resale 5 year old 1 year old VLCC Suezmax Aframax LR LR MR Handymax Upside (%) Newbuild Resale 5 year old 1 year old VLCC % 14 % 2 % 33 % Suezmax % 11 % 8 % 2 % Aframax % 13 % 11 % 23 % LR2 % 7 % 6 % 15 % LR1 % % -8 % -8 % MR % -3 % 4 % % Handymax -2 % 5 % 15 % -14 % asset value forecasts: low/high case (USDm) High case Newbuild Resale 5 year old 1 year old VLCC Suezmax Aframax LR LR MR Handymax Low case Newbuild Resale 5 year old 1 year old VLCC Suezmax Aframax LR LR MR Handymax High case Newbuild Resale 5 year old 1 year old VLCC 2 % 3 % 45 % 67 % Suezmax 3 % 23 % 27 % 48 % Aframax 4 % 23 % 33 % 52 % LR2 4 % 14 % 15 % 27 % LR1 4 % 2 % 36 % 58 % MR 5 % 21 % 42 % 56 % Handymax 16 % 24 % 42 % 1 % Low case Newbuild Resale 5 year old 1 year old VLCC -9 % -1 % -23 % -27 % Suezmax -8 % % -2 % -25 % Aframax -9 % % -24 % -29 % LR2-7 % -7 % -13 % -12 % LR1-11 % -12 % -8 % -8 % MR -11 % -11 % -8 % -11 % Handymax -13 % -8 % -8 % -37 % Model The figure below lays out our views on asset values in shipping. 35

36 How we view asset price formation in shipping Scraping the bottom. In February 213, when the market was still scraping the bottom we published a model based on yard margins. This phase is found in the lower left quadrant in the figure above. At that time yard margins were clearly negative (we estimated by c15%) and with improved sentiment among buyers, we concluded that higher prices were needed to accommodate more orders: yards would not sell additional vessels for less than or the same as previously. From a methodological viewpoint, we built an historical regression model aiming to estimate yard (gross) margins and calculated ~16% potential upside to newbuild prices on a 12-month horizon and translated this (after adjusting for age, depreciation and value of fuel efficiency) to 1 2% upside potential to second-hand prices. Overbuilding. First, we take a step back: to enter a market characterised by this race to the bottom caused by fierce yard competition, the market needs to go through a phase we call overbuilding (lower right quadrant) would fall into this category given its low average rates but high deliveries, implying healthy yard order backlogs. On the way down, this phase is characterised by newbuild prices lagging behind second-hand prices, and at times newbuild prices are quoted far higher than second-hand (resale) prices. Also, liquidity in both the newbuild- and second-hand markets is tightening during this part of the cycle. Sky s the limit. To find the reason (apart from the, in retrospect, very unnecessary mini rally in 21) for the combination of record-low rates and record-high deliveries in , we need to go back to when the sky was the limit (upper right quadrant) and new ordering skyrocketed and the fleet-to-order book ratio peaked at 48%. And much of what was ordered in had record-late delivery, i.e. 3 4 years ahead, as shown below. Also, in the peak year of 26, yards recorded 4.1x as much order intake as delivered tonnage (both measured by dwt). Recovery. We now believe the market is in what we have called the recovery phase. In practice, this would imply that dayrates (we use 1-year TC rates as a proxy) are sufficient to pay both the crew and the bank; i.e. rates are again moving above cash break-even, but there are still available slots at the yards with normal delivery dates. The average leadtime to delivery is about 28 months. To find a good proxy for a vessel s cash flow, we use historical resale prices back to 28. In the cash break-even calculation we assume maturity (tenor) of 15 years, USD1,/day of G&A and USD7,/day for opex. To arrive at a resale price forecast, we use the difference between 1-year TC rates and resale cash break-even as a proxy for cash flow, which we then regress towards resale prices in This is shown in the chart below for VLCCs. There are two main reasons for not using a longer data series: 1) simply because we do not have data on resale prices before April 28 (however, we could use a proxy derived from 5-year-old vessels); and 2) vessel 36