Commodities Compendium 190117 RX e 263687

GLOBAL
Commodities Compendium
Politics, profits and protectionism
The world is positioning for a healthy period of global reflation in 2017, adding
investor support to a healthy demand perspective for commodities as the
industrial recovery matures. Meanwhile, a lack of new supply projects is seeing
the raw material end of value chains tighten in many cases. This is a very
supportive environment for commodities, and given cost structure moves have
severely lagged price gains, many commodity producers are in a sweet spot at
the present time.
So far, so good. However, such periods dont last forever, making duration of
this cycle critical. Commodity prices are cyclical after all, and we are looking at
an industrial economy which is still underperforming the global economy as a
whole, which itself is struggling to accelerate. Visibility in China remains poor
post the 2017 political transition, particularly as many much-needed reforms
have been delayed. In practical terms, we are still worried about the very same
things as we were this time last year; however, while Chinese industrial profits
are growing such concerns become less pressing. More recently, the growing
threat of protectionist policy is casting a shadow over commodity trade flows. In
our opinion, this will lead to two-tier commodity pricing in many markets,
benefitting domestically focused producers over those competing into
international markets.
Cash flow, caution and (a lack of) capex are widespread

Analyst(s) Across all commodity markets there are some common themes at present.
Macquarie Capital (Europe) Limited Producers, who this time last year were fearing for their future, are now seeing
Colin Hamilton
+44 20 3037 4061 colin.hamilton@macquarie.com strong free cash flow generation. And, while some of this will be put back
Matthew Turner towards sustaining capital and operational improvements at existing assets,
+44 20 3037 4340 matt.turner@macquarie.com
Vivienne Lloyd
growth capex is not yet widespread as producers simply lack confidence the
+44 20 3037 4530 vivienne.lloyd@macquarie.com recent gains are sustainable. This is understandable, given the heightened
Patrick Morton political risk environment over 2017 and beyond, with few tangible commodity
+44 20 3037 2014 patrick.morton@macquarie.com
Jim Lennon, Senior Commodities Consultant stories with above-trend demand forecast.
+44 20 3037 4271 jim.lennon@macquarie.com
Macquarie Capital Securities (Singapore) Pte. The focus shifts from demand surprise to supply recovery
Limited
Ian Roper The Q4 commodity price moves set a high base as we enter 2017. As a result,
+65 66010698 ian.roper@macquarie.com even this early in the year there is a high degree of certainty that annual average
Conrad Werner
+65 6601 0182 conrad.werner@macquarie.com prices will be higher in most cases this year. However, this is already starting to
Macquarie Capital Limited attract latent supply capacity back to production across many commodities,
Lynn Zhao
+86 21 2412 9035 lynn.zhao@macquarie.com whether that be oil, iron ore or aluminium, given we are trading out the operating
Macquarie Capital Securities (Malaysia) Sdn. Bhd. cost curve for the majority. This is standard commodity economics, but does
Isaac Chow mean that necessary inventory draws and wider market rebalancing are delayed.
+60 3 2059 8982 isaac.chow6@macquarie.com
Macquarie Securities (NZ) Limited We are cautious on the outlook for all three commodities mentioned above for
Nick Mar these reasons.
+64 9 363 1476 nick.mar@macquarie.com
Macquarie Capital (USA) Inc. Meanwhile, for industrial commodities we remain concerned that Chinese
Vikas Dwivedi
+1 713 275 6352 vikas.dwivedi@macquarie.com construction activity, still a significant demand driver, will be weakening into mid-
Walt Chancellor year, while sustained raw material constraints will still be few and far between.
+1 713 275 6230 walt.chancellor@macquarie.com As a result, from todays levels we see bulk commodities lower at 2017-end than
Jay Tobin
+1 713 275 6123 jay.tobin@macquarie.com current price levels, but oil, gold and agriculture higher. Among base metals we
Julia Surgeon, CFA continue to like exposure to zinc, however the positive nickel story has been
+1 713 275 6461 julia.surgeon@macquarie.com
completely derailed by the Indonesian ore ban reversal. We also see current
19 January 2017 prices for uranium as being too low to be sustainable, while we expect recent
gains in chrome, cobalt, alumina and US natural gas to be sustained through Q2.
Please refer to page 92 for important disclosures and analyst certification, or on our website
www.macquarie.com/research/disclosures.
Macquarie Research Commodities Compendium
Our key commodity recommendations

Firstly, we can consider where many commodities are in the fundamental cycle, as per Fig 1.
The arrows show where we see these commodities moving to on a two-year view. We
continue to see a general shift to the right, but still few are in the right-hand-most green bar
marking raw material constraint. The Q4 acceleration has seen met coal, manganese and
alumina move quickly through the cycle, while iron ore is another we expect to pass back
towards the red supply cuts zone over the coming year.
Fig 1 Positions of key mined/extracted commodities in the fundamental cycle arrow shows two-year progression
Strong Demand Push Demand destruction Market back in balance Strong supply grow th High stocks Stocks Draw ing Supply constraint
or deficit market Capex accelerating Capex peaking Stocks building Strong supply reaction Limited new supply Capex lagging
Price Acceleration Price peaking Prices falling Severe price decline Prices stablise at low level Prices stable Pricing to encourage supply
Diamonds
Palladium Bauxite Oil
Lithium Nitrogen
Gold
LNG Chrome
Met Coal Aluminium Lead
US Natural Gas
Manganese Copper
Tin Iron Ore
Thermal Coal Nickel Zinc

Potash
Steel
Alumina Uranium Platinum Silver
Source: Macquarie Research, January 2017
Outperformers on a 12- to 24-month basis from current levels

US Natural Gas Strong underlying demand will extend period of tightness.
Zinc The raw material constraint story becomes ever clearer.
Silver Not hurt by reflation, helped by inflation, and lacking mine supply.
Cobalt A consistent deficit expected, and major country risk remains.
Uranium Prices are finally invoking a supply response.
Crude Palm Oil Bolstered by low inventory and a strong vegoil complex.
Potential for medium-term raw material constraints

Chrome Too much dependence on underperforming South African supply.
Copper A lack of capex will bite, leaving significant pressure on large existing mines.
Zinc Its hard to solve for the lack of ex-China mine supply growth.
Cobalt Should EV battery technology move in its favour, it would be hard to satiate.
Commodities for which the longer-term challenges are underappreciated

Potash The market share battle is maturing, but overcapacity persists.
Alumina Perhaps the only commodity with significant ex-China capacity still coming.
Aluminium Industry cost deflation is an ongoing headwind.

Steel Chinese supply cuts will help, but a downward demand trend is the main issue.
Nitrogen - Increased efficiency of use and displacement by alternatives present problems.
19 January 2017 2
Table of contents
Executive Summaries
Macro Outlook: Creeping forwards, but at least on all cylinders 5
China Outlook: Demand has surprised to the upside, but a slowing is inevitable 14
Metals & Bulks: Passing through the sweet spot 16
Energy: Co-ordination, compliance and caution 24
Agricultural: Waiting for a weather kick-start 26

Dairy: Prices buoyed by supply squeeze 28
Commodity Summaries
Copper
A corner turned 36
Aluminium
Stubbornly stable pricing 38
Zinc
Higher, faster shorter? 40
Lead
Never a dull moment 42
Tin
2-handle returns 44
Nickel
Indonesian ore export relaxation dashes strong price recovery 46
Stainless Steel
Booming 2016 Chinese production driven by exports and restocking 48
Ferrochrome
In desperate need of a supply response 50
Molybdenum
Big supply cuts rebalance market but medium-term oversupply risks remain 52
Cobalt
Risks remain, but looking ripe for 2017 upside 54
Lithium
The focus shifts to Australian supply 56
Bauxite & Alumina
A big alumina deficit, but abundant bauxite 58
Steel
Starting the drive down a long road to sustainable improvement 60
Iron ore
How long can prices hold up amid abundant supply? 62
Manganese
What goes up.. 64
Metallurgical Coal
Prices come tumbling down, but where will they find a floor? 66
Gold
Medium-term outlook strong, but near-term concerns 68
Silver
Reflation OK, inflation good 70
Platinum and Palladium
Slowing car market, more recycling main headwinds 71
19 January 2017 3
Diamonds
After the rally, theres the after-rally 74
Global Oil
From Binaries to Balancing in 2017 and 2018 76
US Natural Gas
Tightness likely to persist across summer 2017 78
Thermal Coal
Will 276 days return or will prices settle at an acceptable level? 80
Uranium
Theres low, and theres too low 82
Crude Palm Oil
Tight CPO inventory, vegetable oil complex should support prices 84
Milk Powder
Powder prices buoyed by supply contraction 86
Nitrogen
Riding the energy price uplift, but structural concerns remain 88
Potash
The market share battle is maturing 90
19 January 2017 4
Macro Outlook
Creeping forwards, but at least on all cylinders
Global growth accelerated around mid-year 2016, with all key countries/sectors improving,
but especially the developed markets. At year-end this recovery seemed to be strengthening,
raising the prospect of a good 2017. But the new US Administration offers both upside and
downside risks, and Chinas outlook is hazy.
Growth in the world economy has accelerated since its nadir in Q4 2016/Q1 2017. We
calculate from our database of 50 leading economies that GDP, the most comprehensive
albeit lagging indicator, rose by 0.83% QoQ in 3Q, the fastest pace since 3Q 2014 (fig 2). On
a YoY basis this meant GDP was 2.9% higher, the best pace in a year (fig 3).
Fig 2 World GDP, % change QoQ (PPP weights) Fig 3 World GDP, % change YoY (PPP weights)
1.2% 5.0%
4.5%
1.0%
4.0%
0.8% Using YoY
3.5% data
0.6%
3.0%
0.4% 2.5% Using QoQ
data
0.2% 2.0%
1.5%
0.0%
2011 2012 2013 2014 2015 2016
2011 2012 2013 2014 2015 2016
Source: National output data, Macquarie Research, January 2017 Source: National output data, Macquarie Research, January 2017
It would be a stretch to call the recovery overwhelming. Of those 50 countries in our

database, only 21 saw faster growth in 3Q than 2Q. And just over half saw faster growth in
3Q 16 than 3Q 15. But importantly the largest economy, the USA, saw its growth improve
from 0.4% in 3Q to 0.9% in 4Q, while Russia returned to growth and India and Mexico
accelerated. Turkey was the main disappointment, contracting by 1.8% QoQ in 3Q.
While China (fig 4) continues to supply the lions share of global growth, the contribution from
developed economies is behind the uptick. Crucially the recent improvement in US growth
has not come at the expense of Europe/Japan.
Fig 4 World GDP, % change QoQ, by area Fig 5 GDP by country, QoQ, %
1.0%
0.8%
0.6%
0.4%
0.2%
0.0%
USA Other developed
(0.2%) Other emerging China
2011 2012 2013 2014 2015 2016
Source: National output data, Macquarie Research, January 2017 Source: National data, Macquarie Research, January 2017
19 January 2017 5
The recovery is even more evident if we look at industrial production, a series of great
importance to commodity markets and for which we have data up to November.
Using our database of 69 leading industrial countries (which account for 94% of global IP), we
calculate industrial output was 2.8% higher YoY in November, the best performance since
December 2014 and a major swing from flirting with a global industrial recession at the end of
2015. On a 3m moving average basis, which gives a better guide to the trend given data
volatility, growth was more modest but still 2% higher, the best in 18 months (both fig 6).
When assessing the commodity demand implications, however, we need to be cautious. As
fig 7 shows manufacturing output, the largest component of industrial production and the
most obviously commodity-using, is still growing faster than the wider measure, but has not
had the same swings and the gap between the two series has narrowed considerably. This
implies, as we forecast in the last Compendium, an improvement in mining/oil or utility output
has been an important driver of the improvement in wider IP.
Fig 6 Global IP recovers, spikes higher in November Fig 7 Manufacturing continues to outpace wider IP
(% change YoY, 3m moving average) but gap narrowing, % change YoY (3m moving average)
5% 5%
4% 4%
3% 3%
2% 2%
1% 1%
Manufacturing, 3m average IP, 3m average

0% 0%
2012 2013 2014 2015 2016 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16
Source: National data, Macquarie Research, January 2017 Source: National data, Macquarie Research, January 2017
This shift is most obvious in the IP of the US, where mining output remains lower YoY but is
off its depths (fig 6). In China the catch-up of IP has been more to do with more robust utility
(mainly electricity) output (fig 7), which might benefit energy commodities.
Fig 8 US IP growth by type, % YoY Fig 9 China IP growth by type, % YoY
19 January 2017 6
Looking ahead, PMI surveys, we believe, give a short look into the future. Decembers reports
(fig 10) suggest the YoY growth rate in IP/manufacturing will continue to improve into year-
end. We have modestly upgraded our 2016 global estimate (1.8%, from 1.7%) and forecast
for 2017 (now 2.7%, from 2.5%) - fig 11, though there is much uncertainty, especially since
the election of Donald Trump.
Fig 10 Manufacturing PMIs vs IP/manufacturing

output growth, % change YoY Fig 11 World industrial production, % change YoY
56 5%
55 5%
4%
54
4%
53 Manufacturing
3%
52 3%
51 2%
PMI 2%
50
1%
49 Industrial 1%
Production
48 0%
2013 2014 2015 2016
Source: National data, Markit, Macquarie Research, January 2017 Source: National data, Macquarie Research, January 2017
Financial market and policy developments
Trumps election was behind the main shifts in financial markets over the last few months,
although some of the trends were already brewing before then. After an initial shock, traders
reckoned that a Trump Administration would lead to faster US economic growth on the back
of some kind of fiscal stimulus, whether tax cuts or through higher infrastructure spending.
This in turn led to a renewed increase in US 10 year bond yields (fig 12), and indeed
across the curve. Higher US yields also helped drag yields up in Germany, the UK and
Japan, though to a much lesser extent given ongoing central bank action.
Bond yields rise for three reasons - higher real yields, either because of anticipation of higher
interest rates or (less likely in the US case), anticipation of greater issuance, or anticipation of
higher inflation. Inflation-linked bonds suggest the increase has been driven by both higher
inflation and higher real yields, though the latter has fallen back recently.
Fig 12 10yr bond yields, % Fig 13 US 10 year yield, decomposition, %
Source: Macrobond, Macquarie Research, January 2017 Source: Macrobond, Macquarie Research, January 2017
19 January 2017 7
The prospect of higher US interest rates was given a boost by the Feds decision to raise
rates in December for the second time since the financial crisis, and even more by the
policymakers raising of their number of expected rate hikes in 2017 to three, from two.
Market pricing followed suit.
Fig 14 Fed fund futures curve, % Fig 15 US inflation measures, % change YoY
Higher inflation expectations have also played a role. The US headline inflation rate has
risen sharply since 2015 as oil prices have gone from being 50% down YoY to nearly 50%
higher. Other measures, which exclude volatile items, have shown much smaller swings,
though in general have trended higher (fig 15). How the Fed reacts to rising inflation if the US
economy steps up a gear will be one of the key future developments, with uncertainty caused
by President Trumps pre-election comments.
Outside the US, other countries are also seeing rising inflation on the back of higher oil
prices. But except for in special cases that effect only has a little longer to run, and core
rates of inflation have been more stable, with low readings maintained in the Eurozone and
Japan.
Fig 16 Headline inflation rising, % change YoY Fig 17 Core inflation more stable, % change YoY
19 January 2017 8
This difference in core rates has also meant a difference in monetary policy. While the Fed
has been shifting to a tighter stance, the Bank of Japan and ECB have gone the other way.
The BOJ announced in September a policy of targeting 10yr yields around 0%, and the ECB
waited until December before extending its QE policy for another six months (at a slightly
weaker pace of 60bn/year).
A widening interest rate differential, a much higher prospect of fiscal stimulus and lower slack
typically means a rising currency. And indeed the dollars rally, which had stalled significantly
in 2016, came roaring back towards year-end, taking it to new post-2002 highs across a
broad range of indices (fig 18) by early 2017, before falling back. It also rallied against the
Chinese yuan, hitting a peak 6.96 at the start of 2017 before PBoC intervention and a
generally weaker tone saw it fall to 6.85. Against the PBOCs CFETs basket, to which Korea
was added at the start of this year, it has been more stable (fig 19).
Fig 18 Dollar gains post US election to new highs Fig 19 Yuan falling against the dollar, but not its
(1995 = 100) basket (end-2014 = 100)
110
105
100
95
90
85
Dec-14 Jun-15 Dec-15 Jun-16 Dec-16
The rising dollar was, as ever, bad for the gold price (fig 20 - dollar is inversed). But it did not
have a similar effect on base metal prices, as shown by the LMEX in fig 21. We saw three
reasons for this in US dollar stronger but less scary for (most) metal prices:
The value of the dollar is the only factor that drives metal prices, and if S&D
fundamentals warrant a large move, then it can overwhelm FX effects.
The impact of the dollar depends on why it is moving. The post-election trade was a
risk-on move, and this benefits base metals.
Similarly the yuans decline was seen as helping the Chinese economy, not as in
2015, a consequence of a collapsing Chinese economy.
19 January 2017 9
Fig 20 Gold price, $/oz cannot escape DXY surge Fig 21 but base metals (LMEX) can, and how!
Nevertheless we cannot get sanguine about the dollar. Just because it hasnt yet derailed the
commodity rally, this does not mean it cannot. One concern, though one that is hard to
quantify, is the possibility of a border adjustment tax in the US, part of the Congressional
Republicans tax plan. Such a tax is thought by economists not to be protectionist, despite
reducing taxes on firms who export and penalising those who import, because it would be
offset by a rise in the US dollar. There are many hurdles still to pass through before
something like this becomes law, however, and President-elect Trump suggested he was
sceptical of it in an interview shortly before his inauguration, pushing the dollar notably lower
than its peak.
Economic Outlook
So is this time different? The impressive end to the year for the world economy, as shown
by PMI surveys and industrial production data, has raised hopes that finally the world
economy is shaking off the sluggishness seen since the GFC. But by mid-January there were
some signs the Trump reflation trade was going cold - the dollar and yields had both fallen
back as investors focused on the less economically positive potential consequences of the
new Administration.
In its latest Global Economic Outlook, our economics team continue to forecast the long
grinding cycle, though they now see real GDP growth (on a market exchange rate basis) at
the top-end of the recent range at 2.6% in 2016, 3.0% in 2017 and 2.9% in 2018 (from a
September forecast of 2.2%, 2.6% and 2.7%). Using PPP weights the rates are roughly 0.4%
higher.
19 January 2017 10
Fig 22 Global real GDP growth; Macquaries long grinding cycle forecast, YoY
6%
5% Forecasts after
here
4%
3%
2%
1%
0%
(1%) Other developed
(2%) USA
Other developing
(3%)
China
(4%)
2007 2009 2011 2013 2015 2017 2019
Note: The Total-16 is the IMFs 10-advanced and 6 EM economies. Forecasts are Macquarie where available,
alternatively the IMF. The country weights use market exchange rates, not PPP
Source: IMF, Macquarie Research, January 2017
United States: Economic activity is once again expanding at an above potential pace
following a slower rate of growth at the beginning of 2016. While the election of President
Trump introduces an added element of forecasting uncertainty, we anticipate that a cyclical
upswing remains firmly in place. As a result, we have an above-consensus growth outlook for
4Q16 (2.8%, cons. 2.2%), 1H17 (2.3%, cons. 2.2%) and 2H17 (2.6%, cons. 2.3%).
One scenario we are increasingly considering is the potential for an overheating. Several
underlying measures of inflation have been accelerating for much of the past 18 months.
Coupled with firming wages, this is leading to growing inflationary pressures.
On the policy front, we believe that the Federal Reserve is likely to continue with a rate hike
cycle that should prove gradual, cautious, and supportive of risk-taking. The pace of hikes is
likely to be, on average, 50bps per year until the upper end of the fed funds range reaches
our forecasted terminal level of 2% in mid-2019.
Fig 23 Macquaries real GDP forecasts

Calendar Year YoY (%) 4Q on 4Q (%)
2014 2015 2016 2017 2018 2014 2015 2016 2017 2018
US 2.4 2.6 1.7 2.5 2.3 2.5 1.9 2.1 2.5 2.0
Eurozone 0.9 1.9 1.7 1.6 1.6 1.0 1.8 1.8 1.5 1.7
Japan -0.1 1.2 0.7 0.8 0.7 -0.9 1.2 0.4 0.6 0.7
UK 2.9 2.2 2.0 1.6 1.3 2.9 2.2 2.0 1.6 1.3
Canada 2.5 0.9 1.3 1.7 1.6 2.4 0.4 1.6 1.7 1.5
Australia 2.6 2.4 2.3 1.9 2.7 2.2 2.6 1.8 2.6 2.8
New Zealand 3.7 2.5 3.4 3.3 2.4 4.1 2.3 3.5 3.0 2.2
China 7.2 6.8 6.7 6.5 6.0 7.2 6.8 6.7 6.5 5.7
S. Korea 3.3 2.6 2.9 2.9 2.7 2.7 3.1 2.7 2.9 2.7
Taiwan 3.9 0.6 1.3 2.0 2.3 3.4 -0.9 2.3 2.0 2.3
Hong Kong 2.6 2.4 1.2 1.5 2.0 2.5 1.9 1.2 1.5 2.0
Indonesia 5.0 4.8 5.0 5.2 5.2 5.0 5.0 4.6 5.2 5.2
Malaysia 6.0 5.0 4.1 4.5 5.1 5.7 4.5 4.4 4.5 5.1
Singapore 3.3 2.0 1.7 1.6 2.6 2.9 1.7 1.8 1.6 2.4
Philippines 6.2 5.9 6.6 6.1 6.3 6.6 6.6 5.6 6.1 6.3
Thailand 0.8 2.8 3.2 3.3 3.5 2.1 2.8 3.0 3.3 3.5
India 6.7 7.2 7.1 7.1 7.7 6.6 7.2 6.1 7.5 7.5
South Africa 1.5 1.2 0.4 1.4 2.0 1.3 0.0 0.2 1.4 2.0
Source: Bloomberg, Macquarie Research, January 2017
19 January 2017 11
China: Chinas economic growth ended 2016 stable, with an earnings recovery and inventory
restocking. We expect GDP in 4Q to be a 6.7% YoY again.
After the US Presidential inauguration, the most important political event in 2017 is the
political transition in China. Its unsurprising that Chinas policy-makers put stability as the top
priority for this year, so that they can focus on politics. What could surprise is whether and
how the market reassesses Chinas growth and reform outlook after the transition, when the
power and the responsibility for Chinas top leader could both reach the highest point in the
past forty years.
That said, the economy is likely to see a modest slowdown in 2017 as China enters a
property down-cycle. Press reports say that President Xi is open to a larger slowdown, with
growth falling below 6.5% (link), but we believe it is more likely policy-makers will maintain a
relatively stable growth trajectory around 6.5% in 2017, but will accept more bumpy growth
beyond that to carry out long-awaited and painful reforms.
The biggest surprise could be a stable yuan, compared to market expectations of another
devaluation. We expect it to end in 2017 at 6.9 to $1, near the current level, supported by
capital controls.
Eurozone: The Eurozone economy appears to have ended 2016 strongly and despite
numerous political risks another solid year is a prospect in 2017. We believe rising inflation
will prove uncomfortable for the ECB but it will be short-lived, and we expect monetary policy
to remain very accommodative.
Japan: We are expecting the global industrial recovery and moderate export growth to lead
to positive revisions over the coming year, but strong domestic growth remains elusive. Only
reform would jolt the economy onto a higher growth trajectory, in our opinion.
India: We expect growth to slow in 2H fiscal year 2017 to 6% YoY, from 7.2% in 1H. We
expect demonetisation measures to hurt consumer discretionary spending in the short term.
Small & medium enterprises and the rural economy where cash transactions are quite
prevalent will be at a disadvantage. The supply chain disruptions and adverse impact on
productivity is also likely to delay the recovery in the investment cycle.
That said, we expect the slowdown in economic activity to be largely limited, spread over the
next 6-9 months, with growth to bounce back from the second half of FY18. We continue to
expect inflationary pressures to remain contained (~4.5% CPI inflation estimated in FY18)
and see scope for a further 50-75bps cut in the repo rate over the next 12 months.
19 January 2017 12
Fig 24 What a difference a year makes! Asset prices between 19 th January 2016 and 19th January 2017
Commodity & financial asset price moves, Y-on-Y, 19th Jan, US$ terms, %
Iron ore Iron Ore +142%

Coking coal Coking coal +125%
Brent
Zinc
WTI
Heating Oil
Russian stocks
Thermal coal
Natural Gas
Tin
Palladium
Coffee robusta
Sugar 11
GSCI Spot
Lead
Orange Juice
LMEX index
BCOM spot Equity indices
RUB FX
Lumber Bonds
Copper Industrial metals/raw materials
Rhodium Precious metals
Coffee C Agricommodities
Moly Energy
Sugar Commodity indices
Soybean meal
BZL
GSCI TR
Oats
Nasdaq
Aluminium
Dow Jones
ZAR
Soybean
Silver
S&P 500
BCOM TR
Nickel
Soybean Oil
Cotton
Platinum
Nikkei
NZD
Gold
CLP
CAD
AUD
EuroSTOXX
European banks
Japan bonds
TIPS
Lean Hogs
JPY
Dollar Index (inv)
PEN
US bonds
German Bonds
CHF
INR
Corn
Eurozone bonds
Shanghai Comp
Euro
CNY
Live cattle
UK bonds
Rough Rice
Wheat
GBP
Feeder Cattle
Uranium
(40%) (20%) 0% 20% 40% 60% 80% 100%

Source: Bloomberg, Macquarie Research, January 2017
19 January 2017 13
China Outlook
Demand has surprised to the upside, but we see slowing as inevitable
Chinas economic performance in 2016 surprised to the upside, and while demand over the
first half of the year was driven largely by stimulus impacts to infrastructure and property,
demand over the second half clearly became more broad based. Our proprietary steel and
copper surveys have given us extremely good insights into where Chinas economic demand
has been coming from in 2016, and showed a clear improvement in demand from the autos,
appliances and machinery sectors in 2H16.
Looking ahead we maintain our view that commodities demand from China should be robust
through mid-2017, but potentially fade towards the year end. This view has been based on an
expectation of follow through effects on 1H17 construction activity from infrastructure
spending and housing market strength in 2016, together with our macro teams view that the
government should remain biased to keep policy reasonably supportive ahead of the 2017
mid-year leadership reshuffle at the end of President Xis first five year term.
The property market as always remains key to Chinas commodity demand, and this clearly
looks set to slow this year as the government has become increasingly unhappy with the
pace of price appreciation. So far tightening measures have been largely seen on a city-by-
city basis, but they are clearly having an impact. Property sales growth slowed to 7.9% YoY in
Nov from 34% in Sep. Through Jan-Nov, investment and new starts rose by 6.5% and 7.6%
YoY, compared with +1% and -14% in 2015, but the pace has been slowing since 1H16,
particularly for new starts, and floor space completions showed a big decline in its growth
rate, down from 12% Jan-Sep to 6.4% Jan-Nov. If property sales growth continues to slow
down, and our property team forecasts -10% sales volume for this year, it should transfer to a
slower growth in new starts and construction volumes in 2H17.
Property was a major driver for Chinese banks new lending last year, i.e. a major source for
liquidity increase. Household medium-long term loans, which is mainly mortgages, accounted
for 45% of total new lending in 2016 and this ratio was as high as 58% from April to October.
Mortgage lending will be a key indicator to watch as a slowdown may indicate a broader
tightening policy toward property, and thus accelerate any decline in sales and potentially
construction activity later in the year.
Fig 25 Infrastructure, auto and machinery the main Fig 26 FAI was driven by infrastructure in 1H16, but
drivers for domestic steel orders in 2H16 improvement seen in other sectors over 2H16
Net change in orders at mills and traders - end use sectors Total Manufacturing
45
Infrastructure Real estate
FAI growth breakdown, YoY % 3mma
50 40
40 35
30 30
20 25
10
20
0
15
-10
10
-20
5
-30
Construction Infrastructure Machinery 0
-40
Autos White goods Shipbuilding
-50 -5
Sep-10
Sep-11
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
May-10
May-11
May-12
May-13
May-14
May-15
May-16
Jul-11
Jan-12
Jul-12
Jan-13
Jul-13
Oct-13
Jan-14
Jul-14
Jan-15
Jul-15
Jan-16
Jul-16
Jan-17
Apr-12
Apr-13
Apr-14
Apr-15
Apr-16
Oct-11
Oct-12
Oct-14
Oct-15
Oct-16
Source: Macquarie China Steel Survey, January 2017 Source: NBS, CEIC, Macquarie Research, January 2017
Infrastructure spending accelerated strongly in 2016, but this has resulted in significant front
loading of what is essentially a very bearish 13th five year plan. FAI growth projections look
increasingly challenging at this elevated pace, and we see a slowdown in infrastructure
spending as inevitable this year. With economic growth looking broader based, and ex-China
economic optimism also on the upturn, policymakers in Beijing should have greater
confidence to slow their support to the economy via infrastructure spending, especially as
concern around inflationary pressures is clearly on the rise.
19 January 2017 14
Fig 27 Daily property sales cooling Fig 28 Inflation is clearly picking up
60 Total Residential YoY % PPI CPI

15
50
40
10
Sales Volume, YoY %
30
20 5
10
0
0
-10
-5
-20
-30 -10
Dec-07
Nov-10
Oct-06
Apr-10
Oct-13
Dec-14
Aug-05
Sep-09
Jan-05
May-07
Aug-12
Sep-16
Jun-11
Jan-12
May-14
Mar-06
Jul-08
Feb-09
Mar-13
Jul-15
Feb-16
Jan-11
Jan-16
Jan-12
Jan-13
Jan-14
Jan-15
Sep-11
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
May-11
May-12
May-13
May-14
May-15
May-16
Source: WIND, Macquarie Research, January 2017 Source: CEIC, NBS, Macquarie Research, January 2017
Indeed we believe policymakers are increasingly concerned with keeping inflation under
control. PPI inflation turned positive in September for the first time in 54 months, mainly
driven by higher price indices from the steel, base metal and coal industries. While CPI has
remained steady thus far, the recent rise in PPI is expected to lead to a higher CPI in 2017 as
costs are passed through, and a higher CPI could lead to a more prudent macro policy once it
gets close to the government target of 3%. Of course the upside from inflation is that nominal
GDP has to work harder to achieve the underlying real GDP targets, and that is normally
good news for commodities demand.
The direction of the RMB remains a key theme for commodities in general. In 2016 the PBoC
managed to continue with a slow, gradual depreciation of the RMB while trying to stem the
consequent capital outflows, though as depreciation expectations built pressures on outflows
again it intensified towards the year end. Our economics team expects to see more volatility
in the RMB this year, in order to lessen depreciation expectations, but have a year-end
forecast of 6.9 RMB/USD, essentially looking for no depreciation versus end 2016. Potential
further RMB devaluation will remain a key theme to watch in the commodities world,
especially as it pertains to Chinas export competitiveness in steel and potentially aluminium.
News coverage on Chinese supply-side reforms remains highly topical, but while plenty of
closures have been publicised, we still believe little effective capacity has closed and will be
closed near term. We remain highly sceptical that there is any real underlying change to the
way heavy industries are operating. Indeed, the latest statements from the government
towards steel have outlined a three stage process, which doesnt look for industrial
restructuring to take place until the 2020-2025 period.
Fig 29 RMB has seen a gradual depreciation, but will Fig 30 Liquidity remains plentiful, and lending
this continue? continues to rise
3,000 2012 2013 2014 2015 2016

8.25
CNYUSD
8.00 2,500
New Rmb loans, Rmb bn
7.75
2,000
7.50
7.25 1,500
7.00
1,000
6.75
6.50
500
6.25
6.00 0
Jul
Oct
Nov
Dec
Jan
Jun
Feb
Mar
May
Apr
Aug
Sep
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
Source: Bloomberg, Macquarie Research, January 2017 Source: CEIC, Macquarie Research, January 2017
19 January 2017 15
Metals and Bulks Executive Summary

Passing through the sweet spot
The past few months have seen yet another leg higher for many metals and bulk commodity
prices, offering further salvation to resource producers. We believe many of the factors which
have been helping over the past six months will still be doing so in the initial part of 2017.
Global manufacturing continues to recover, while the Chinese government has repeated their
desire for market stability, which necessitates ongoing focus in infrastructure spending to
underpin the 6.5% lower threshold for real GDP. Moreover, while property sales are now
trending negative, we feel there is enough momentum in Chinese construction activity to
sustain positive demand conditions through mid-year. With this, in our December price
changes we once more raised our demand forecasts for most metals and bulk commodities
for the coming twelve months, and thus our price expectations.
The window of opportunity for resource producers remains open at present, but the pertinent
questions now surround the duration of this event. We feel a return to cost curve normality
(but certainly not January 2016 lows) is likely over the coming 12 months, most likely through
fading Chinese construction activity into mid-year. The risk that this happens faster would be
the Chinese government worrying about inflation, while for it to happen slower would involve
further confidence in healthy reflation flowing through to manufacturing and construction
globally. Given the lack of visibility, we see little reason to change to our medium and long
term price assumptions, which are based on the return to a low industrial growth environment.
We also see no reason for the global metals and bulks industry to be spending growth capex.
At present, we would look for exposure to those commodities with raw material constraints or
risks thereof, or those with cyclical margin expansion. These include stainless steel, chrome,
alumina, cobalt and steel. Following the relaxation of the Indonesian ore ban, we now see
downside rather than upside for nickel. While our base case Q1 copper price is below todays
levels, given the tightening market balance and risk of supply disruption, the skew is to the
upside something perhaps already reflected in recent pricing. However, into H2 2017 we
see most commodity prices below todays levels. In particular, we feel bulk commodities will
see consistent downward pressure, though in the main producers will still be generating
significant free cash flow. For gold, 2017 is likely to see some early year relative weakness
before outperformance in H2.
Our longer-term preferences for nickel, copper, silver and gold persist, while we are also
more bullish than consensus on iron ore into the end of the decade. Meanwhile, uranium is
now where other commodities were at this time last year trading too far into the cost curve
to be sustainable. We would avoid sustained exposure to steel, aluminium, alumina and
thermal coal, where we think medium-term pricing expectations are too robust relative to the
structural challenges these industries face.
Fig 31 As demand expectations have been upped, we Fig 32 Spot price moves have ranged from strong to
continue to be in a price upgrade cycle spectacular compared to this time last year
Changes in price forecast over coming 12 months Spot price change over last 365 days
20
148%
15
UP 160%
125%
10 140%
5 120%
87%
0 100%
71%
63%
68%
57%
-5 80%
51%
43%
60%
32%
-10
25%
20%
19%
17%
40%
-15
9%
20%
-20
DOWN 0%
Platinum
Aluminium
Zinc
Tin
Gold
Alumina
Iron Ore
Nickel
Palladium
Lead
Steel
Met Coal
Copper
Silver
Thermal Coal
-25
Jan 11
Jan 12
Jan 13
Jan 14
Jan 15
Jun 16
Dec 15
May 11
Sep 11
May 12
Sep 12
Jul 13
Sep 13
May 14
Sep 14
May 15
Sep 15
Mar 16
Sep 16
Dec 16
Apr 13
Source: LME, MB, CRU, TSI, Macquarie Research, January 2017 Source: LME, MB, CRU, TSI, Macquarie Research, January 2017
19 January 2017 16
Fig 33 Macquarie metals and bulk commodity price forecasts

2015 2016 2017 2017 2017 2017 2017 2018 2019 2020 2021
Unit CY CY Q1 Q2 Q3 Q4 CY CY CY CY CY LT $2016
Copper $/tonne 5,503 4,863 5,350 5,500 5,350 5,200 5,350 5,100 5,188 5,725 6,250 5,900
Aluminium $/tonne 1,663 1,604 1,700 1,650 1,550 1,500 1,600 1,494 1,500 1,500 1,550 1,350
Zinc $/tonne 1,932 2,092 2,600 2,750 2,850 2,950 2,788 2,775 2,475 2,325 2,188 2,300
Nickel $/tonne 11,836 9,599 10,250 9,750 9,750 9,500 9,813 11,000 12,000 13,000 14,000 15,000
Lead $/tonne 1,786 1,871 2,100 2,300 2,450 2,550 2,350 2,351 2,103 1,945 1,928 1,950
Tin $/tonne 16,077 17,991 21,500 23,000 22,000 20,500 21,750 21,375 23,000 21,000 20,500 18,000
Manganese ore $/mtu CIF 2.9 4.6 7.0 5.5 4.8 3.8 5.3 3.5 3.8 3.8 4.0 2.8
FeCr (EU contract) c/lb 107 96 165 170 150 130 154 140 125 125 133 110
Molybdenum oxide $/lb 7 7 8 8 7 7 7 7 7 8 8 10
Cobalt (99.8% ) $/lb 13 12 14 14 14 14 14 14 15 17 18 13
Alumina $/t FOB 301 254 350 335 325 325 334 276 250 230 240 225
Steel - Average HRC $/tonne 416 451 539 491 449 429 477 410 418 423 423 410
Steel Scrap - average #1HMS $/tonne 210 204 243 227 207 207 221 197 207 243 243 205
Iron ore - Australian fines c/mtu fob 81 86 88 79 71 70 77 65 69 84 83 86

Iron ore - Australian lump c/mtu fob 96 102 102 93 83 82 90 80 86 101 100 101
Spot 62% Fe iron ore China $/t cfr 56 58 60 55 50 50 54 47 50 60 60 60
Hard coking coal $/t fob 102 114 285 230 175 155 211 140 130 135 140 115
Semi-soft coking coal $/t fob 78 85 171 161 123 109 141 98 91 95 98 81
LV PCI coal $/t fob 84 88 180 173 131 116 150 105 98 101 105 86
Coke - China export spot $/t fob 144 192 290 240 190 170 223 160 150 160 160 140
Gold $/oz 1,160 1,248 1,115 1,125 1,275 1,350 1,216 1,375 1,363 1,400 1,375 1,250
Silver $/oz 15.7 17.1 16.8 17.0 19.0 20.5 18 21.3 21.8 22.0 22.8 18.0
Platinum $/oz 1,053 986 950 975 1,025 1,075 1,006 1,188 1,306 1,338 1,300 1,400
Palladium $/oz 692 612 700 650 625 650 656 756 813 769 738 800
Source: LME, TSI, CRU, Metal Bulletin, Macquarie Research, January 2017
Fig 34 Our forecasts show the vast majority of metals Fig 35 but on a Q4 comparison basis, precious will
and bulk commodities higher on average in 2017 be higher, bulks lower
2017 average price vs. 2016 Q4 2017 vs. Q4 2016 forecast price
25%
100% 19% 19%
85% 17%
14%
80% 15% 12% 11%
6%
60%
5%
40% 33% 31%

26% -5% -1%
20% -5%
10% 9% 7% 7% 6% 6%
2% 2% -11% -12%-12%
-15%
0%
-15%
0% -3%
-8%
-20% -25% -23%
-21% -26%
-40% -30%
-35%
Copper
Tin
Lead
Nickel
Iron Ore
Alumina
Silver
Platinum
Zinc
Steel
Aluminium
Gold
Uranium
Palladium
Thermal Coal
Hard Coking Coal
Uranium
Lead
Tin
Copper
Iron Ore
Silver
Platinum
Alumina
Aluminium
Steel
Zinc
Gold
Nickel
Palladium
Thermal Coal
Hard Coking Coal
Source: LME, TSI, Bloomberg, Macquarie Research, January 2017 Source: LME, TSI, Bloomberg, Macquarie Research, January 2017
19 January 2017 17
The factors central to metals and bulk commodity markets over the coming year
2017 is beginning with many of the same supportive factors for metals and bulks, both from
fundamental and financial markets. However, in an environment of elevated political turmoil
where global economic growth is yet to meaningfully accelerate, the duration of such support
is the key 2017 factor, particularly given many prices are now invoking a supply response.
We consider the following themes particularly important for metals and bulk commodity
markets over the coming period.
Plenty of free cash flow: Resource producers are currently having a good time of it at
present. From being happy to be here after the commodity price slump, many are now in a
position where free cash flow is accelerating quickly. Commodity prices have risen, and costs
have not (yet). As our European resources analyst Alon Olsha noted recently, even without
the benefit of spot commodity prices, we estimate that RIO, South32 and Glencore in
particular could boost yields significantly whether in the form of dividends or buy-backs
due to low gearing levels or dividend policy flexibility. The market appears to be under-
appreciating the cash yield potential of the diversified miners and while scepticism is
understandable given the sectors woeful capital allocation track record, we believe the more
restrained approach to spending adopted by the miners will persist, partly because of a lack
of confidence in the sustainability of the recent price rally.
but only a modest capex uplift: This lack of confidence means that, while sustaining
capex may recover, we have less confidence in growth capex. To be clear, in many of these
markets given our medium-term demand projections and lack of permanent supply closures,
we simply dont need major new supply projects to be incentivised. This is particularly true
given the lack of visibility on Chinese demand trends beyond the latter half of 2017. Mining
capex has likely passed its nadir, but is not set to rise sharply. Meanwhile, companies focus
on incremental operational gains and further debottlenecking is accelerating, such that some
existing assets will see planned gains.
Peak overcapacity?: Our concerns over excess capacity in many markets have not faded,
however we may now be passing the worst of the situation. After the capex void limited new
projects are coming to market, but perhaps more importantly the Chinese government is
ramping up efforts to close obsolete capacity. It will take a long time, but capacity utilisation
in downstream industries should start to steadily improve.
Fig 36 Higher prices and stagnant costs mean Fig 37 with almost all commodities now trading out
significant free cash flow upside for miners of the cost curve unlike this time last year
70,000 Proportion of supply losing money - cash basis

80%
70% As of Jan 16 Current
60%
South32 50%
Glencore 40%
US$mn
Vale 30%
Anglo American 20%
Rio Tinto 10%
BHP Billiton 0%
Copper
Mn Ore
Met Coal
Zinc
Alumina
Gold
Aluminium
PGM Basket
Thermal Coal
Nickel
Iron Ore
Uranium
-
2016F 2017F 2017F (spot)
Source: Company Data, Macquarie Research, January 2017 Source: Wood Mackenzie, Macquarie Research, January 2017
The fiscal stimulus hope: At present, the world seems fixated on fiscal-led reflation, which
is certainly benefitting commodity markets from an asset allocation standpoint, and is likely to
over the coming months. While fiscal stimulus itself is likely to have only minimal impact on
demand, and even then with a lag, rising inflation expectations will tend to support interest in
commodities. In particular, wed highlight the potential for copper demand from the much-
needed rebuilding of distribution grids in global cities as the obvious beneficiary of such state
spending. However, should expectations shift from healthy reflation towards stagflation or
even disinflation, industrial metals would suffer while gold prospers.
19 January 2017 18
We certainly worry that a bout of global fiscal stimulus will simply lead to a front-loading of
future metals demand. Certainly, there is an argument that developed world infrastructure is
in dire need of upgrading through (particularly in the US, which has been waiting ~40 years
for infrastructure spend). However, there is a limit to the extent and duration of such
spending. To be clear, we are not saying fiscal stimulus will stop anytime soon, and if we
move to an environment where governments can borrow for free then debt will never be
called, however every incremental dollar is likely to be increasingly less metals intensive. For
example, rebuilding a bridge is easy and highly metals intensive, but once it is done it is done.
Spending money on R&D or space programs, perhaps the next stage, will still drive growth
but without the same metals intensity.
Fig 38 We expect a positive response from marginal Fig 39 We also expect expansion capex to pick
capacity to higher prices if supply flexibility persists upslightly
10% Forecast supply growth, 2016 & 2017 Global Mining Expansion Capex, $bn
160
8%
6% 140
4% 120
2%
100
0%
-2% 80
-4%
60
-6%
-8% 40
Zinc
Silver
Gold (mined)
Aluminium
Platinum
Nickel
Stainless Steel
Thermal Coal
Steel
Met coal
Copper (mined)
Uranium
Palladium
Lead (mined)
Iron Ore
20
2007
2008
2009
2010
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2011
2012
2013
2014
2015
2016
2017F
2018F
2019F
Source: Wood Mackenzie, Macquarie Research, January 2017 Source: Wood Mackenzie, Macquarie Research, January 2017
The broad-based nature of demand: The rise in commodity markets is not all about the
reflation trade, nor is it solely to do with Chinese property. That demand continually surprised
to the upside throughout 2016 owes much to the age-old driver of global manufacturing.
Certainly, this is hardly booming at a ~2% YoY (but accelerating) growth rate. It is however
better than the close to zero rate seen this time last year. As the industrial recovery reaches
a more mature stage, we see further gains from improving oilfield activity and mining itself
being less of a drag, and thus supporting commodities demand. Meanwhile, Decembers
global manufacturing PMI was the highest since 2014. There is still a risk that recovery is
derailed by either rampant inflation or political shocks; however, it does look reasonably
robust at the current time.
Fig 40 Notwithstanding the base effect, apparent Fig 41 to semiconductors, as the industrial
demand has been recovering, from steel recovery has matured
15%
Global steel consumption Global semiconductor shipments
15%
3MMA, YoY % change
10% 10%
5%
YoY change
5%
0% 0%
-5%
-5% Value basis
Volume basis
-10%
-10% 2011 2012 2013 2014 2015 2016
2011 2012 2013 2014 2015 2016
Source: worldsteel, Macquarie Research, January 2017 Source: Company Data, Macquarie Research, January 2017
19 January 2017 19
The turn in the Chinese property and construction cycle: In terms of end use drivers,
Chinese construction activity is still the single most important for metals and bulk
commodities. A key part of the 2016 recovery in metals demand was the relative strength in
the Chinese property market contributing to a more commodity-friendly growth environment.
However, we feel the cycle is now turning.
We expect the Chinese government to prolong its more commodity-intensive phase of growth
through the political transition in 2017. We expect new construction starts will still accelerate
out of Chinese New Year, and when supported by ongoing strength in infrastructure will still
see decent YoY apparent demand growth from China for industrial metals in H1 2017 (helped
by a low base). Into H2 however, the construction sector is likely to once more prove negative
for steel (and its raw materials) while rates of base metal demand growth will slow. As such,
we see a relatively short window of opportunity for commodity prices and commodity
producers before the demand tailwind fades. And, given the slowing urbanisation rate, the
property sector is likely to be an ongoing headwind to steel demand we see steel into
residential property in 2020 at a level ~15% below 2013s peak.
Chinese construction data will be of paramount importance in Q2 2017, with any sign of
sequential weakness a red flag. However, should construction activity continue to surprise on
the upside, the main supportive factor for global commodities demand will persist.
Fig 42 Our proprietary China surveys have been Fig 43 and a period of sustained profitability has
showing consistently strong orders undoubtedly helped sentiment
Have domestic orders increased or decreased over the last What is the current level of profitability?
month?
100 Increasing no. of mills seeing orders rise 100
Total
90 90 Total
Increasing no. of mills making money
80 80
70 70
60 60
50 50
40 40
30 30
20 20
10 Increasing no. of mills seeing orders fall 10
Increasing no. of mills losing money
0 0
Jul-11
Jan-12
Jul-12
Jan-13
Jul-13
Jan-14
Jul-14
Jan-15
Jul-15
Jan-16
Jul-16
Apr-13
Apr-15
Oct-11
Apr-12
Oct-12
Oct-13
Apr-14
Oct-14
Oct-15
Apr-16
Oct-16
Apr-12
Apr-13
Apr-14
Apr-15
Apr-16
Jul-11
Oct-11
Jul-12
Oct-12
Jul-13
Oct-13
Jul-14
Oct-14
Jul-15
Oct-15
Jul-16
Oct-16
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Source: Macquarie China Steel and Copper Surveys, January 2017 Source: Macquarie China Steel and Copper Surveys, January 2017
The scale and speed of marginal supply response: 2016 was always going to be the year
of weak or negative supply growth, owing to the lack of new projects post-capex cuts and the
closure of existing operations at the low prices we saw at the start of the year. For 2017,
however, we do expect some reaction, particularly given pretty much every metal (with the
notable exceptions of nickel and uranium) is trading at levels where pretty much every unit of
supply is making money.
Over the past decade, supply-demand modelling in industrial metals and bulk commodities
has had to incorporate a major balancing factor flexible Chinese supply which reacts to
price prompts. This supply is generally low capex, high opex and operates on a short time
horizon for decision making. Or to put this another way, rather than 20-year asset lives, if it
makes money on a 10-week time horizon output will increase, and if it doesnt it will exit the
market. Such cycles in and out have become an important aspect of supply.
However, it is fair to say that we were underwhelmed by the flexible reaction in 2016, in a
year where we expected weak or negative supply growth, owing to the lack of new projects
post-capex cuts and the closure of existing operations. With many commodities having been
trading at a level where a decent amount of the marginal supply is making money, including
latent Chinese capacity, a supply reaction would again be expected over H1 2017. And in
certain cases we have already seen this, such as in iron ore and crude steel, where domestic
supply has been reacting in an efficient manner to price.
19 January 2017 20
However, this hasnt been the case for all metals. Most notably, the Chinese zinc mine
supply response we were expecting to pricing above $2,000/t did not emerge, while Chinese
aluminium restarts took until Q4 to accelerate despite economics suggesting they were viable
all year. In our view there were four factors behind this: a lack of confidence in sustainability
of demand, alternative opportunities for capital, a lack of available finance for operations and
enhanced environmental regulation. To a greater and lesser extent, they are all still present.
Enhanced Chinese environmental regulation: In our view, this is perhaps the most
serious restriction on latent capacity returning to market. Unfortunately, it is also perhaps the
hardest to get a solid handle on. Certainly, the anti-corruption push in 2015, coupled with the
clear focus on Chinas conspicuous environmental problems, has seen local governments
take a more robust approach. As noted above, this certainly played a major role in zinc mine
output disappointment last year, and also in the slow nature of alumina restarts versus
aluminium. And if anything, this factor is only going to be becoming more stringent over the
coming year.
All in all, we still expect some response from Chinese supply. High prices in commodity
markets do find a solution, and economics still work in China. It is fair to say that they work
less efficiently than before, however, and thus the Chinese supply response we factor in is
perhaps as much at risk of disruption as that from international peers. Together with zinc and
alumina, nickel, tin and iron ore lump/pellet would be relative beneficiaries of a muted supply
response owing to environmental pressures.
Cost curves matter again: At present, cost curves barely matter given the underlying
support from commodities. However, with costs set to rise and demand growth merely
decent rather than aggressive, this doesnt feel like a sustainable situation. We expect a
return to cost curve-based normality over the coming year given most markets should be
adequately supplied. To be clear, we are not expecting to see prices trading aggressively
into these curves la January 2016, but rather to a level where marginal producers are again
under some pressure. Meanwhile, the game for the majors will be to move asset portfolios
down these curves to be best positioned for the longer cycle, where operational gains really
matter.
The wildcard of government policy: Theres no getting away from it in commodities, just
as with pretty much any other asset class, government policy is garnering ever more
influence. In a world where governments are struggling with disenchantment among the
electorate and stagnant global growth, policies are increasingly working to a new playbook.
And as government policy changes, so does the lay of the land for many commodities.
Of course, this is not a new thing for metals and bulk commodity markets. Following the
Chinese government has been a core part of analysis here for the past fifteen years. Take
the strong liquidity push right at the start of 2016 or the 276-day policy for the domestic coal
industry for example as to the impact this can have. Meanwhile, the recent surprise from
Indonesias relaxation on nickel ore and bauxite shipments highlights that government
decisions can be founded on non-economic grounds and have the propensity to change
markets overnight.
2017 is going to be particularly pertinent for policy in the worlds two largest economies. In
the first part of the year we will see what a patriotic Trump government strategy will actually
look like for the US. In the second half, we will see what a patriotic President Xis plan is for
China for the coming five years. Add to that the fact that 2017 will almost certainly determine
whether the Eurozone huddles together or breaks apart means policy risk remains top of
most agendas. This will be important for future demand projections if nothing else.
Protectionism and two-tier commodity pricing: As with any period of political uncertainty,
governments become more insular and less global in their thinking and actions.
Protectionism is no longer a maybe, it is now active and growing, with trade and commodity
markets at the forefront. Trumps protectionist campaign is a case in point, and something we
discussed in this note, but other countries are also upping the ante given the heavy focus on
manufacturing (i.e. high productivity) jobs. This benefits domestically-focused producers over
those who rely on exports.
19 January 2017 21
We expect the protectionist push to lead to further fragmentations of metals (and perhaps
energy) market pricing. We foresee a situation where domestic commodity prices are in
many cases artificially amplified relative to those seen internationally via trade barriers and
elevated customs checks, creating a two-tier pricing system. In base metals, this might be
absorbed via premiums over LME, but for others regional arbitrages will develop. Meanwhile,
in coal we expect China to continue to support the domestic coal price but perhaps return to
enhanced customs checks on imports (under the environmental protection banner), leading to
a working capital and operational arbitrage between a higher Chinese domestic price and
lower international one.
From a wider perspective, the push towards de-globalisation will only delay further the
necessary capacity reset many commodity markets still need. This is most notable for steel
and aluminium, as governments increasingly step in to support uneconomic domestic assets,
keeping these markets in a perpetual state of overcapacity. The US steel industry is an
example of what we may face, having been strongly protectionist already for the past 18
months. This led to a situation where, in early 2016, US buyers of steel hot rolled coil were
paying double what their peers in China were for the same material. Protectionism is an
effective tax that ends up being passed to the consumer, and for the US will partially offset
the actual tax cuts which seem certain to come through.
Chinese cost structures: For most refined metals, the cost structure of the Chinese
industry now plays a crucial role in global price setting. This is because in our view, the
Chinese business model in commodities remains pretty clear. Source the raw material, either
domestically or internationally, and have suitable capacity to produce the refined product. And
when there is plenty of raw material available, this is generally deflationary given Chinas
ever-increasing competitiveness in smelting and refining.
However, when this business model is disrupted through a lack of raw material supply, the
result is price inflation. This is what we saw in the stainless steel raw materials of nickel and
chrome, which are highly susceptible to this, given Chinas lack of domestic raw material
supply potential. In our view, the cost structure of producing nickel pig iron (NPI) and
ferrochrome in China from imported raw materials goes a long way to setting the price for
these products.
Fig 44 For refined markets, the feed through from raw Fig 45 We have already seen the creation of two-tier
material prices to costs in China remains crucial steel pricing through protectionism
LME Price vs 10%Ni RKEF costs, $/tonne HRC price premium

21000 120%
19000 100% US over China
17000
80%
15000
60%
13000
11000 40%
9000 20%
7000 LME Price 10%Ni RKEF Costs 0%

Jul-12
Jul-13
Jul-14
Jul-15
Apr-16
Jul-16
Jan-14
Apr-12
Oct-12
Jan-13
Apr-13
Oct-13
Apr-14
Oct-14
Jan-15
Apr-15
Oct-15
Jan-16
Oct-16
5000
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
Jul-16
Jul-13
Jul-14
Jul-15
Nov-13
Jan-16
Nov-12
Jan-13
Jan-14
Nov-14
Jan-15
Nov-15
Nov-16
Mar-13
May-13
Mar-14
May-14
Mar-15
May-15
Mar-16
May-16
Source: SMM, LME, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
Financial market evolution: There is a lot of attention being paid to Chinese financial
market influence on commodities at the present time. Certainly, Chinese players are
significantly more active in financial markets than they were 2-3 years ago, and in terms of
positioning are much more aggressive than developed world counterparts. Traded volumes
on Dalian contracts for bulks are multiples of those traded outside China indeed iron ore
has become a financial market as the marginal physical buyer is now influenced by the daily
moves in futures prices.
19 January 2017 22
Part of the Chinese interest in commodities is down to a lack of attractive alternatives for
asset allocation in China at the moment. Property is being clamped down on, the bond
market isnt developed, real interest rates are negative, and the A-share market is still volatile.
The government are still spending strongly in FAI and reflation is expected. The natural
default is commodities. As a result, this also drives natural physical speculation (i.e.
restocking) particularly while future order books look healthy. But a restock can quickly turn
to a destock should the cycle turn.
Perhaps owing to this, towards the end of last year the Chinese government has shown some
irritation towards the role of domestic retail investors (i.e. speculators) in commodity futures
markets. They have successfully flushed them from the coal market, but it took multiple
interventions to raise margins to achieve this. At present, we have only seen rhetoric that the
government are not happy but given what happened with A-shares last year more aggressive
regulation is certainly possible. This would reduce volatility in commodity markets, but would
also see an abrupt removal of retail support for commodity prices.
Meanwhile, as on-exchange stocks (though not necessarily total stocks) continue to draw for
base metals, these markets will look optically tighter to financial market participants.
A risk premium for certain long-term prices: 2016 highlighted that acute supply shocks
(or at least inelasticity) can have a dramatic upward effect on pricing. Thus, for many
commodities where we are happy with a long-term normal, given geographical concentration
of supply leaves them prone to such problems once every few years, you could argue for a
risk premium to a standard long run equilibrium price. Those most in this camp would be met
coal, chrome, cobalt and manganese ore.
Fig 46 When Chinese supply-side reform starts in Fig 47 While overall inventories are not necessarily
earnest, it will help ease the overcapacity in steel falling, those on exchange are for zinc and aluminium
Global steel overcapacity - set to improve Aluminium stocks on exchange

300
6000
kt
250
5000
200
million tonnes
4000
150
3000
100
50 2000
0 1000
-50 0
-100
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018f
2020f
LME CME SHFE
Source: worldsteel, Macquarie Research, January 2017 Source: LME, CME, SHFE, Macquarie Research, January 2017
19 January 2017 23
Energy Executive Summary

Co-ordination, compliance and caution
2016 was a year that saw a plethora of improbable binary events and the energy market was
no exception. We were surprised be the OPEC agreement, the extent of the lithium price
surge, Chinas extent of coal supply intervention and the underperformance of uranium.
In a maturing industrial recovery, we are comfortable with the wider demand perspective for
global energy markets. However, depending on government policy, the outcomes of the
supply side can be wide. Compliance to the OPEC agreement will be closely monitored for
any slippage, as will President-elect Trumps energy policy, particularly should we see a coal
resurgence on the back of growing LNG supply. Meanwhile, US onshore oil production is
reacting quickly to the OPEC deal, something we expect to continue particularly with the
volume of drilled and uncompleted wells available. In the potential for a large production
response from the US, we see significant risk of OPEC reverting back towards dysfunction
from the current spirit of coordination, leading to an eventual reversal from 2017 production
cuts.
Like other commodity markets, energy producers are feeling much better about life. And yet,
given the high level of political and trade risk, they are not willing to recommit to growth capex
plans. As a result, cash flow will remain strong, and even in markets such as coal and lithium
where spot prices are set to fall as supply grows, there is high confidence they will not revert
to end-2015 levels. Meanwhile, uranium is now where many other commodities were at this
time last year trading too far into the cost curve to be sustainable.
Fig 48 Major oil producers are still in capex reduction Fig 49 but following recent events we expect a US
mode production response - to record levels in 2018
Source: Factset, Macquarie Research, January 2017 Source: IHS, Macquarie Research, January 2017
A 2017 oil focus on deal compliance and extension, concerns mount for 2018: OPEC
defied expectations and not only managed to coordinate the first production cut of the
decade, but also corralled a significant group of Non-OPEC producers to agree to reduce
supply by a combined ~1.8 MBD. If the OPEC/Non-OPEC group can execute on the deal,
the global supply-demand balance will move into deficit for the first time since early 2014.
With a robust shale response mounting and an eventual reversal of cuts expected, we believe
the bill for 2017 relief will come due in 2018, where we see inventories again building.
US natural gas all set for a tight summer: After moving to an uncomfortably bullish posture
on natural gas ahead of winter, given stalling supply and rising real demand, we continue to
see support for natural gas prices through summer 2017. With strong underlying gas demand
growth likely offsetting price-related impacts to power demand (reverse coal to gas switching)
healthy production growth will be required to achieve an acceptable level of storage at the
end of the 2017 injection season.
19 January 2017 24
Thermal coal = Chinese government control: Thermal coal prices have finally eased since
November on the back of the numerous changes to the 276 days policy seen since
September. Spot thermal coal prices in Qinhuangdao have fallen from highs in November of
over 730RMB/t, to just under 600RMB/t in mid-January, as domestic mine supply has clearly
picked up and inventory is rising. In terms of the outlook for prices in 2017, all eyes remain
focused on what happens next regarding Chinese policy. If prices continue to slide, we expect
that some form of output restrictions will come back with the 276 days policy, which in
November was suspended until the end of 1Q, being reapplied to some mines. Beijing
policymakers now essentially provide a floor to global coal prices, with a Beijing put
underwriting the market.
Fig 50 The relaxation of Chinese thermal coal policy Fig 51 Rising global EV sales point to a stronger
has had a prompt production reaction lithium demand outlook
China thermal coal production run rate EV sales as % of total
3.5%
Btpa
2.6
3.0% US China
2.4 Japan Europe
2.5% Canada
2.2
2.0%
2.0
1.5%
1.8
1.0%
1.6
0.5%
1.4
0.0%
Jan-14 Jul-14 Jan-15 Jul-15 Jan-16 Jul-16
1.2
2012 2013 2014 2015 2016
Source: SXcoal, Macquarie Research, January 2017 Source: National data, Macquarie Research, January 2017
Uranium price too low to be sustainable: 2017 has gotten off to a decent start for 2016's
worst performing commodity, with expectations of supply adjustment seeing prices up >10%
from the start of the year. However, when the initial base was so low, that only equates to a
$2/lb move in the spot price. To put things in context, uranium pricing is currently trading
50% of where it was 40 years ago in nominal terms never mind adjusting for inflation.
There is no other commodity for which this is true. Essentially, this has put uranium in the
situation where many peer commodities were at this time last year, trading too far into the
cost curve for pricing to be sustainable.
Lithium focus shifts to Australian supply: Lithium was certainly last years main
commodity story, given the surge in pricing and excitement around future demand growth.
That demand story remains extremely robust indeed the strongest among all the
commodities covered in this report. However, the surge in pricing has been met by a surge in
projects from a pent-up mining industry looking for a positive story. Lithium is unlikely to
return to pre-boom norms, however the Australian supply growth set to hit the market in 2017
should result in some downward pressure on spot prices. It remains a good story, particularly
from the demand side, but the period of maximum excitement is now likely to be behind us.
Fig 52 Macquarie oil, gas, coal, uranium and lithium price forecasts
2015 2016 2017 2017 2017 2017 2017 2018 2019 2020 2021
Unit CY CY Q1 Q2 Q3 Q4 CY CY CY CY CY LT $2016
Energy
Crude Oil - Brent $/barrel 52 43 56 62 65 62 61 68 73 75 77 68
Crude Oil - WTI $/barrel 49 43 51 57 60 57 56 62 67 69 71 63
Henry Hub Gas $/MMBTU 2.6 2.5 3.2 3.0 3.5 3.0 3.2 3.3 3.3 3.2 3.2 2.9
Thermal coal - Aus Spot $/t FOB 59 66 75 73 70 70 72 65 63 60 58 48
Uranium $/lb 37 26 20 20 22 22 21 24 27 30 33 33
Lithium carbonate $/t CFR China 5,190 8,447 10,000 9,000 8,000 8,000 8,750 7,500 7,000 6,750 6,750 6,000
Source: Bloomberg, Trade Data, Metal Bulletin, Macquarie Research, January 2017
19 January 2017 25
Agricultural Executive Summary

Waiting for a weather kick-start
Compared to their volatile history, global agricultural prices have had an ongoing period of
benign pricing on average. Indeed, Macquaries proprietary MacPI agricultural commodity
index ended December 2016 at approximately the same level as December 2015, which in
turn matched May 2015. Lows have been tested but not broken, but it is fair to say many
agricultural commodities missed out on the 2016 rally their peers enjoyed. In particular,
grains prices continued to suffer, though vegetable oil prices gained steadily all through 2016
while sugar rose strongly before a strong late year correction. We struggle to be bearish from
current levels. However, even though the path of least resistance seems to be to the upside,
catalysts may take until H2 to emerge.
One reason for the lack of sustained moves is that we have now gone four years without a
significant weather issue for the major grains. And, given high global inventory levels, this is
something we would need for price upside. Concerns over a poor Brazilian soybean crop on
the shift to La Nia conditions are dissipating by the day, with this latest weather event set to
have eased by February. Meanwhile, any fears over the US corn crop are still months away.
As ever, there will be much focus on US plantings over the coming year. Farmer margins
have slightly improved over recent times, though they are far from good. It is fair to say
however that, just like commodity peers, the risk of mass bankruptcies (in either the US or
Latin America) is fading. We expect a small switch from corn to soybeans in US plantings,
which is slightly negative nitrogen demand. Meanwhile, given the heavy snowpack has led to
a wetter planting period, total area may be lost. Among the grains, we would see wheat as
having the greatest upside over the year given the dependence on Black Sea supply and
potential upward pressures on the Rouble from higher oil prices.
Chinas influence continues to grow in agricultural markets, with domestic grains production
falling and demand from an ever wealthier population rising. Were agricultural prices to rise,
impacting CPI, the likelihood of China being concerned about inflation rather than enjoying
happy reflation grows. Key to this will be hog margins, which are currently close to record
levels. Should these normalise through a hog price fall, inflation is less of a concern. Should
higher margins drive higher feed prices and a stronger pull on global markets, inflation
concerns may return. We expect growth in Chinas soybean imports at mid-single-digit
percentage rates over the coming years.
The impact of protectionist measures on international trade is difficult to call, particularly
should a US border adjustment tax be imposed. Generally, grains balance sheets are solved
by the US being the last exporter given their enhanced ability to store inventory, though we
could see meaningful restrictions on fertilisers. Overall, we expect limited tit-for-tat measures
around agricultural trade given all governments like to keep their population fed unless they
have a strong desire for social unrest.
Fig 53 The MacPI index has been trending sideways Fig 54 Chinas soybean imports are on a steady
since early 2015 upward trajectory, supporting global demand
MacPI index, 1997-2000=100 China's soybean imports

230
50
220
210 40
million tonnes
200
190 30
180
20
170
160 10
150
0
140
H1 2012
H2 2012
H1 2013
H2 2013
H1 2014
H1 2006
H2 2006
H1 2007
H2 2007
H1 2008
H2 2008
H1 2009
H2 2009
H1 2010
H2 2010
H1 2011
H2 2011
H2 2014
H1 2015
H2 2015
H1 2016
H2 2016
130
Jul-13
Jul-14
Jul-15
Jul-16
Jan-15
Jan-13
Jan-14
Jan-16
Brazil Argentina USA Other

Source: Bloomberg, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
19 January 2017 26
Fig 55 Fertiliser price forecasts past the nadir, but little fundamental reason for sustainable upside
Fertilisers - Quarterly & Annual Average Price Forecasts
2015 1Q16 2Q16 3Q16 4Q16 1Q17f 2Q17f 3Q17f 4Q17f 2018f 2019f 2020f 2021f LT
Ammonia FOB Black Sea - US $/t 385 263 273 204 197 285 250 250 230 225 230 230 245 230
Urea prilled FOB Black Sea - US $/t 273 208 198 184 207 235 230 220 220 220 220 230 240 230
st Potash FOB Vancouver - US $/t 303 282 262 220 215 225 230 230 220 230 240 250 250 280
Source: CRU, Macquarie Research, January 2017
Nitrogen Riding the energy price uplift, but structural concerns remain: In recent
months we have certainly had a turnaround in nitrogen prices arguably too strong of one.
2017 is unlikely to be a year of strong global nitrogen demand, while current pricing is
attracting marginal supply back into the market, most notably in Chinese urea. Thus the
current urea rally is unlikely to be sustained, particularly now that Chinese anthracite pricing is
following coal lower. Global ammonia prices have reacted strongly to higher gas prices, with
the notable exception of the US. Even given the combination of domestic production ramp-
ups and the threat of protectionism, the US import price looks too low compared to other
global benchmarks. Longer term, the challenges to industrial nitrogen from increased
efficiency of use and displacement by industrial by-product alternatives are growing more
pertinent.
Potash the market share battle is maturing: The potash market is one we have long held
structural concerns about, and we are still in the middle innings of the necessary market
adjustment. Quite justifiably for a market which is seeing a market share battle into the key
consumer, prices are trading at a level to cause pressure on supply. However, unlike others
(such as uranium) where spot prices are unsustainably low following a 2016 collapse, for
potash spot prices have been relatively stable post the mid-year contract settlement. We are
not bearish on potash pricing from todays levels, and it finally seems the key market players
have moved towards acceptance of the wider problems to be faced. However, any recovery
will be slow and will need a boost from wider agricultural markets, which may be some time
off yet.
Crude Palm Oil Tight inventory, strong vegetable oil: CPO prices strengthened in 4Q16
to end the year strongly at US$713/t (+36% YoY), driven by: (i) lower palm oil inventory.
Malaysias Dec16 palm oil inventory fell by 37% YoY to 1.66m T, the lowest for December
since 2010, due to weak production and robust export demand; (ii) soybean complex firmed
up in 4Q16 as uncertainties in South American crop prospects offset the impact of record US
crop; (iii) soybean oils value share remained stable at 40%; and (iv) Indonesia remains
committed to its biodiesel plan (solid renewal of biodiesel tender). Moving into 2017, we
forecast the CPO stock-to-usage ratio to remain tight at 12.5% (versus 5-year historical
average of 19%), as lower starting inventory and robust demand growth should more than
offset recovery in production.
19 January 2017 27
Dairy Executive Summary

Prices buoyed by supply squeeze
The dairy price recovery is well underway, with significant increases in prices over 2H16
following a down market since Jan-14. The recovery in prices has been supply side driven,
both through farmers moderating output due to weak farmgate returns, as well as adverse
weather conditions. Over 2016, prices of the four main dairy commodities were up around
40% and currently trading around long run average levels. While milk output is falling, there
remain some inventories in the market and planned sales out of intervention stocks to come.
Demand has recovered a little, with small increases in milk powder import volumes by China,
while foodservice and ready-to-consumer volumes have increased more materially. Latin
America has seen a little growth in dairy imports, while Middle East and Africa have been
impacted by the economic flow through of low oil prices which is starting to improve. Russias
embargos on imported agricultural products from a number of regions continues.
Dairy processors are passing the higher product prices through to farmers, and helping to
improve returns on farm. This improved revenues, and reasonable feed costs could see
farmers looking to increase output at the margin but herd sizes remain a little lower.
Fig 57 but a little ahead of that implied by feed

Fig 56 Prices back towards long-run averages... prices
US$/MT
6,000 US cents /Bu 1,200
US$/MT
6,000
2012: -8% 2013: +51% 2014: -48% 2015: -6% 2016: +41% 5,000 1,000
5,500
5,000 4,000 800
4,500
4,000 3,000 600
3,500 2,000 400

3,000
2,500 1,000 200
2,000
0 0
1,500
Nov-02
Nov-07
Nov-12
Nov-17
Sep-03
Jan-02
Jan-07
Sep-08
Sep-13
Sep-18
Jan-12
Jan-17
May-05
May-10
May-15
Mar-06
Mar-11
Mar-16
Jul-04
Jul-09
Jul-14
1,000
Apr-12
Apr-14
Apr-16
Oct-12
Apr-13
Oct-13
Oct-14
Apr-15
Oct-15
Jul-16
Oct-16
Jul-12
Jul-13
Jul-14
Jul-15
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
WMP WMP average

WMP SMP Cheese Butter WMP implied (fwd feed) 70% Corn, 30% Soy
Source: GDT, Macquarie Research, January 2017 Source: GDT, Bloomberg, Macquarie Research, January 2017
Change in milk production out of key export regions was negative over the second half of
2016, with an acceleration in the decline in recent months. European production is down
around 3.5% in November reflecting softer farmgate prices, while Australia (-11% in Oct), NZ
(-5% in Nov) and Argentina (-18% in Nov) are weak due to adverse weather conditions.
Chinese demand for WMP has increased, with the 12 month run rate up by 10% as at
November. There has been no change in Russian imports with the agricultural ban extended.
Fig 58 Supply side continues to deteriorate Fig 59 Chinese imports improving, Russia still low
US EU NZ '000 MT, Chinese WMP demand still

8.0% Aus Argentina Brazil 1,800 800 low but starting to track up.
rolling 12
GDT (RHS) Some substitution into more
months
1,600 700 finished products
6.0%
1,400 600 China and Russia
4.0% combined represented
1,200 500
around 20% of global
2.0% 400 dairy imports in 2014
1,000
800 300
0.0%
200 Russian cheese imports yet to
600 recover with bans remaining in
-2.0% place until at least Dec-17
100
400
-4.0% % change in 0
exporter Price index 200
Jan-09
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Sep-09
Sep-10
Sep-11
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
May-09
May-10
May-11
May-12
May-13
May-14
May-15
May-16
supply
-6.0% 0
Oct-09
Oct-10
Oct-11
Oct-12
Oct-13
Oct-14
Oct-15
Oct-16
Apr-09
Apr-10
Apr-11
Apr-12
Apr-13
Apr-14
Apr-15
Apr-16
Jan-09
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Jul-09
Jul-10
Jul-11
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
China - WMP imports Russia - Cheese imports
Source: GDT, Eurostat, USDA, DairyNZ, Dairy Australia, January 2017 Source: GTIS, Macquarie Research, January 2017
19 January 2017 28
Supply side contraction continues

The majority of key dairy exporting regions are seeing a pullback in milk production due to the
impact of lower on-farm returns as well as adverse weather events. The EU saw milk
production grow at a run rate of 4-6% following the removal of quotas, but output has now
slowed, and growth turned negative in the last six months (November -3.5% pcp). Conditions
in Australia have been challenging, with flooding driving a significant reduction in production
(Oct -11%). NZ has seen similar challenges, with the season expected to be down around 5-
6%. The US continues to grow 1-3%, but with the majority of product being consumer
domestically hence having limited impact on trade flows.
We estimate that around half of the decline in production is linked to adverse weather events
across Australia, New Zealand, Argentina and Brazil, which should normalise over time. Herd
sizes have also declined following a period of weak raw milk prices and farm profitability,
which creates a more medium term headwind to production growth.
EU production tracking lower
Milk production in the EU is now responding strongly to the low farmgate price signals, and
the latest data showed declines of around 3.5% in November. This is now more realistically
reflecting the historical relationship between the underlying changes in both milk price and
output. Production has declined in the majority of regions including the UK (-7.2% Nov on
pcp), France (-7.5%), and Germany (-5.2%) while Netherlands (+0.1%), Ireland (+0.5%) and
a few other smaller producing regions are still seeing flat or marginal growth.
The decline in output has been facilitated by the EU milk production reduction scheme, with
the initial tranche being 98.9% subscribed representing around 1.06m tonnes of milk to not be
produced (between Oct-16 and Mar-17). In terms of production, this represents an average
reduction of 16.5% by applicant, and 2.9% of EU quarterly production. In exchange for the
reduction, farmers receive around EUR14cents/litre from the European Commission. In
addition to this scheme, Friesland Campina had been running a similar structure with its own
farmers for non-production for six months from October 2016.
The reduced output has had a positive impact on commodity prices. This uplift is quickly
translating into higher farmgate prices. The average EU milk price is up around 25% in the
last six months since the lows in July, and processor data indicates further strength in
January (Friesland guaranteed milk price EUR34.5 in Jan). Rising on-farm revenues, coupled
with low feed and energy costs could see farmers looking to expand output, however, recent
data shows some herd reduction, including Germany down 1.6% on pcp. Further,
Netherlands faces challenges near term, with around 200,000 cows expected to be culled in
order for Dutch authorities to avoid losing their nitrate derogations due to breach of phosphate
limits for the last three years.
Fig 60 EU farmgate milk prices seeing strong

rebound Fig 61 Output showing lagged response to low FMP
EUR/100kg 40% 8%
50
30% 6%
45
20%
4%
40
10%
2%
35 0%
0%
30 -10%
-2%
25 -20%
-30% -4%
20
Nov-09
Oct-05
Dec-06
Oct-12
Dec-13
Aug-04
Apr-09
Sep-08
Aug-11
Sep-15
Apr-16
Jan-04
Jun-10
Jan-11
May-06
May-13
Mar-05
Feb-08
Mar-12
Feb-15
Jul-07
Jul-14
-40% -6%
Jan-16
Jan-06
Jan-07
Jan-08
Jan-09
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jul-06
Jul-07
Jul-08
Jul-09
Jul-10
Jul-11
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
Germany France Ireland

Netherlands Weighted avg EU Price change (m/pcp, LHS) Production growth (m/pcp, 4m lag, RHS)
Source: European Commission, Macquarie Research, January 2017 Source: EC, Eurostat, Macquarie Research, January 2017
19 January 2017 29
US milk production continues to track up

Milk production growth over 2016 has been about 1.8% for the US, a slight acceleration 2015
seeing 1.3% growth. The growth has been running a little faster in recent months, up 2.5% in
October and November on pcp. This has been helped through a stabilisation of production in
California after significant declines from drought conditions, and continued expansion out of
Wisconsin and other smaller states. The overall increase has been driven largely by
productivity gains, with total herd size up only 0.2% on pcp.
While milk prices have been weak over 2016, there has been solid rebound in the last six
months, with the November price up around 18% on June. The higher farmgate prices and
steady input costs have seen milk margins expand and stay well clear of support levels. The
average milk price of US$17.6/cwt remains above the estimated nationwide variable cost of
production of ~US$13-14/cwt, so marginal production remains profitable. When including
fixed costs/overheads, total costs are closer to US$21-22/cwt meaning new operations are
unlikely to be established with current pricing. The USDA expects growth of around 2.1% in
2017 compared to 2016 levels, or an additional 2b litres of milk.
The strong demand for fat-based products, including butter and cheese, has meant the
increased production has not translated into higher export volumes of such products (rolling
12 months), while milk powders and whey have increased in export volume.
Fig 62 US growth accelerated to 2.5% in November Fig 63 Milk margins trending up and are healthy
US$/cwt
100% 10% 4.0x 18
80% 8% 3.5x 16
3.0x 14
60% 6%
12
2.5x
40% 4%
10
2.0x
20% 2% 8
1.5x
0% 0% 6
1.0x 4
-20% -2%
0.5x 2
-40% -4%
0.0x 0
Oct-00
Oct-03
Oct-06
Oct-09
Oct-12
Oct-15
Apr-02
Apr-05
Apr-08
Apr-11
Apr-14
Jan-00
Jan-03
Jan-06
Jan-09
Jan-12
Jan-15
Jul-01
Jul-04
Jul-07
Jul-10
Jul-13
Jul-16
-60% -6%
Oct-07
Dec-08
Apr-11
Nov-11
Oct-14
Dec-15
Aug-06
Sep-10
Jan-13
Aug-13
Jan-06
May-08
Jun-12
May-15
Mar-07
Jul-09
Feb-10
Mar-14
Jul-16
Price change (m/pcp, LHS) Production growth (m/pcp, 6m lag, RHS) Milk/feed Margin: milk less feed (RHS)
Source: USDA, Macquarie Research, January 2017 Source: USDA, Macquarie Research, January 2017
NZ production impacted by unfavourable weather over peak period

The 2016/17 milk production season has been a little softer to-date, with declines of around
3% against pcp, and the previous season being down around 2%. Fonterras expectation
remains for a 6-7% decline in its NZ milk collection following a weak peak production period.
This was caused by unfavourable weather conditions across most dairying regions, and the
lower peak output is expected to flow through to the remainder of the season. The most
recent data point for November indicates around a 4.5% decline for NZ output.
Cow and heifer cull rates have started to slow a little, and off elevated levels in the previous
year. This is likely reflective of the improvement in farmgate milk prices and outlook for farm
profitability. The NZ dairy herd has declined for the first time in ten years, with total number of
cows falling by 0.4%, compared to the 10-year growth rate of 2.6% p.a. Given the weak
profitability for the previous two seasons, and associated increase in on-farm debt, it is likely
that farmers will be looking to repay this as profitability improves as opposed to expanding
herd materially in coming seasons.
Fonterra has revised the forecast 2016/17 farmgate milk price up three times this season,
starting from $4.25/kgMS opening in May 2016, to $6.00/kgMS at its November update. The
market would expect further upside if commodity prices hold around current levels, with spot
commodity and FX implying a number closer to $6.50-7.00/kgMS. The improvement in milk
prices could see a little extra supplementary feed used to boost output or mitigate any further
adverse weather impacts.
19 January 2017 30
Fig 64 Low payout and weather dragging on 2016/17 Fig 65 Cull rates falling from elevated levels
14% Change pcp Rolling 12m

40%
change pcp
12%
30%
10%
8% 20%
6% 10%
4%
0%
2%
0% -10%
-2% -20%
-4%
-30%
-6%
Jan-11
Jan-00
Jan-01
Jan-02
Jan-03
Jan-04
Jan-05
Jan-06
Jan-07
Jan-08
Jan-09
Jan-10
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
2000/01
2001/02
2002/03
2003/04
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
2010/11
2011/12
2012/13
2013/14
2014/15
2015/16
2016/17F
Heifer cull rate Cow cull rate Total (cow + heifer)
Growth Average
Source: DairyNZ, Fonterra, Macquarie Research, January 2017 Source: Stats NZ, Macquarie Research, January 2017
Dairy industry still tough in Australia

Dairy farming conditions in Australia remain challenging following farmgate milk price cuts in
the 2015/16 season. Total production at present is down around 11% across Australia,
compared to a 2% decline in the previous season. The climatic conditions swung from too dry
in the previous season to too wet, with flooding being an issue in the current season. The
weather appears to have improved recently and moved into summer with good soil moisture.
The weak end to the previous season in terms of farmgate milk price, with an effective step
down required by the two largest processors, negatively impacted farmer confidence in the
industry and led to accelerated culling and farmer retirements. This has contributed towards
further declines in production during the 2016/17 season, worsened further by unfavourable
weather reducing pasture growth and milk yields.
While commodity prices have moved up materially lately, the Australian FMP is yet to be
revised upwards to the same extent. Murray Goulburn in October reduced its FMP forecast to
A$4.70/kgMS (from $4.88) due to falling intake, while Fonterra increased its full year forecast
to A$5.20/kgMS on improving revenues. We note that Australias product mix sees a
significantly lower exposure to commodity volumes (against NZ), and the export volumes
skewed towards SMP which has been the weaker performer.
While the FMP remains tough (historic breakeven around A$5.50/kgMS), feed prices are
favourable with a strong grain and hay season leading to increased supply and lower prices.
Irrigation and water storage also improved over the year, meaning better availability should
dry conditions arise over 2017. Given the increased debt on farm from a weak season and
recoupment structures from a few co-ops, investment in farm assets is likely to be subdued.
Fig 66 Farmer exit and culling driving lower output Fig 67 FMP still weak
A$/kgMS
8.0%
New Season
8.0 stepped and

6.0% final price
7.5
4.0%
7.0
2.0% 6.5
0.0% 6.0
-2.0% 5.5
-4.0% 5.0
-6.0% 4.5
-8.0% 4.0
-10.0% 3.5
Vic Aus 3.0
-12.0%
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jul-09
Jul-10
Jul-11
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
Aug-15
Sep-15
Oct-15
Nov-15
Dec-15
Jan-16
Apr-16
Aug-16
Sep-16
Oct-16
Jun-16
May-16
Mar-16
Jul-15
Feb-16
Jul-16
Source: Dairy Australia, Macquarie Research, January 2017 Source: Company data, Macquarie Research, January 2017
19 January 2017 31
Argentinian dairy industry facing challenges

The dairy industry in Argentina has seen production decline by around 11% in the last 12
months, and the monthly decline in November accelerated to -18%. There are a number of
factors impacting output including low farm profitability, further exacerbated by flooding in a
number of provinces.
Farmgate prices in Argentina have been increasing in local currency terms over the last 12
months, however, the continued devaluation of the Argentine Peso has caused high inflation
and increased production costs. Accordingly, producers have shifted some production into
less intensive feeding systems to reduce costs, but this is at the expense of milk yields per
cow. The lower production has translated into declining volume of exports by Argentina.
The flood damage and weak financial performance of the sector has seen increased farmer
exits and a reduction in herd size (USDA estimates -5%). It would be expected that
production increases a little over 2017 through the normalisation of weather patterns, as well
as recovering domestic prices. The tough season, likely the worst in 20 years, may
discourage further investment in the industry and creates challenges around competitiveness.
Fig 68 Milk production declining materially Fig 69 Domestic raw milk prices tracking up
Change pcp
20% 5.00 40.00
15% 4.50 35.00

4.00
10% 30.00
3.50
5% 3.00 25.00
0% 2.50 20.00
2.00 15.00
-5%
1.50
-10% 10.00
1.00
-15% 0.50 5.00
-20% 0.00 0.00

Sep-10
Sep-11
Jan-12
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
Jan-10
Jan-11
Jan-13
Jan-14
Jan-15
Jan-16
May-10
May-11
May-12
May-13
May-14
May-15
May-16
-25%
Apr-11
Oct-11
Apr-12
Oct-12
Apr-13
Oct-13
Apr-14
Oct-14
Apr-15
Oct-15
Apr-16
Oct-16
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jul-11
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
ARS$/lt EUR/100lt (RHS)
Source: CLAL, Macquarie Research, January 2017 Source: CLAL, Macquarie Research, January 2017
Milk collection growth for key exporters slowing to 1%

Below is a summary of the production and export statistics for the key exporters. Milk
production is slowing with low farmgate prices flowing through to output and adverse weather
impacting a number of regions. Export volumes have been mixed, but overall down for milk
powders, and up for butter and cheese.
Fig 70 Milk flow solid for last 12 months driving more commodity production and exports
Milk collection, production and exports (12m to November 2016)
EU US NZ Australia Key exporters
'000 % chg pcp '000 % chg pcp '000 % chg pcp '000 % chg pcp '000 % chg pcp
Tonnes Tonnes Tonnes Tonnes Tonnes
Milk collection 152,752 1.1% 96,194 1.8% 21,250 -1.5% 9,165 -4.7% 279,361 1.0%
Production - WMP 670 5.1% 44 -6.3% N/A N/A 59 -38.8% 2,229 -2.3%
Production - SMP 1,580 6.1% 1,039 0.5% N/A N/A 245 -3.4% 3,410 6.2%
Production - Cheese 8,989 2.0% 5,470 3.3% N/A N/A 323 -1.7% 15,141 2.5%
Production - Butter 2,166 3.9% 862 3.4% N/A N/A 78 -6.0% 3,524 3.9%
Exports - WMP 372 1.4% 53 57.1% 1,343 -3.0% 67 3.1% 1,835 -0.9%
Exports - SMP 582 -14.7% 566 3.5% 519 26.0% 169 -16.7% 1,836 -0.4%
Exports - Cheese 802 12.5% 287 -10.0% 351 9.0% 169 1.5% 1,609 5.8%
Exports - Butter 170 28.8% 17 -5.6% 293 6.9% 21 -22.1% 501 11.1%
Note: production data of individual commodities is not available for New Zealand but the majority of production is exported.
Source: Eurostat, USDA, DairyNZ, Dairy Australia, GTIS, Macquarie Research, January 2017
19 January 2017 32
Key importer demand growth modest

Import demand from China has improved slightly after the significant declines over 2014-15,
while Russia remains largely out of the market due to agricultural bans being extended.
Fig 71 Movements in dairy import demand by key regions
300 '000MT change

200
100
-100
-200
-300
Sep-09
Sep-10
Sep-11
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
Jun-09
Dec-09
Jun-10
Dec-10
Jun-11
Dec-11
Jun-12
Dec-12
Jun-13
Dec-13
Jun-14
Dec-14
Jun-15
Dec-15
Jun-16
Mar-09
Mar-10
Mar-11
Mar-12
Mar-13
Mar-14
Mar-15
Mar-16
China Russia SE Asia Africa Middle East Latin America ROTW
Note: data derived from exports of key regions to import destinations. Products include WMP, SMP, Butter,
Cheese and Fluid Milk.
Source: GTIS, Macquarie Research, January 2017
Chinese imports seeing a little recovery

It appears that the stock that was accumulated over 2014/15 has been largely worked
through in the market given the run up to expiry, and demand for milk powder imports has
started to rebound from lows. While this has underpinned improved milk powder prices,
demand continues to shift into more finished products such as UHT (+52% 12 months to Nov)
and more Western products including cheese (+20%) and butter (+32%).
Fig 72 China milk powder imports up a little Fig 73 Raw milk prices lower than imported WMP
MT milk RMB/kg
180,000 powder 6.0
imports
160,000
5.0
140,000
4.0
120,000
100,000 3.0
80,000 2.0
60,000
1.0
40,000
20,000 -
Apr-10
Apr-15
Nov-09
Dec-11
Oct-12
Nov-14
Dec-16
Aug-08
Sep-10
Aug-13
Sep-15
Jan-09
Jun-09
Jan-14
Jun-14
May-12
Feb-11
Jul-11
Mar-13
Feb-16
Jul-16
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2014 2015 2016 Average raw milk price WMP price equilibriated
Source: GTIS, Macquarie Research, January 2017 Source: MOA, GDT, Macquarie Research, January 2017
Corn prices in China have softened further over the last few months which should continue to
lower dairy farm input costs and improve margins for domestic producers (negative for
imports). With the recent rebound in WMP prices, the cost of importing milk powder (adjusting
for tariffs, VAT and transportation) is now higher than the raw milk price in China. This could
see some substitution away from imported milk if there is sufficiently high quality raw milk
available, and over time, should drive higher raw milk prices in China. Over the last few
months, Chinese raw milk prices have risen by around 4%.
Recent liquid milk production data (i.e. downstream consumer products such as UHT) has
shown increased output by 8% in October and 13% in November which should imply strong
consumption trends. This could also indicate the utilisation of milk powder inventory as this
growth should be running ahead of domestic raw milk production (USDA est. -5% in 2016).
19 January 2017 33
The weak raw milk prices in China have seen the exit of smaller players as well as lower
growth (and culling) of some larger dairy herds. The USDA estimate that herd size in China
could have reduced by as much as 10% over 2016. With improving milk margins as prices
recover, this could see addition feeding to improve milk yields.
Fig 74 Corn prices track lower in China Fig 75 Downstream liquid milk volume still growing
RMB/MT
2,800 50%
2,600 40%
2,400
30%
2,200
20%
2,000
10%
1,800
1,600 0%
1,400 -10%
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Sep-11
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
May-11
May-12
May-13
May-14
May-15
May-16
Sep-09
Sep-10
Sep-11
Sep-12
Sep-13
Sep-14
Sep-15
Sep-16
Jan-09
May-09
Jan-10
May-10
Jan-11
May-11
Jan-12
May-12
Jan-13
May-13
Jan-14
May-14
Jan-15
May-15
Jan-16
May-16
Source: Bloomberg, Macquarie Research, January 2017 Source: CLAL, Macquarie Research, January 2017
Russian sanctions mean no change to import situation

With the ban on agricultural products from a number of regions remaining in place until at
least 31 December 2017, there is limited change to the import demand situation from Russia.
Prior to import bans, Russia accounted for around 18% of the butter exports and 22% of
cheese exports of the top 5 exporters. The proportion from the EU was even greater, with
Russia previously taking around 23% of butter exports and 30% of cheese. Since the bans,
export volumes to Russia have roughly halved, with EU exports heading to zero but offset by
slightly higher volumes from Belarus. Russia has been establishing partnerships with other
nations such as Vietnam, Thailand and China, to help deliver the products it requires and
removing reliance on other milk producers. It is likely a large amount of the export volume
redirection in place from embargoed regions will persist longer term.
Production of raw milk in Russia and Belarus has been largely flat and Russian milk
production is expected to hit all-time lows next year through further declines in the dairy herd
(USDA). The decline in herd is attributable to low investment over the last few years, but is
offset by improving yields through better genetics and herd management. Consumption is
also expected to fall due to high pricing and substitution into other products, especially when
good-quality domestically produced dairy products cannot be found.
Fig 76 Exports of cheese to Russia still impacted Fig 77 Milk production in Russia and Belarus flat
'000 MT Monthly
50 6% change vs.
Cheese
45 exports to pcp
40 Russia 4%
35
30 2%
25
20 0%
15
10 -2%
5 Russia Belarus
-4%
0
Jan-13
Jan-14
Jan-15
Jan-16
Nov-12
Sep-13
Nov-13
Sep-14
Nov-14
Sep-15
Nov-15
Sep-16
May-13
May-14
May-15
May-16
Mar-13
Mar-14
Mar-15
Mar-16
Jul-13
Jul-14
Jul-15
Jul-16
-6%
Sep-13
Nov-13
Sep-14
Nov-14
Sep-15
Nov-15
Sep-16
Nov-16
Jan-13
May-13
Jan-14
May-14
Jan-15
May-15
Jan-16
May-16
Jul-15
Mar-13
Jul-13
Mar-14
Jul-14
Mar-15
Mar-16
Jul-16
Argentina Australia New Zealand US Belarus EU Others
Source: CLAL, Macquarie Research, January 2017 Source: CLAL, Macquarie Research, January 2017
19 January 2017 34
Fig 78 Global WMP pricing Fig 79 Global SMP pricing
WMP US$/MT SMP US$/MT

6,000 6,000
5,000 5,000
4,000 4,000
3,000 3,000
2,000 2,000
1,000 1,000
0 0
Aug-04
Sep-08
Aug-11
Sep-15
Mar-05
Dec-06
Mar-12
Dec-13
Aug-04
Oct-05
Sep-08
Apr-09
Nov-09
Aug-11
Oct-12
Sep-15
Apr-16
Nov-16
Oct-05
Dec-06
Apr-09
Nov-09
Oct-12
Dec-13
Apr-16
Nov-16
Jan-11
Jan-04
May-06
Jun-10
May-13
Jun-10
Jan-04
May-06
Jan-11
May-13
Jul-07
Feb-08
Jul-14
Feb-15
Jul-14
Mar-05
Jul-07
Feb-08
Mar-12
Feb-15
US EU Oceania US EU Oceania
Source: USDA, EC, GDT, Macquarie Research, January 2017 Source: USDA, EC, GDT, Macquarie Research, January 2017
Fig 80 Global Cheese pricing Fig 81 Global Butter pricing
Cheddar Butter
7,000 US$/MT 7,000 US$/MT
6,000 6,000
5,000 5,000
4,000 4,000
3,000 3,000
2,000 2,000
1,000 1,000
0 0
Mar-05
Mar-12
Mar-05
Mar-12
Aug-04
Oct-05
Dec-06
Sep-08
Apr-09
Nov-09
Aug-11
Oct-12
Dec-13
Sep-15
Apr-16
Nov-16
Aug-04
Oct-05
Dec-06
Sep-08
Apr-09
Nov-09
Aug-11
Oct-12
Dec-13
Sep-15
Apr-16
Nov-16
Jun-10
Jan-04
May-06
Jun-10
Jan-11
May-13
Jan-04
May-06
Jan-11
May-13
Jul-07
Feb-08
Jul-14
Feb-15
Jul-07
Jul-14
Feb-08
Feb-15
US EU Oceania US EU Oceania
Source: USDA, EC, GDT, Macquarie Research, January 2017 Source: USDA, EC, GDT, Macquarie Research, January 2017
Fig 82 Global Whey pricing Fig 83 Global raw milk pricing (standardised)
Whey US$/MT US$/100kg

2,000 60
1,800 55
1,600 50
1,400 45
1,200 40
1,000 35
800 30
600 25
400 20
200 15
0 10
Aug-04
Sep-08
Aug-11
Sep-15
Oct-05
Dec-06
Apr-09
Nov-09
Oct-12
Dec-13
Apr-16
Nov-16
Jun-10
Jan-04
May-06
Jan-11
May-13
Jul-07
Jul-14
Mar-05
Feb-08
Mar-12
Feb-15
Oct-00
Apr-02
Oct-03
Apr-05
Oct-06
Apr-08
Oct-09
Apr-11
Oct-12
Apr-14
Oct-15
Jan-03
Jan-00
Jan-06
Jan-09
Jan-12
Jan-15
Jul-01
Jul-04
Jul-07
Jul-10
Jul-13
Jul-16
US EU EU US NZ Aus
Source: USDA, EC, GDT, Macquarie Research, January 2017 Source: USDA, EC, Fonterra, Macquarie Research, January 2017
19 January 2017 35
Copper
A corner turned
Copper abruptly shifted from an ever-decreasing trading range in the year to October to
massive volatility in Q4, after bullish signals emanating from LME Week suddenly turned into
a buying frenzy. The huge lurch upwards was eventually tempered, with two separate
exploratory stabs above $6,000/t being firmly sold, and prices settling in the mid $5,000s/t
which is where we find ourselves today. We think the key catalyst was the collective
acknowledgement in London of copper having been left behind by the price rises which had
been seen in other industrial commodities (oil, iron ore, coal and zinc most notably). Chinese
shorts had been turning neutral through 2016, and this accelerated in the days after LME
Week meanwhile enough shorts were still sitting there with option structures putting them
deeper in the red the more the price rallied. As they capitulated one by one, copper broke
new ground, and later frightened off sellers as it held through Asia CESCO Week in Shanghai
(the week following the rally).
Chinese demand strength was a key factor as in all commodities 2016 surprised to the
upside, and with a political transition year ahead we see little chance of a 2015-style slump in
demand. Consumer products have notably rebounded since a product destocking left growth
in negative territory earlier in the year, while power-driven consumption remains in positive
territory, if slowing due to base effects lately.
Fig 85 Demand improvement story holds, with

Fig 84 Copper price makes a move upwards! consumer now looking up
LME copper cash price ($/t) China grid investment, consumer product
output & construction starts YoY % MAs
10,000
50% Grid investment, 3mma YoY
9,000 Refrigerator output, 3mma YoY
40% AC generator output, 3mma YoY
AC output, 3mma YoY
8,000 30% Construction completions, 5mma YoY
20%
7,000
10%
6,000
0%
5,000 -10%
-20%
4,000
Jan 11 Jan 12 Jan 13 Jan 14 Jan 15 Jan 16 Jan 17 -30%
LME Copper Cash Jan-13 Jul-13 Jan-14 Jul-14 Jan-15 Jul-15 Jan-16 Jul-16
Source: LME, Macquarie Research, January 2017 Source: NBS, CEIC, Macquarie Research, January 2017
Fig 86 Global copper market balance (000 tonnes copper contained)

'000t copper 2013 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
Mine production 17,921 18,249 19,059 19,691 19,336 20,148 20,477 20,604 20,398
% Change YoY 7.6% 1.8% 4.4% 3.3% -1.8% 4.2% 1.6% 0.6% -1.0%
Refined production 21,120 21,807 22,339 22,495 22,914 23,694 24,062 24,239 24,202
% Change YoY 5.0% 3.3% 2.4% 0.7% 1.9% 3.4% 1.6% 0.7% -0.2%
Consumption 20,991 21,723 21,989 22,242 22,814 23,350 23,875 24,411 24,868
% Change YoY 5.6% 3.5% 1.2% 1.2% 2.6% 2.4% 2.3% 2.2% 1.9%
Refined balance 128 84 350 253 100 344 187 -171 -666
SRB/bonded stocking -50 309 140 150
Adjusted balance 178 -225 210 103 100 344 187 -171 -666
LME Cash ($/t) 7,322 6,862 5,503 4,863 5,350 5,100 5,188 5,725 6,250
Stocks (Weeks) 6.0 7.3 8.6 9.1 9.1 9.6 9.8 9.2 7.7
Source: LME, Comex, SHFE, ICSG, CRU, Wood Mackenzie, Macquarie Research, January 2017
19 January 2017 36
Meanwhile, the supply-side lost its invulnerable H1 glow and reverted to the shaky, flaky
copper mining sector we all know and love. While new projects Las Bambas and Cerro Verde
continued to impress, a number of older assets began to look weaker including Escondida
and Grasberg (together ~10% of global output), with the world #1 mine in particular seeing
disappointing output in H2. Q4 data is yet to be released, but a glance at Chilean government
data in October implies that the problems have continued at the desert behemoth.
This shift began to tighten the concentrates market in H2 at exactly the wrong moment for the
Chinese smelters, who ended up having to settle for almost $5/5c lower TCRC benchmark of
$92.5/9.25c with their suppliers. Meanwhile on the metal side, the opposite took place, with
sudden LME warehouse inflows in mid-December increasing total inventories by 62% to
345kt in just over a week. This softer refined market complements our expectation of a
surplus in Q1, and prices to fall from the current levels to average ~$5,350/t.
In the rest of the year we see a tighter balance, and some recovery in prices. We note that
more optimistic sentiment from Chinese speculators has continued into 2017 and steady
demand in 2017 is likely to feed into this atmosphere beyond Chinese New Year. However,
our numbers (mainly slowing demand growth) indicate a softer balance of >300kt of excess
output in 2018, which should act to reduce prices to average ~$5,100/t. 2019 also represents
a surplus year, though by a smaller amount, before we expect to see the market tipping into a
deep deficit by 2021 on a lack of new production to offset declines, most particularly in Chile.
Fig 87 Escondida begins to struggle Fig 88 LME stocks came up quickly in mid-Dec
Chile and Escondida combined mined copper LME copper stocks (kt)
output monthly (kt) 400
160 550
140 350
500
120 300
100 450 250
80
200
60 400
150
40 Escondida 350
100
20 Total Chile (RHS)
0 300 50
Jan-10
Jan-08
Jan-09
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jul-08
Jul-09
Jul-10
Jul-11
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
-
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
16 16 16 16 16 16 16 16 16 16 16 16 17
Source: Cochilco, Macquarie Research January 2016 Source: LME, Macquarie Research, January 2017
Fig 89 Falling output at big mining countries begins Fig 90 Price turnaround seen holding, but 2018
to take its toll at the end of the decade surplus to dent prices before the deficit opens later
Copper mine output*, Chile and Peru (kt) Refined copper supply/demand (kt, LHS) vs
2000
400 LME cash price ($/t, RHS) 8,500
8,000
200
1000 7,500
0 7,000
6,500
0 -200
6,000
-400 5,500
-1000 5,000
2011 2013 2015 2017f 2019f 2021f -600
4,500
Chile Peru China
USA Congo DR Australia -800 4,000
Indonesia Zambia Mexico 2013 2015 2017F 2019F 2021F
Refined Balance New LME Cash Price ($/t)
Source: *before disruption, WM, CRU, ICSG, Macquarie Research, Source: LME, CRU, WM, ICSG, Macquarie Research, January 2017
January 2017
19 January 2017 37
Aluminium
Stubbornly stable pricing
Aluminium had a strong end to 2016, ending up 17% on the year, and has shown further
strength in the early days of 2017 with prices now staking territory above $1,800/t (LME cash
basis). This came despite our concerns that rising levels of Chinese production from a raft of
restarts and new capacity commissioning in the second half of last year would weigh on the
market, and ingot in particular would start to soften. The production (Fig. 86) did rise
according to various trade media reports of the plants firing up, and the IAIs own reckoning of
output levels in the last few months but the stocks did not show material gains (Fig. 87).
Ingot tightness still seems to be a factor in China, though slightly higher stocks and much
lower premiums show that the situation has eased since the worst moments in late
September, at which point premiums had blown out to >600RMB/t. This was partly driven by
the switch to melt rather than ingot, and partly by the abrupt crunch in truck availability due to
a crackdown on overloading, which created a bottleneck situation for Xinjiang material
intended for eastern and southern consumer centres. Trucking has become less tight since
then, but into the Chinese New Year period the transit of people, food and fuel take priority,
meaning aluminium stock is still being delayed out of the far West, and allowing premiums to
remain positive for now. That said, premiums are only just above flat, and we note that SHFE
is not leading LME in the recent rally, implying weakening Chinese price support.
Fig 91 Aluminium climbs back to $1,800/t Fig 92 ... as Chinese output growth slowed sharply
LME aluminium cash price ($/t) Aluminium output (LHS, Mtpa) vs LME cash
2200 40 (RHS, $/t) 3,500
China 2016:
2100 35 Low prices, 3,000
ex-China smelter cuts
2000 30
Lme price 2,500
1900 25
2,000
1800 20
1,500
1700 15
2010: 1,000
1600 10 5yr plan
2009: power 500
1500
5 GFC efficiency
cuts
0 -
1400
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Jan Jun Nov Apr Sep Feb Jul Dec May Oct
2013 2013 2013 2014 2014 2015 2015 2015 2016 2016
Source: LME, Macquarie Research, January 2017 Source: CRU, IAI, Macquarie Research, January 2017
Fig 93 Chinese aluminium inventories have lifted only

slightly since the truck trouble Fig 94 US and Japan looking a little tighter lately
China Al ingot and billet stocks (LHS, kt) vs SHFE Aluminium premiums ($/t)
active month price (RHS, $/t) 350
1,000 14,500
900 14,000
300
800 13,500
700 13,000 250
600
12,500
500 200
12,000
400
11,500 150
300
200 11,000
100
100 10,500
0 10,000 50
0
2011 2012 2013 2014 2015 2016
SHFE Stock Non-SHFE stock
Japan CIF US M.West del. NW Eur DDP
Aluminium billet inventory SHFE price
Source: SHFE, SMM, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
19 January 2017 38
Fig 95 US needs net imports of ~5Mtpa Al units Fig 96 We still see global aluminium oversupply
what happens if the Border Adjustment Tax comes in? arriving from 2019, but tighter 2017/2018
US primary aluminium supply/demand/net import Global Al SD balance (LHS, Mt) vs price forecast
requirements (Mt) 3.00 (RHS, $/t) 2,200
8
2.50
6 2,000
2.00
4 1,800
1.50
2
1.00 1,600
0
0.50
1,400
-2 0.00
1,200
-4 -0.50
-6 -1.00 1,000
2011 2012 2013 2014 2015 2016E 2017F 2018F 2019F 2020F 2021F 2012 2013 2014 2015 2016E2017F 2018F 2019F 2020F 2021F
Production Primary consumption Required Al unit net imports Global balance LME Cash Price ($/t)
Source: CRU, Macquarie Research, January 2017 Source: CRU, IAI, Macquarie Research, January 2017
Meanwhile elsewhere premiums have lifted, with Q1 MJP $20/t higher at $95/t and US mid-
west now over 8c/lb, though Europe softened slightly. In Japan we note that stocks have
fallen at the ports, while in the US there is some interest in the ramifications for aluminium of
the so called Border Adjustment tax. Or more accurately, the effects on the CME all-in price,
which includes a premium and would therefore capture the rally in physical market premiums
that would naturally follow should the tax be enacted and begin to block the entry of the
~4.8Mt of aluminium units required to fill the US deficit (Fig 89).
Regional dislocations aside such as the possibility above, we see the global aluminium
market in a slight surplus in 2017, with the continued Chinese and Indian capacity build-out
meeting increasing reluctance to cede further ground from Western smelters (see the
attempts to rescue Alcoas Portland smelter in Australia) and this together with a softening
Chinese physical landscape means we see prices falling to ~$1,600/t for the year.
Further out, the question as ever is the extent of Chinas capacity build-out, and how much
older capacity will be displaced by these new gleaming furnaces. Recent media reports have
begun to discuss a supposed plan at the national level to order closures of aluminium
capacity on environmental and supply-side reform grounds, in much the same way as has
taken place in steel and coal, though we have discovered no hard evidence or detail for these
claims. Nevertheless we have raised our disruption expectations for China from 5% to 8%
from 2018 on the view that more old capacity will be removed, with the result that our growth
rates have slowed to a CAGR of 4.1% to 2021. This still eclipses the 2.5% consumption
growth expected, and so we continue to see the market sliding into oversupply, though now
from 2019, and prices to struggle.
Fig 97 Global aluminium market balance (million tonnes aluminium)

m illion tonnes 2013 2014 2015 2016E 2017F 2018F 2019F 2020F 2021F
World Production 50.6 54.2 57.1 58.8 60.8 62.9 65.4 67.7 67.7
% Change YoY 5.4% 7.0% 5.5% 2.9% 3.4% 3.4% 4.0% 3.4% 0.1%
of w hich: China 24.9 28.3 30.8 32.0 34.4 35.6 37.7 39.0 39.1
World Consum ption 50.8 54.0 56.5 59.4 60.8 62.8 64.0 65.2 66.6
% Change YoY 6.9% 6.4% 4.7% 5.1% 2.3% 3.3% 1.9% 2.0% 2.1%
of w hich: China 22.7 25.8 27.4 29.5 30.2 31.6 32.0 32.6 33.3
Global balance -0.16 0.16 0.61 -0.61 0.02 0.11 1.47 2.47 1.17
Estim ated total stocks 12.5 12.6 13.2 12.6 12.6 12.8 14.2 16.7 17.9
Weeks of consumption 12.8 12.2 12.2 11.1 10.8 10.6 11.6 13.3 14.0
LME Cash Price ($/t) 1,845 1,867 1,663 1,604 1,600 1,494 1,500 1,500 1,550
LME Cash Price (c/lb) 83.7 84.7 75.4 72.8 72.6 67.8 68.0 68.0 70.3
Source: CRU, IAI, LME, Wood Mackenzie, Macquarie Research, January 2017
19 January 2017 39
Zinc
Higher, faster shorter?
In our last compendium in September, zinc had rallied 45% YTD to just over $2,300/t. It has
hardly wasted time since then rallying again to over $2,700/t at the time of writing, having
flirted with $3,000/t already back in December. The long standing cap of $2,400/t has thus
been categorically breached, and positive Chinese speculator momentum into 2017 has
begun to carry the price upwards again, after a somewhat heady stab at $3,000/t was
rejected late in 2016.
Though critical to the metals outperformance, it is not just the familiar story of mine cuts that
has assisted zinc: Galvanised steel output showed a strong recovery in output through 2016,
correlating with rising prices of both steel and zinc. China has provided the biggest delta as a
result of its mass liquidity injections, but we note that ex-China also turned positive, after
contracting in 2015. Declining areas including Russia and Brazil have effected a turnaround,
while the broader rest-of-world has seen much strong car sales and production rates,
matching stronger PMIs. Galv accounts for around 60% of demand, with diecast alloys (13%),
brass/bronze (11%) and chemicals (9%) the main other first uses. Looking ahead, we expect
continued strength driven by China to support demand in 2017, before falling growth as
construction and infrastructure investment begins to taper into the end of the decade.
Fig 98 It keeps on going Fig 99 Palpable galv. turnaround aids demand
LME cash vs 3M prices ($/t) Galv. steel qtly output growth, YoY
3,000 30%
2,800 25%
$2,400/t
2,600 finally 20%
breaks
2,400 15%
2,200 10%
2,000 5%
-
1,800
-5%
1,600
2012 2012 2013 2013 2014 2014 2015 2015 2016 2016
1,400 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3
Jan 14 Jul 14 Jan 15 Jul 15 Jan 16 Jul 16 Jan 17 China Ex-China World
LME Zinc Cash LME Zinc 3M
Source: LME, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
Fig 100 Mine output tanked as expected Fig 101 ...but China got some extra conc at year-end
Reporting mines contained Zn output (kt) China net zinc imports (kt)
1600
230
1400 Refined zinc Zn-in-zinc conc
1200
180
1000
800
130
600
400
80
200
0
30
-20
2014 2015 2016E Jan-14 Jun-14 Nov-14 Apr-15 Sep-15 Feb-16 Jul-16
Source: Company reports, Macquarie Research, January 2017 Source: LME, SHFE, CRU, USGS, Macquarie Research, January 2017
19 January 2017 40
Fig 102 Are premiums beginning to respond? Fig 103 Price peak seen sooner, and fading faster
Zinc spot concentrate TCs and premiums Zinc in concentrate vs metal balances (kt) vs
250 600 cash price ($/t, rhs) 3000
400 2800
200 200 2600
0 2400
150
-200 2200
100 -400 2000
-600 1800
50 Spot treatment charge ($/dmt) -800 1600
CIF Shanghai premium ($/t) 2013 2015 2017F 2019F 2021F
- Concentrate Balance Refined Balance
Jan 2014 Jul 2014 Jan 2015 Jul 2015 Jan 2016 Jul 2016 LME Cash Price
Source: ILZSG, Wood Mac, CRU, LME, Macquarie Research, January

Source: CRU, Macquarie Research, January 2017 2017
We are especially interested in the fate of diecast alloys and potentially zinc coating weights
given this strong price rally. With supply insufficient to halt deep deficits, we think that
demand destruction will have to be the undoing of zincs bull run, with rival products like
aluminium trailing the price by ~$1000/t by now. Demand destruction usually takes a few
months to kick in, and so as we begin to see Chinese smelters announcing production cuts,
ostensibly due to either a lack of feed or uneconomic spot concentrate treatment charges
(TCs), we think prices should safely rally to average close to $3,000/t by year-end. We do
need to see evidence of refined metal getting tighter, however, something that neither spot
ingot premiums not Chinese imports have been able to demonstrate so far.
The collapse in treatment charges is undeniable, and as we head into mating season proper
for zinc miners, smelters and traders, the question of 2017 terms looms large. How hard will
the miners push the smelters (and other buyers), given their clear advantage, but also
knowing that in the course of a few years all this will have blown over, and long-term
relationships are important in this business. In any case, despite moves by Vedanta and
MMG amongst others to accelerate new zinc projects, and expectations that Glencore begins
to restart their closed operations, for 2017 at least there is not enough concentrate for all
those who want it. Smelter output will be forced lower, hence our expected deficits, and prices
should continue to benefit into year-end. Moving into 2018 should see some more shortages
of metal, but change in the air as demand growth begins to subside should see prices
retracing, before more concrete market balancing arrives into decade-end to bring zinc prices
back down to earth.
Fig 104 Global zinc market balance ('000 tonnes zinc contained)
'000t zinc 2013 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
Mine production 12,902 12,812 13,217 12,383 13,206 13,579 14,194 14,910 15,123
% Change YoY 3.3% -0.7% 3.2% -6.3% 6.6% 2.8% 4.5% 5.0% 1.4%
Concs Balance 454 -45 -35 -643 -118 -34 10 11 -18
% Change YoY 3.3% 2.9% 3.3% -0.5% 1.8% 2.0% 3.8% 4.6% 1.5%
Consum ption 13,044 13,598 13,603 13,928 14,267 14,516 14,793 15,085 15,408
% Change YoY 4.6% 4.2% 0.0% 2.4% 2.4% 1.7% 1.9% 2.0% 2.1%
Refined Balance -224 -403 23 -370 -465 -438 -180 200 106
Estim ated total stocks 2595 2298 2210 1840 1375 937 757 957 1063
LME Cash Price ($/t) 1,909 2,164 1,932 2,092 2,788 2,775 2,475 2,325 2,188
LME Cash Price (c/lb) 87 98 88 95 126 126 112 105 99
Source: CRU, ILZSG, LME, Wood Mackenzie, Macquarie Research, January 2017
19 January 2017 41
Lead
Never a dull moment
Lead finally began to show some interest in rallying as H2 2016 began, and just days away
from the end of the year it was showing the most impressive H2 performance of the complex,
before a swift sell-off brought it back to earth (<$2,000/t on 30 December). The selling
seemed to be linked to end-of-year squaring up amongst speculators as no fundamental
newsflow occurred at the time. Indeed, in the New Year prices have recovered impressively
and are now testing $2,300/t once again.
Whether there is anything solid (i.e. fundamental) behind the moves is harder to say.
Aggregating the handful of reporting mines plus official Chinese NBS data gives us ~65% of
total estimated mine output, and in the first nine months we counted 2.36Mt of contained lead,
down 11% compared with the 2.65Mt reported a year earlier. The TCs have sunk like a stone
as concentrate supply has tightened sharply over the past year, underscoring the primary
supply tightness point. This factor is probably from whence lead draws most of its supporters.
However, scrap batteries comprise over half the supply-side for lead, and a glance at the
price moves (Fig 100) as a proportion of LME prices show spot pricing is clearly lagging the
upswing, suggesting that this market is much looser with buyers not interested in bidding after
the LME.
Fig 105 Lead rushes to catch up, slips, comes back Fig 106 Battery prices certainly not a driver
LME lead cash price ($/t) Scrap lead-acid battery prices (% LME
2,500
85% prices)
2,400
80%
2,300
2,200 75%
2,100 70%
2,000
65%
1,900
60%
1,800
1,700 55%
1,600 50%
1,500 Jan 2013 Jan 2014 Jan 2015 Jan 2016 Jan 2017
Jan-16 Mar-16 May-16 Jul-16 Sep-16 Nov-16 Jan-17 USA EU UK
Source: LME, Macquarie Research, January 2017 Source: LME, Macquarie Research, January 2017
Fig 107 TCs off a cliff shows weak mine output Fig 108 ... but spec volume is almost certainly the key
China import lead concentrate TCs ($/t) SHFE front month lead volume (LHS, k lots) vs
200 3000 avg. price (RHS, RMB/t) 20000
19000
180 2500
18000
160
17000
2000
140 16000
120 1500 15000
14000
100 1000
13000
80 12000
500
60 11000
0 10000
40
Nov-14
Nov-15
Nov-16
Sep-14
Sep-15
Sep-16
Jan-14
Jan-15
Jan-16
May-14
May-15
May-16
Mar-14
Jul-14
Mar-15
Jul-15
Mar-16
Jul-16
20
-
Jan 2014 Jul 2014 Jan 2015 Jul 2015 Jan 2016 Jul 2016 SHFE vols SHFE front month px
Source: CRU, Macquarie Research, January 2017 Source: LME, CRU, Macquarie Research, January 2017
19 January 2017 42
Fig 109 Batteries lifted, except China Fig 110 Lead seen tighter over next few years
Battery production and exports 3MMA YoY, Kt Refined lead supply/demand vs LME cash price $/t
monthly
20% 40% 80 2,400
15% 30% 2,300

60
2,200
10% 20% 40
2,100
5% 10% 20 2,000
0% - 0 1,900
-5% -10% 1,800

-20
1,700
-10% -20%
-40
Jan-15 Jul-15 Jan-16 Jul-16 1,600
N America battery shipments, units -60 1,500
Japan Pb consumption in battery output, kt 2013 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
China battery exports, RHS kt
Refined balance LME Cash Price ($/t)
Source: BCI, CEIC, CRU, Macquarie Research, January 2017 Source: ILZSG, CRU, Wood Mac, Macquarie Research, January 2017
These prices are also pointing to unexciting demand conditions (Fig 103). Ex-China battery
shipments seem to be improving and have been helped by OEM demand from the strong
autos seen worldwide, but Chinese exports have been more negative recently. Recalling that
OEM is only ~15% of lead acid battery demand (which in turn is around 83% of total lead
demand), we note that replacement batteries have been less brisk, with winter weather in the
northern hemisphere unexceptional in most regions. This leaves us turning to the speculators,
and one of the interesting things that has changed in the last few months, which is Chinese
interest: SHFE volumes exploded from averaging 219,000 lots of the front month contract per
month Jan-Oct to 2.23M lots in November-December. Most significantly, experience from
aluminium tells us that once a contract becomes popular in China, volumes may subside but
will not return to previous levels of apathy. This is likely to be more bullish for lead, at least
while zinc remains a favoured long.
Opening up into 2017 and beyond, on the fundamentals we see a deeper deficit this year as
concentrates supply issues restrain refined output. Environmental pressures have already
begun to impact Chinese smelter output, and so we find a market chiefly reliant on secondary
to actually grow. This and the zinc story which lead benefits from means we see prices
heading up to peak around the end of 2017/early 2018, as with its sister metal, before supply
of secondary begins to increase while autos take a step back from recent highs, allowing
prices to begin to retrace. As with most base, we believe the lows have been seen however in
this cycle, and that shallower downside to $1,900-2,000/t will be the worst of it.
Fig 111 Global lead market balance ('000 tonnes lead contained)
'000t lead 2013 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
Mine production 4,967 5,376 5,328 5,110 5,243 5,435 5,663 6,031 6,197
% Change YoY 10.5% 8.2% -0.9% -4.1% 2.6% 3.7% 4.2% 6.5% 2.8%
% Change YoY 3.3% 1.9% 1.3% 2.5% 2.2% 2.5% 2.8% 2.7% 2.1%
Consum ption 11,021 11,250 11,470 11,767 12,052 12,332 12,620 12,927 13,234
% Change YoY 4.4% 2.1% 2.0% 2.6% 2.4% 2.3% 2.3% 2.4% 2.4%
Refined balance 74 55 -15 -22 -47 -22 40 73 36
Estimated total stocks 629 684 669 647 600 577 617 689 726
LME Cash Price ($/t) 2,141 2,096 1,786 1,871 2,350 2,351 2,103 1,945 1,928
LME Cash Price (/lb) 97.1 95.1 81.0 84.9 106.6 106.7 95.4 88.2 87.4
Source: CRU, ILZSG, LME, Wood Mackenzie, Macquarie Research, January 2017
19 January 2017 43
Tin
2-handle returns
Tin was a star performer in 2016 only second to zinc in its recovery rally back above
$20,000/t, having spent time below $14,000/t early in 2016. Aside from the broader
commodities rebound driven by better than expected Chinese demand levels, some key
reasons behind the recovery for this particular metal include the low visible stock levels on the
exchanges and reduced export volumes from key producer Indonesia and lower production in
China and Myanmar. However, like most metals, demand has been the most decisive factor,
as semiconductor shipments have picked up strongly worldwide in recent months.
We estimate that Indonesian mine production was down 13% YoY in 2016 to 61kt, after price-
related mine closures at the start of the year, effective export quotas via government
regulation relating to illegal mining and reserves depletion. Exports of metal fell 9% YoY to
63.6kt over the year. However, shipments of ores and concentrates rocketed out of Myanmar,
up 94% YT November based on Chinese trade data (metal-contained basis), as both volumes
and implied tin content rose into year-end. ITRI cautions, however that the increases are
chiefly related to stock releases, and that actual mine output is now struggling as exploitation
moves underground. We thus have Myanmar production growth slowing from 2017 onwards.
As previously noted, Chinese metal output began to be affected from July by environmental
shutdowns enforced by the central government, and the data to October implies no real
recovery with volumes just above 14ktpm, which implies stocking of Myanmar concentrates.
Fig 112 Back over $20k/t; was it all a bad dream? Fig 113 Backwardation persists with low stocks
LME tin cash price ($/t) Tin LME stocks (LHS, t) vs cash-3 month price
14,000 spread (RHS, $/t) 600
24,000
12,000 500
22,000 10,000 400
8,000 300
20,000
6,000 200
18,000
4,000 100
16,000 2,000 0
- -100
14,000 Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan
15 15 15 15 15 15 16 16 16 16 16 16 17
12,000
LME stocks Cash-3 month
Jan-13 Aug-13 Mar-14 Oct-14 May-15 Dec-15 Jul-16
Source: LME, Macquarie Research, January 2017 Source: LME, Macquarie Research, January 2017
Fig 114 2016 Indo export volumes fell again Fig 115 but inventories at key producers rose
Indonesian tin surveyed for export (ktpa) Timah + Minsur refined tin inventories (t Sn)
180 Tin checked for export 10,000
YTD monthly average
160
9,000
140
8,000
120
100 7,000
80
6,000
60
5,000
40
20 4,000
0
3,000
2011 2012 2013 2014 2015 2016
Q113 Q313 Q114 Q314 Q115 Q315 Q116 Q316
Source: ITRI, Macquarie Research, January 2017 Source: Company Reports, Macquarie Research, January 2017
19 January 2017 44
Turning to refined inventory, however, we see a more constrained picture, with metal supply
issues meaning that LME, SHFE, and the producer stocks we trace are all down to historically
low levels. These supply constraints meant stock moved down to historically low levels. LME
stocks are ~4kt, while SHFE stocks have tended to oscillate between 2 and 3 kt in 2016. This
led to extreme backwardations flaring out in the forward LME curve (Fig 107) over the last
couple of years, as dominant positions emerged amongst the warrants and squeezed the
near dates. Tin is particularly susceptible to overextensions like this due to its lack of liquidity,
but we must note that the cash-3 has more lately softened to a $11/t back after steady
reverse cancellations, plus some muted new inventory inflow, and lending along the near
dates allowed the front of the curve to soften.
While these stocks and supply developments are interesting they are not very dynamic, and
we therefore conclude that the price rally witnessed was mostly driven by the one factor that
did change dramatically: demand. Coming off a 3% YoY contraction in 2015 as consumer
goods including electronics slumped, 2016 did not initially demonstrate promise. However, as
Figure 111 shows, a strong turnaround for semiconductors (50% demand) began in mid-year
and has continued into year-end, while tinplate production has been boosted by a stronger-
than-anticipated Chinese steel sector. Together accounting for 2/3 of demand, these market
sector improvements lead us to estimate +1% growth for 2016, and an overall deficit of 600t.
Moreover, Macquaries semiconductor analysts are projecting global sales growth of 9% in
2017 and 3% in 2018 on a continued recovery in the sector and wafer fabrication capacity
being added to aggressively. China has been the key growth area (as ever), but other Asia
including Japan also looks healthy, and so we have lifted demand growth to 2.3% (from flat)
in 2017. This pushes the market into deeper deficit from 2017 on of 2.6-3.8ktpa, allowing us
to lift our price expectations to $21,000-23,000/t over the next four years.
Fig 116 Myanmar ores and concs exports maintained Fig 117 Solder alloys (~50% tin consumption) have
strong levels in 2016, but from destocking turned very bullish lately
China's tin concentrate imports (kt gross) Global semiconductor shipments

80 15%
RoW
70
3MMA, YoY % change
Myanmar
10%
60
50 5%
40
0%
30
20
-5%
Value basis
10
Volume basis
0 -10%
Jan 11 Jan 12 Jan 13 Jan 14 Jan 15 Jan 16 2011 2012 2013 2014 2015 2016
Source: Customs data, Macquarie Research, January 2017 Source: WSTS, Macquarie Research, January 2017
Fig 118 Global tin market balance (000 tonnes tin contained)
'000t tin contained 2013 2014 2015 2016E 2017F 2018F 2019F 2020F 2021F
Tin mine production 285.3 296.3 270.9 272.3 274.8 275.3 275.7 275.8 276.6
YoY change 3% 4% -9% 1% 1% 0% 0% - 0%
Tin metal production 340.8 363.5 341.3 345.0 350.7 353.7 355.0 356.1 357.7
YoY change 2% 7% -6% 1% 2% 1% 0% 0% 0%
Tin consumption 344.0 352.7 340.6 345.6 353.4 357.3 358.8 359.7 361.1
YoY change 3% 3% -3% 1% 2% 1% 0% 0% 0%
Tin m arket balance -3.2 10.8 0.7 -0.6 -2.6 -3.6 -3.8 -3.6 -3.4
Reported stocks 29.9 33.0 25.4 24.8 22.2 18.6 14.8 11.2 7.8
Stocks (w eeks) 4.5 4.9 3.9 3.7 3.3 2.7 2.1 1.6 1.1
LME tin cash price ($/t) 22,305 21,893 16,077 17,991 21,750 21,375 23,000 21,000 20,500
Source: ITRI, CRU, WBMS, LME, Macquarie Research, January 2017
19 January 2017 45
Nickel
Indonesian ore export relaxation dashes strong price recovery
2016 was a turning point in the nickel market as supply cuts and a better-than-anticipated
demand recovery pushed the market from structural surplus (of over 500kt Ni combined over
the past five years) to what looks like to be a structural deficit. The Q4 2016 average of
$10,810/t was 27% higher than the first quarter average. However, prices sold off sharply in
December amid fund end-year profit-taking and, as it turns out, well-placed fears of an
Indonesian ban reversal.
Following the surprise Indonesian announcement last week to pave the way for nickel ore and
bauxite exports to recommence, further details of the impact are coming to light. Over the
weekend, the government stated that 5.2mt of nickel ore per annum would be allowed to be
exported roughly a tenth of the levels seen in 2013. After adjusting for moisture, grade and
yield loss this would equate to just over 50kt of nickel contained. However in our view this is
still enough to cause downward pressure on ore prices, NPI production costs and LME
prices. For further details of the regulation changes and Indonesias importance to the global
nickel market please refer to our 12th January Commodities Comment.
Fig 119 Nickel market moves into deficit in 2016 for Fig 120 Rollercoaster price recovery in 2016 as
first time since 2010 following stock build of 500kt+ doubts remained about Indonesian supply potential
Nickel supply/demand balance and stocks LME nickel cash price ($/t)
13,000
200 27
160 25
150 12,000
121 120 23
104 101
100 90 21
19 11,000
48
weeks' use
50 32 34
'000t Ni
31 17
9 4
15 10,000
0
13
-50 -26 9,000
-36 11
-56
9
-100
-91 7 8,000
-150 5
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016E
7,000
Jan-16 Mar-16 May-16 Jul-16 Sep-16 Nov-16 Jan-17
Balance (LHS) Stocks in weeks of use (RHS)
Source: INSG, Macquarie Research, January 2017 Source: LME, Macquarie Research, January 2017
The stronger 2016 performance reflected booming demand with estimated 2016 annual
global demand growth of 7.8% YoY for refined nickel, driven by strong Chinese stainless steel
production and strong demand growth for nickel in batteries for electric vehicles.
Low nickel prices in late-2015 and early 2016 led to over 70% of the industry moving into
loss-making and amid large price-related production cuts, which more than offset increases in
production in Indonesia and a number of other new suppliers. Overall, global refined nickel
production fell by an estimated 1% in 2016 following a similar 1% fall in 2015. Global supply
should return to growth from 2017 onwards, due mainly to new NPI capacity in Indonesia and
the small resumption of ore exports, but the lack of new supply growth elsewhere should lead
to supply growth of only 2.3% this year.
Government policy in 2017 in the Philippines (an environmental audit of nickel mines) is still
unclear but ore supply last year was more a function of ore price (low in H1 and higher in H2)
than policy. The partial relaxation of the Indonesian nickel ore ban will lower ore prices and
lead to lower exports from the Philippines and New Caledonia of mid-grade ore.
In 2016, Indonesian nickel pig iron production grew to an estimated 93kt from only 28kt in
2015. It is expected to grow strongly again in 2017 (to 180kt) as new supply is commissioned
from Tsingshan, PT Virtue Dragon (Delong) and the PT Harita/Xingxing joint venture. The
pace of building should accelerate in the coming years due to the low cost of making nickel
pig iron in Indonesia when compared with China (a cost advantage of $2,500-3,000/t of
nickel) we expect Indonesian NPI production to exceed 300ktpa by 2020.
19 January 2017 46
The one advantage Chinese NPI producers do have is integration into stainless steel
production by hot metal transfer. The Chinese NPI producers are heavily reliant on ore from
the Philippines. They are, however, increasing their diversification to other feed sources
including nickel concentrates and other intermediate products from the sulphide ore industry.
Tsingshan is now getting feed from Zimbabwe, Russian, Finland and Australia (potentially 35-
40kt Ni in 2017) to offset losses of feed sources from the Philippines.
Exports of recoverable nickel in ore from the Philippines fell by an estimated 67kt Ni in 2016
to 327kt. The year-on-year fall was entirely in the first half of the year and reflected the
depletion of high-grade reserves at the Tawi Tawi mine and deliberate production cuts due to
low nickel ore prices. Surprisingly, however, exports rose strongly in the second half of 2016
despite a central governmental environmental audit of the industry and the suspension of
production at several small mines.
We now expect a year-on-year fall in Pilipino exports in 2017 to 290kt due mainly to lower
nickel ore prices. Chinese NPI production should fall slightly this year to 370kt due to the
expected depletion of ore stocks in the Chinese ports and at plants. We estimate that around
40kt of nickel in 2016 NPI production came from destocking and this cannot continue in 2017.
The more immediate challenge for Chinese NPI producers, however, is rising costs. Due to
rises in coal/coke costs, freight costs and ore costs, we estimate that the breakeven for
efficient Chinese NPI producers (using RKEF technology) rises from $8,500-9,000/t in
February 2016 to $11,000-11,500/t (LME equivalent basis) by December, forcing most
Chinese producers into loss-making. With ore and coal prices now expected to fall in 2017,
the upward pressure on nickel prices will now dissipate, pushing prices back to the $9,500-
10,000/t range.
Fig 121 Nickel ore exports from Philippines price Fig 122 The key assumption in nickel: NPI production
elastic: up YoY in 2H 2016 as nickel ore prices recover growth in China and Indonesia
180
700
159
160 2015 2016
600
'000t recoverable Ni in ore
140
122 500 320 350
118 230 280
120 111 180
28 93
400
'000t Ni
100
80 300
59 60 500 480
58
60 200 400 381
361 370 360 350 340 340
40 35
100
20
0
2012 2013 2014 2015 2016E 2017F 2018F 2019F 2020F 2021F
0
Q1 Q2 Q3 Q4 China Indonesia
Source: Antaike, Macquarie Research, January 2017 Source: INSG, Macquarie Research, January 2017
Fig 123 Global nickel supply/demand balance

`000t 2014 2015 2016 2017f 2018f 2019f 2020f 2021f
Total SS production 42924 42208 45713 46977 48288 49452 50823 52050
% Change 8.7% -1.7% 8.3% 2.8% 2.8% 2.4% 2.8% 2.4%
300-series SS prod. 23252 23581 25059 25930 26812 27475 28504 29316
% Change 7.2% 1.4% 6.3% 3.5% 3.4% 2.5% 3.7% 2.9%
Primary nickel consumption 1889 1877 2024 2065 2152 2206 2288 2341
% Change 4.7% -0.6% 7.8% 2.0% 4.2% 2.5% 3.7% 2.3%
Nickel Supply 1993 1978 1968 2012 2110 2182 2226 2286
% Change 1.5% -0.8% -0.5% 2.3% 4.9% 3.4% 2.0% 2.7%
(of which NPI) (483) (428) (474) (550) (590) (630) (660) (690)
World Market Balance 104 101 -56 -53 -41 -24 -62 -55
Estimated total stocks 813 908 852 799 758 734 672 617
Weeks' world demand 22.4 25.1 21.9 20.1 18.3 17.3 15.3 13.7
LME Cash Price ($/tonne) 16867 11836 9599 9813 11000 12000 13000 14000
Source: INSG, Macquarie Research, January 2017
19 January 2017 47
Stainless Steel
Booming 2016 Chinese production driven by exports and restocking
2016 proved to be a much stronger year for global stainless steel production that had been
expected at the start of the year. Since January 2016, we have raised our forecast of global
growth from 0.7% to 8.3%. This rise has been due to China where our forecast has been
raised from -0.7% in January to +13.4%. Chinese growth was driven by a remarkable rise in
stainless steel exports, which grew an estimated 19.4% YoY. Virtually all of the growth in
exports has been to other Asian countries with Taiwan accounting for half of the growth. In
the second half of the year, non-Chinese production also grew, by over 7% YoY.
Fig 124 World stainless steel production by major area (000t melt stainless)
2016 2016 2016 2016 Year Year Year Year Year Year Year Year
'000t Q1 Q2 Q3 Q4 2014 2015 2016 2017F 2018F 2019F 2020F 2021F
USA 530 613 687 665 2389 2346 2495 2532 2532 2532 2481 2432
Japan 775 745 791 792 3331 3061 3104 3042 3011 2981 2892 2877
Europe 1861 1926 1659 1789 7237 7162 7234 7261 7148 7016 6793 6789
Korea 526 564 593 580 2038 2231 2263 2263 2218 2262 2307 2331
Taiwan 299 324 309 285 1103 1126 1217 1156 1087 1054 1033 1012
India 844 860 868 870 2973 3209 3441 3682 3958 4275 4617 4894
China 5242 6471 6310 6709 22736 21933 24732 25314 26189 26819 27669 28416
Indonesia 0 13 0 0 13 500 900 1250 1750 2000
Other 299 330 308 291 1116 1141 1214 1227 1244 1262 1280 1299
Total (inc. others) 10375 11833 11524 11981 42924 42208 45713 46977 48288 49452 50823 52050
Total Ex-China 5133 5362 5214 5272 20187 20275 20981 21663 22099 22633 23154 23634
Chinese net exports 817 965 1,102 895 3563 3167 3780 3647 3529 3059 2706 2706
% change YoY
USA -19.7% 6.0% 16.1% 28.9% 17.7% -1.8% 6.3% 1.5% 0.0% 0.0% -2.0% -2.0%
Japan -4.0% 3.2% 0.5% 6.5% 4.9% -8.1% 1.4% -2.0% -1.0% -1.0% -3.0% -0.5%
Europe -3.0% -1.8% 4.3% 5.7% 0.9% -1.0% 1.0% 0.4% -1.5% -1.8% -3.2% -0.1%
Korea -1.9% -0.9% 6.2% 2.3% -4.9% 9.4% 1.5% 0.0% -2.0% 2.0% 2.0% 1.0%
Taiwan 5.8% 9.9% 16.1% 0.8% 2.0% 2.1% 8.0% -5.0% -6.0% -3.0% -2.0% -2.0%
India 6.6% 7.8% 8.0% 6.7% 10.9% 7.9% 7.3% 7.0% 7.5% 8.0% 8.0% 6.0%
China -0.2% 10.9% 15.2% 24.9% 13.4% -3.5% 12.8% 2.4% 3.5% 2.4% 3.2% 2.7%
Indonesia 3746% 80.0% 38.9% 40.0% 14.3%
Other 3.5% 9.8% 12.8% 4.3% -3.2% 2.2% 6.4% 1.0% 1.4% 1.4% 1.4% 1.5%
Total World -1.5% 7.0% 11.4% 16.7% 8.7% -1.7% 8.3% 2.8% 2.8% 2.4% 2.8% 2.4%
Total Ex-China -2.9% 2.6% 7.1% 7.7% 3.9% 0.4% 3.5% 3.2% 2.0% 2.4% 2.3% 2.1%
Chinese net exports 11.3% 28.3% 34.6% 3.9% 58.4% -11.1% 19.4% -3.5% -3.2% -13.3% -11.5% 0.0%
Source: ISSF, CRU, Tex Report, Macquarie Research, January 2017
Fig 125 Chinese stainless growth driven by soaring Fig 126 Large investment in new integrated stainless
exports in 2016, not by over-stocking steel and nickel pig iron facilities
'000t '000t '000t % change YoY
Production Net exports App. Cons Production Net exports App. Cons Capacity 400 and Integrated
Q114 5372 702 4671 15.3% 40.9% 12.2% mtpa melt 300 series 200 series Start-up NPI
Q214 5854 1059 4795 26.5% 70.8% 19.7% China
Q314 5853 962 4891 12.4% 76.5% 4.9%
Q414 5657 841 4816 1.8% 43.2% -3.1%
Shangtai 1.0 1.0 0.0 mid-2016 Yes
Year 22736 3563 19173 13.4% 58.4% 7.7% Delong 2.0 2.0 0.0 Q4 2015 Yes
Behei Chengde 0.9 0.9 0.0 2015/16 Yes
Q115 5251 734 4517 -2.3% 4.6% -3.3% Behei Chengde II 0.9 0.9 0.0 2017/18 Yes
Q215 5837 752 5085 -0.3% -29.0% 6.0% Jinghai Hui 0.5 0.5 0.0 2016 Yes
Q315 5476 819 4657 -6.4% -14.9% -4.8% Zhongjin 0.5 0.0 0.5 2015/16 Yes
Q415 5369 862 4508 -5.1% 2.5% -6.4% Jinhui Shengyang 1.6 0.0 1.6 2016/17 Yes
Year 21933 3167 18767 -3.5% -11.1% -2.1% Baosteel Desheng 0.6 0.0 0.6 2017/18 Yes
Xinhai 2.0 2.0 0.0 H2 2017 Yes
Q116 5242 817 4425 -0.2% 11.3% -2.0%
Q216 6471 965 5507 10.9% 28.3% 8.3% Sub-total China 10.0 7.3 2.7
Q315 6310 1102 5207 15.2% 34.6% 11.8% Indonesia
Q416E 6709 895 5813 24.9% 3.9% 29.0% Tsingshan 1.0 1.0 1.0 Q1 2017 Yes
Year 24732 3780 20952 12.8% 19.4% 11.6%
Total above 11.0 8.3 3.7
2017F 25314 3647 21667 2.4% -3.5% 3.4%
Source: CSSA, Chinese Customs, Macquarie Research, January 2017 Source: ISSF, CRU, Macquarie Research, January 2017
19 January 2017 48
The strong growth in supply reflects a recovery in nickel prices, which has induced some
restocking by steel distributors and traders but also better underlying demand growth in
China, Europe, India and more recently, the USA. Reports from the USA and Europe indicate
a strong improvement in stainless orders in Q2 2016 and for Q1 2017, so this augurs well for
at least the first half of 2017.
In China, booming demand was already apparent in 2016, in part due to stimulative measures
by the Chinese government. Chinese data shows apparent consumption growth in China of
9.9% in 2016 but this accelerated to an unsustainable 29% YoY by Q4 2016, and there are
reports of slowing Chinese production as we enter 2017. For 2017 as a whole we are
projecting Chinese apparent consumption to slow to only 3.4% this year and for production
growth to slow to 2.4%, lower than for the rest of the world.
Our 2017 forecasts assume some slowdown in global production growth to reflect some
destocking, and we have conservatively projected 2017 growth in global production at only
2.6% in both 2017 and 2018. This may be too conservative, especially if global growth
continues to reaccelerate in 2017.
In our forecasts for 2017 onwards, we assume a fall in Chinese exports in part due to the start
of Chinese-owned stainless production in Indonesia. In our forecasts of ex-Chinese
production growth to 2021, we assume virtually all the growth is in India and Indonesia.
We continue to be cautious about the growth outlook to 2021 despite the historical reality that
whenever the global economy recovers, world production often grows at double-digit rates
due to heavy restocking. We had previously assumed that Chinese exports would decline in
2016 due to China losing its strong competitive advantage in nickel.
However, to offset this several Chinese producers are now making massive investments in
integrated NPI/Stainless capacity to try and regain competiveness. We have identified
11mtpa of new melt capacity in China and Indonesia, of which 8mtpa is for 300 series (see
Figure 120). This is an unstoppable trend, but it may well be that these producers will
transfer more capacity to Indonesia if the initial Indonesian investment by Tsingshan proves
to be successful.
We think 2016s Chinese export surge was temporary and exports will decline in part as some
Chinese exports are displaced by growing Chinese investment in stainless steel capacity in
Indonesia (starting this year with Tsingshans new 1mtpa slab facility, which was originally
due to start in 2016).
An important trend in our global forecasts is a faster rise in 300 series stainless steel
production than the total, although it appears that Q4 2016 saw a small reversal in that trend.
This reflects a move away from inferior low nickel 200 series in China and some claw back
from no nickel 400 series elsewhere. It is a partial reversal of the massive loss of 300 series
production share in 2006/07 due to extremely high nickel prices. Virtually all of the new
capacity is being built in China and Indonesia for 300 series production.
Fig 127 Summary of global stainless forecasts by grade and nickel use (mt)
2014 2015 2016f 2017f 2018f 2019f 2020f 2021f
Total stainless production (mt) 42.9 42.2 45.7 47.0 48.3 49.5 50.8 52.1
of which:
Non-nickel containing (400-series) 10.1 9.8 10.5 10.8 11.1 11.4 11.6 12.0
200 series (Mn) austenitic (1-3% Ni) 9.6 8.8 10.1 10.2 10.4 10.6 10.7 10.8
300 series austenitic (8-9% Ni) 23.3 23.6 25.1 25.9 26.8 27.5 28.5 29.3
Total austenitic production (mt) 32.8 32.4 35.2 36.2 37.2 38.1 39.2 40.1
Austenitic ratio % 76.5% 76.7% 77.0% 77.0% 77.1% 77.0% 77.1% 77.0%
Average scrap ratio (in austenitic) 38.8% 38.2% 36.6% 37.5% 36.7% 36.7% 36.3% 36.4%
Primary nickel in stainless ('000t) 1317 1330 1454 1477 1538 1570 1630 1667
Year-on-year changes
Total stainless 8.7% -1.7% 8.3% 2.8% 2.8% 2.4% 2.8% 2.4%
200 series 14.8% -8.5% 15.4% 0.9% 1.8% 1.8% 1.0% 0.7%
300 series 7.2% 1.4% 6.3% 3.5% 3.4% 2.5% 3.7% 2.9%
400 series 6.9% -2.3% 6.8% 2.8% 2.3% 2.8% 2.0% 2.9%
Primary nickel in stainless 5.6% 1.0% 9.3% 1.6% 4.1% 2.1% 3.8% 2.3%
Source: ISSF, CRU, Macquarie Research, January 2017
19 January 2017 49
Ferrochrome
In desperate need of a supply response
While the Indonesian nickel ore ban relaxation is set to ease one potential raw material
constraint for Chinese stainless steel producers, a more significant one remains. Both
chrome ore and ferrochrome prices are at record levels in Rand terms, having risen very
strongly over the course of Q4 2016. Just like many other commodities, such a strong rise is
unlikely to be sustainable as the market adjusts, however without a discernible supply
response we are looking at a chrome market which will be consistently running down
inventories over the coming years, both in ore and metal.
Chrome ore prices rose four-fold from March to December last year, highlighting the raw
material constraint in this market. Thus, ferrochrome production costs in China have risen
quickly, and these correlate very closely with the spot ferrochrome price, making this an
efficient market. With the ore price gains, production costs have more than doubled, and so
has the spot ferrochrome price.
Certainly, the chrome market has been tightened significantly by rising stainless steel output.
And unlike nickel, there is no new supply to help satiate this. Moreover, there is no chance of
direct substitution. Stainless steel needs ferrochrome to make it stainless. When stainless
steel output is growing, and South African ore supply isnt, the chrome market has a major
problem. And 2016 saw South African ex-UG2 supply fall 4% YoY while global demand from
stainless steel rose 7.5% YoY. South Africa makes up >55% of global mined chrome, and
China now gets 75% of its chrome ore from South Africa, a level which has steadily risen from
11% a decade ago. Roughly half this growth however has come from UG2 ore, which is by-
product material extracted from tailings of PGM mines, some of which had been lying above
ground for decades. UG2 extraction looks to have plateaued, and as such cannot provide
another leg of growth. As a result, Chinas chrome ore port stocks are at multi-year lows.
The strength was affirmed by the Q1 2017 European high carbon ferrochrome benchmark
settlement at $1.65/lb, up 50% QoQ to the highest level since Q4 2008. In Rand terms, this is
the highest ever quarterly ferrochrome contract. However, domestic ferrochrome prices in
China are now starting to ease as stainless steel output sequentially weakens. Having
declined marginally in January, leading Chinese steelmakers are cutting their February
ferrochrome purchase price by 8% to RMB9,535/t. This remains an extremely strong price for
producers.
We assume the contract price falls from Q3 2017 as the industrial cycle eases and as current
price levels allow chrome ore producers to squeeze out a little more medium- and low- grade
ore from South Africa, abetted by increases in other regions. However, even assuming a
strong production rebound from South Africa, and strong efficiencies in ferrochrome use, we
still end up with a very tight chrome market by the end of the decade.
Fig 128 Of all commodity moves, that in ferrochrome Fig 129 The surge in chrome ore prices has driven the
has perhaps been most dramatic Chinese industry cost structure higher
Chinese FeCr spot prices Chinese charge chrome costs vs imported FeCr price, c/lb
1.20
130 19
Imports CIF 58%-60% Cr Imported FeCr price
120 1.10
Imports - ZAR/lb CIF (RHS) 17
FeCr production costs from imported ore
110 1.00
US/lb Cr, ex-VAT
15
100
0.90
90 13
0.80
80 11
70 0.70
9
60 0.60
7
50
0.50
40 5
S 2012
S 2010
S 2011
S 2013
S 2014
S 2015
S 2016
M 2010
J 2010
J 2011
M 2011
J 2012
M 2012
J 2013
M 2013
J 2014
M 2014
J 2015
M 2015
J 2016
M 2016
0.40
Jan-13
Jan-12
Jan-14
Jan-15
Jan-16
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
Nov-12
Sep-12
Mar-13
Sep-13
Sep-14
Sep-15
Mar-12
May-13
Nov-13
Mar-14
Nov-14
Mar-15
Nov-15
Sep-16
Mar-16
Nov-16
May-12
May-14
May-15
May-16
Source: CRU, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
19 January 2017 50
Fig 130 Chinese chrome ore port inventories remain Fig 131 ...and unless we get a more aggressive supply
at multi-year lows... reaction, global inventories will continue to fall
China's Cr ore port stocks FeCr equivalent stocks - base case

3.5 25
3.0
20 FeCr
3MMA, m tonnes
2.5
weeks of consumption
Chrome Ore
2.0 15
1.5 10
1.0
5
0.5
0.0 0
2014 2015 2016 2017F 2018F 2019F 2020F 2021F
Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16
Source: Mysteel, Macquarie Research, January 2017 Source: CRU, ICDA, Macquarie Research, January 2017
Fig 132 South African ore output did start to respond Fig 133 India has potential to become more important
to higher prices in late 2016 in solving the chrome supply problem
Reported South African Cr Ore output (ex-UG2)

20
18
16
14
12
10
6MMA
8
6
Apr-12
Jul-12
Apr-13
Jul-13
Apr-14
Jul-14
Oct-14
Apr-15
Jul-15
Apr-16
Jul-16
Oct-12
Oct-13
Oct-15
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Source: DTI, Macquarie Research, January 2017 Source: Steel Mint, Macquarie Research, January 2017
So who suffers from a lack of chrome? The main growth in stainless output over the coming
years comes from three countries India, China and Indonesia. India has access to
domestic chrome ore, however the others dont China has next to no domestic resource
(only 1% of the world total). Thus, we expect a push for China to seek chrome ore supply
alternatives from bordering countries like Kazakhstan and Pakistan, plus other international
suppliers such as Oman and Zimbabwe. Even so, on our base case we see a continual draw
on chrome ore (and ferrochrome) stocks over the coming years.
Fig 134 Global ferrochrome and chrome ore balance

Ferrochrome (kt) 2015 2016 2017F 2018F 2019F 2020F 2021F
Production 10,699 10,806 11,320 11,773 12,018 12,125 12,271

YoY change -4.8% 1.0% 4.8% 4.0% 2.1% 0.9% 1.2%
Consumption 10,840 11,654 11,774 11,928 12,028 12,182 12,283
YoY change -1.6% 7.5% 1.0% 1.3% 0.8% 1.3% 0.8%
Balance -141 -848 -454 -155 -10 -57 -12
Stocks 2,204 1,356 901 746 736 679 667
Weeks of consumption 11 6 4 3 3 3 3
Chrome ore (kt gross weight) 2015 2016 2017F 2018F 2019F 2020F 2021F
Balance 46 -360 -598 -972 -668 -300 -600

Stocks 6,458 6,098 5,500 4,529 3,861 3,560 2,960
Weeks of consumption 10 9.6 8.3 6.6 5.5 5.0 4.1
Source: CRU, ICDA, Macquarie Research, January 2017
19 January 2017 51
Molybdenum
Big supply cuts rebalance market but medium-term oversupply risks
remain
2016 was an adjustment year for the troubled molybdenum market as global production was
cut savagely for the second consecutive year to rebalance a massively oversupplied market
hit by a collapse in investment by the oil and gas industries.
In addition to weaker demand, the industry had to cope with the start-up of several expansion
projects (notably Sierra Gorda, China Gold, Luming, Buenavista, Cere Verde and
Caserones). Global supply was estimated to have been cut by 8.9% YoY in 2015 and 7.7%
YoY in 2016, and the market swung from a large surplus in 2015 to a small deficit in H1 2016,
allowing prices to recover strongly from multi-year lows by mid-2016.
However, a massive surge in Chinese molybdenum exports in Q3 2016 stalled the recovery
in its tracks and served as a timely reminder that supply discipline has to be maintained by
the global industry for many years to prevent large surpluses re-emerging again.
The closures included primary, co- and by-product supplies with Freeport McMoran, the
worlds largest supplier, leading the way with an estimated 17m lbs YoY cut in 2016. In
China, small-scale mines reportedly slashed production amidst a collapse in Chinese demand
with estimated mine production falling 6% YoY in 2015 and a further 10% in 2016 to around
168m lbs. The process of adjustment is being helped by involuntary shortfalls at Sierra
Gorda, Rio Tinto, Antofagasta and Caserones of around 40mlbs from plan in 2016.
Even though oil and gas investment is under 20% of total molybdenum use, segments of this
use have reportedly fallen by 30-50%, especially in low alloy steels. Demand falls stabilised in
2016 and a recovery in oil prices should lead to a recovery in demand, but probably more in
2018 than 2017.
In 2017 and 2018, potential supply growth is quite strong from new copper by-product
projects and the return to normal or planned production at existing producers. We are
projecting that supply should recover strongly in 2017 (we are forecasting growth of 9.8%
YoY). The market looks broadly balanced in both 2017 and 2018 on the assumption that
supply remains below potential in those years.
We have made some downward revisions to our supply forecasts out to 2020 to reflect the
recent weakness in copper prices and the poor financial state of the global mining industry
the pressure on miners to cut capital expenditure and delay projects remains, and this
sometimes means delays in installing moly by-product recovery circuits.
Chinese supply growth will remain price dependent and could surprise on the upside but only
if prices moved above the $8/lb level. In the $7-8/lb range, we assume that Chinese supply
rises to meet ongoing Chinese demand growth. In late 2015/early 2016 the bulk of primary
supply was operating at a loss, and the industry voluntarily agreed to 10% across-the-board
cuts in 2016, which to their credit, they adhered to. The Chinese government has reportedly
intervened to a certain extent reportedly buying 20mlbs+ or so for its strategic stockpile since
July 2014.
The Chinese government scrapped export quotas on molybdenum products in 2015 and net
exports of molybdenum products (excluding unroasted concentrate) rose to just over 19m lbs
last year, up from 5.7m lbs in 2015, a 235% YoY rise. While net imports of concentrates also
rose (from 6.1m lbs in 2015 to around 12m lbs in 2016) partly offsetting the product growth,
the sudden rise in Chinese exports certainly underscores that there is a lid on potential price
rises at around $8/lb. China was a large net exporter in the 2000s and the large over-capacity
in China means that it could return to such a situation again, but only if the price is right.
We expect average prices in 2017 to be higher than in 2016 and for prices to stay in the $7-
8/lb range over the forecast period. We lowered our 2016 deficit estimate from our previous
report from 25m lbs to only 5mlbs (following a small H2 recovery in supply in China, Peru and
Mexico). Each year from 2017-2021 is projected to be a surplus year, but cost factors should
keep the price mainly in the $7-8/lb range over this period with a rise to the top of the range
likely towards the end of the forecast period.
19 January 2017 52
Fig 135 World molybdenum supply/demand balance (m lb Mo)

m lbs Mo 2013 2014 2015 2016 2017f 2018F 2019F 2020F 2021F
Demand
Europe 140 144 135 136 140 144 147 150 152
USA 56 60 54 52 53 56 60 62 64
Japan 60 60 52 51 52 54 55 57 58
China 196 202 178 170 185 202 206 210 213
Other 85 96 94 93 95 101 106 113 118
Total Demand 538 561 513 501 526 557 574 592 605
Change YoY 2.7% 4.4% -8.7% -2.3% 5.0% 6.0% 3.1% 3.1% 2.1%
Supply
Primary mine production 291 276 229 177 194 213 218 230 239
By-product mine production 254 299 293 310 345 370 394 391 388
Catalysts 11 12 12 12 13 13 14 14 14
Disruption allowance/yield losses -8 -9 -8 -8 -24 -32 -33 -34 -34
Total Supply of products 548 577 526 493 527 564 592 600 607
Change YoY 1.4% 5.3% -8.9% -6.3% 7.0% 6.9% 5.0% 1.4% 1.1%
Market Balance 11 16 13 -8 1 7 17 8 2
Price $/lb Mo oxide 10.32 11.43 6.80 6.52 7.25 7.00 7.00 7.50 8.00
Source: WBMS, Macquarie Research, January 2017
Fig 136 Moly oxide prices recovered in 2016 after a Fig 137 Global moly supply and demand Deficit
disastrous two years of falls, but rally stalls emerges in 2016 after seven years of surplus
18 15
16
10
14
Nickel: cents/lb
m lbs Mo balance
5
12
10 0
8
-5
6
-10
Q114 Q214 Q314 Q414 Q115 Q215 Q315 Q415 Q116 Q216 Q316 Q416E
4
Implied market balance
2010 2011 2012 2013 2014 2015 2016
Source: Platts, Macquarie Research, January 2017 Source: IMOA, Macquarie Research, January 2017
Fig 138 Supply follows demand down in 2015/16 Fig 139 A huge rise in Chinese moly product exports
610
10
590 9
8
570 Net exports
6
550 4.2
4.5
4.0
m lbs Mo content
m lbs Mo
4 3.2 3.3
530 2.7 2.6
2.4
2 1.6 1.6 1.4
510
0
490 -0.9 -0.7 -0.4
-1.1 -1.2
-2 -1.6
-2
470 -2.8
-2.4 -2.5
-4 Net imports
-4.1
450 -4.9
-6
Q114 Q214 Q314 Q414 Q115 Q215 Q315 Q415 Q116 Q216 Q316 Q416E
Products Concentrates (unroasted)

Supply Demand
Source: Company reports, Macquarie Research, January 2017 Source: IMOA, Macquarie Research, January 2017
19 January 2017 53
Cobalt
Risks remain, but looking ripe for 2017 upside
There has been a long wait for cobalt prices to start moving upwards in a conspicuous
fashion. The excitement surrounding battery demand has been there, but so has a raw
material overhang from previous years. Only recently, with the global industrial recovery
maturing into a more broad-based one, have we seen cobalt prices move back to 2011 levels.
And quickly too, with current spot prices ~$16/lb, a level we only expected over the medium
term. There remains a large degree of risk to the cobalt story, both from the supply and
demand side, however the fundamental outlook does look to be improving for 2017, and we
see a sustained deficit and stock draw through the end of the decade.
Just as with many other markets, cobalt has limited new supply projects coming through.
Meanwhile, refined output in key supply countries such as Australia, Russia and Zambia are
well down on levels seen a decade ago. As a result, the cobalt market is becoming ever
more dependent on supply from the Democratic Republic of Congo (DRC), a country where
geopolitical risk is once again rising after a period of relative stability. Moreover, it is also a
country under increasing pressure domestically to add more value to mined products in the
country, and under increasing pressure internationally to clamp down on artisanal mining
with cobalt extraction the poster child for this.
All of these factors create significant risk for Chinese buyers, who increasingly dominate the
purchase of cobalt raw materials from the DRC. Of the ~54kt of recoverable cobalt units
imported into China last year, 48kt came from the DRC, making this Chinas most
concentrated supply risk in an individual commodity, helped by the distinct lack of Chinese
domestic resource. We reiterate our view that Chinese purchases of DRC assets (most
notably Tenke Fungurume) are with one eye on securing supply of cobalt units.
We see cobalt supply growing 2.9% this year, the lowest rate since 2012, with the swing up
and down from this number highly dependent on the performance of small-scale DRC supply.
The next leg of incremental supply growth is set to come from the restart of Glencores
copperbelt operations in 2018. However, the raw material overhang in China from 2015s
large import volumes has been partially run down.
While the DRC-China link clearly brings supply side risk, it also creates potential problems on
the demand side. Material buyers rarely feel comfortable being beholden to such
concentrated supply, and when coupled with concerns over the social and environmental
impact of DRC mining, there is certainly potential that cobalt is thrifted or engineered out of
battery technology, which remains the dominant end use of cobalt in the world.
This exposure to batteries has attracted a new source of cobalt demand in recent times
investors. With strong expectations of growth in rechargeable batteries and energy storage,
and cobalt having lagged lithium pricing significantly, it is viewed as the catch-up trade. And
with limited equity exposure options, this has often resulted in taking physical metal positions.
Fig 140 Cobalt prices are starting to gain momentum Fig 141 We see a sustained market deficit over the
as the industrial recovery matures coming years
$/t
99.8% cobalt prices Cobalt stocks vs price (RHS)
26 25 23.0
24 21.0
20 19.0
22
17.0
20 15
15.0
18 13.0
10
$/lb
16 11.0
14 5 9.0
7.0
12
0 5.0
10 2010 2012 2014 2016 2018F 2020F
Apr-10
Apr-11
Apr-12
Apr-13
Apr-14
Apr-15
Apr-16
Jan-10
Jan-11
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jul-10
Jul-11
Jul-12
Jul-13
Oct-10
Oct-11
Oct-12
Oct-13
Jul-14
Jul-15
Jul-16
Oct-14
Oct-15
Oct-16
Weeks of Wold Consumption Price (RHS)
Source: Metal Bulletin, Macquarie Research, January 2017 Source: LME, CDI, Macquarie Research, January 2017
19 January 2017 54
Fig 142 Cobalt supply and demand balance

tonnes 2015 2016 2017F 2018F 2019F 2020F 2021F
Total Demand 90,064 93,464 99,115 102,436 106,011 109,233 112,201

% change YoY 3.3% 3.8% 6.0% 3.4% 3.5% 3.0% 2.7%
Total Supply 92,349 96,755 99,601 101,551 102,806 103,893 105,007
% change YoY 6.8% 4.8% 2.9% 2.0% 1.2% 1.1% 1.1%
Balance 2,285 3,291 486 -885 -3,205 -5,340 -7,194
Stocks 31,123 34,414 34,900 34,015 30,810 25,470 18,277
Weeks of consumption 18.0 19.1 18.3 17.3 15.1 12.1 8.5
Source: CRU, CDI, Macquarie Research, January 2017
There is a difference in these battery materials however. While for lithium the growth focus is
new energy vehicles, cobalt is all about portable electronics LCO batteries, with cobalt
making up 60% of the cathode, dominate this segment. 2016 was not a good year for
demand in this area, with smartphone sales barely growing and tablets and laptops falling
~10% YoY. However, with the widening industrial recovery, production of these items does
look to have reaccelerated into year end. While we are cautious on cobalt use in NEV
batteries, Chinas recent announcement that subsidies would focus on more efficient
technologies does offer some upside. Should this lead to a replacement of the fast growing
(but low quality) LFP technology in the bus fleet, this would be to cobalts benefit.
Fig 143 Chinas cobalt unit imports have swung Fig 144 The DRC remains Chinas dominant source of
aggressively in recent times cobalt units
China's cobalt unit imports

70000 Source of China's imported cobalt
65000 60,000
60000
Imports
50,000
55000
tonnes, 6MMA
Consumption
40,000
50000
t Co
45000 30,000
40000
20,000
35000
10,000
30000
25000 0
June
May
June
May
Mar
Apr
May
Aug
Sep
Feb
Mar
Apr
Aug
Sep
Feb
May
June
Mar
Apr
Aug
Sep
Feb
Mar
Apr
May
Aug
Sep
Feb
Mar
Apr
May
Aug
Sep
Feb
Mar
Apr
Aug
Sep
Feb
May
Mar
Apr
Aug
Sep
Oct
Nov
Dec
Jan
July
Oct
Nov
Dec
Jan
July
Oct
Nov
Dec
July
Jan
Jan
Oct
Nov
Dec
Oct
Nov
Dec
Oct
Nov
Dec
Oct
Nov
Jun
Jul
Jan
Jun
Jul
Jan
Jun
Jul
Jan
Jun
Jul
Feb
2010 2011 2012 2013 2014 2015 2016e

2010 2011 2012 2013 2014 2015 2016
DRC ore DRC intermediate Other ore Other intermediate
Source: China Customs, Macquarie Research, January 2017 Source: CDI, Macquarie Research, January 2017
Fig 145 LCO batteries are most important for cobalt Fig 146 where shipment growth had been struggling
demand, and dominate portable electronics through 2016
Composition of major battery cathodes Portable electronics shipment growth - 9M16

YoY
100% 4%
90%
33% 35% 33% 2%
80% 36%
70% 0%
7% 4% 7%
60% 7% -2%
50% 96% 20% 9%
-4%
40%
30% 60% 61%
-6%
19%
48% -8%
20%
10% 20% -10%
0% 4%
LCO LMO LFP NMC NCA -12%
Ni Co Mn Li Other -14%
Smartphone Tablet Portable PC
Source: Avicenne, Macquarie Research, January 2017 Source: IDC, Macquarie Research, January 2017
19 January 2017 55
Lithium
The focus shifts to Australian supply
Lithium was certainly last years main commodity story, given the surge in pricing and
excitement around future demand growth. That demand story remains extremely robust
indeed the strongest among all the commodities covered in this report. However, the surge in
pricing has been met by a surge in projects from a pent-up mining industry looking for a
positive story. Lithium is unlikely to return to pre-boom norms, however the Australian supply
growth set to hit the market in 2017 should result in some downward pressure on spot prices.
It remains a good story, particularly from the demand side, but the period of maximum
excitement is now likely to be behind us.
In December we made a 6% upgrade to our 2017 price forecast due to a higher base for
2016 year-end pricing. New Australian supply volumes have not arrived as quickly as
anticipated, and hence the true test for the market will come in 2017. We have already seen
more volume from Chile and Argentina in recent times, and our base case remains that in
spite of a strong demand story that has played out in line with our expectation, this wont be
enough to absorb the big incremental supply additions.
The average Chinese import price of lithium carbonate was up 66% YoY over 2016, with
prices roughly doubling over the course of the year as the scramble for supply became
frenzied. However, this price now looks to be peaking out, with October marking the high
point. This follows the Chinese domestic spot quote, which peaked in Q2 16 before trending
lower. If we use domestic prices as a lead indicator to import prices (which suffer a contract
sales lag), they have since stabilised around 30% below peak but are significantly higher than
their pre-boom level. And hence, while we do expect some further downside in pricing
through 2017, compared to historical norms this will still mark a very good lithium price, and
one where brine assets in particular are highly profitable.
One issue central to lithium markets is the share being taken by electric vehicles of various
types. We calculate this was 1.33% in 2016, up from 1.05% in 2015 (Fig 146). This is a slow
pace of growth, and was despite another surge in China which took the (estimated) share of
EVs in December to over 2%. Excluding China the global share of EVs in 2016 remained
under 1% in the markets we cover (which means it is likely to be lower globally given we
focus on markets which have an established EV presence). Nevertheless ignoring the year-
end spikes, which tend to be due to expiring incentive schemes, the trend is clearly upwards,
and it could strengthen in 2017 following on from a number of high profile EV launches such
as the Chevrolet Bolt and Tesla Model 3.
We see lithium battery demand rising at 17% CAGR through 2021, with EVs being the main
driver as portable electronic growth stalls. This takes battery demand to 141kt of lithium
carbonate equivalent by 2021, accounting for ~50% of global lithium demand at that point.
Fig 147 After the surge, Chinese customs values for Fig 148 while the Chinese domestic spot market is
lithium carbonate imports look to be peaking now stabilising higher than the pre-boom level
China lithium carbonate CIF import price China spot lithium carbonate
US$/t US$/t
12,500
30,000
11,500
10,500 25,000
9,500 20,000
8,500
15,000
7,500
6,500 10,000
5,500
5,000
4,500
3,500 0
Jan-17
Jan-14
Jul-14
Jan-16
Jan-15
Jul-15
Jul-16
2,500
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Source: China Customs, Macquarie Research, January 2017 Source: SMM, Macquarie Research, January 2017
19 January 2017 56
We see that 2017 will in many ways be an acid test for the lithium market. For the past few
years, the market has been highly consolidated with producers maintaining supply discipline
in an effort to control the price. However, Galaxy and Mineral Resources are currently
ramping up their spodumene assets or Mt Cattlin and Mt Marion respectively, which plan to
add a combined 74kt LCE equivalent over the next two years. This equates to roughly 40%
of the current market size, and would significantly dilute the market share of incumbents. We
have seen this script before in other commodity markets, and it generally ends up with a
market share battle of some form. As such, while for lithium last year the focus was on strong
demand growth, for 2017 the delivery of Australian projects is set to be the central topic.
Fig 149 Expected aggressive ramp-ups from new Fig 150 present a major challenge to the market
Australian operations share of incumbent producers
Estimated Australian lithium supply by producer Supply share of Talison, SQM, ALB and
kt LCE FMC
95%
70
90%
60
85%
50
80%
40 Mineral Resources 75%
30 Galaxy Mt Cattlin 70%
20 Talison 65%
10 60%
0 55%
2017F
2011
2012
2013
2014
2015
2016
2018F
50%
2014 2015 2016 2017F 2018F 2019F 2020F 2021F
Source: Company Data, Macquarie Research, January 2017 Source: Company Data, Macquarie Research, January 2017
Fig 151 Forecast lithium demand growth in batteries Fig 152 Electric vehicle share of main auto markets
2015-2025F battery cathode material 2.5%
growth (CAGR)
16%
2.0% EV PHEV
14%
12%
1.5% 2016 = 1.3%
10%
8% 1.0% 2015 = 1.1%
6%
2014 = 0.5%
4% 0.5%
2%
0% 0.0%
LCO NMC NCA LMO LFP Jan-14 Jul-14 Jan-15 Jul-15 Jan-16 Jul-16
Source: Signumbox, Avicenne, Macquarie Research, January 2017 Source: National data, Macquarie Research, January 2017
Fig 153 Global lithium supply and demand (tonnes LCE)

Demand 2015 2016 2017F 2018F 2019F 2020F 2021F
Batteries 55,325 63,442 76,079 89,759 105,066 121,744 140,716
Non-battery 118,511 121,737 125,495 129,443 133,393 137,416 141,559
Total 173,836 185,179 201,573 219,202 238,459 259,160 282,275
YoY 9% 7% 9% 9% 9% 9% 9%
Supply 2015 2016 2017F 2018F 2019F 2020F 2021F
Argentina 19,210 30,024 32,169 34,828 35,299 35,781 36,276
Australia 55,079 69,772 124,081 134,072 134,072 134,072 134,072
Chile 55,735 63,098 67,512 72,483 77,002 79,713 79,713
Other 24,065 24,370 24,720 24,970 25,220 25,470 25,720
Total 154,088 187,264 248,481 266,352 271,592 275,036 275,780
YoY -4% 22% 33% 7% 2% 1% 0%
Additional supply required 19,748 -2,086 -46,908 -47,150 -33,133 -15,876 6,495
Spare capacity at existing assets 42,993 61,045 52,886 47,415 43,933 43,438
Planned from most likely projects 0 0 43,884 100,377 138,900 152,900
Source Customs Data, Company Data, Macquarie Research, January 2017
19 January 2017 57
Bauxite & Alumina

A big alumina deficit, but abundant bauxite
Alumina remains a commodity where we have medium-term concerns, given it has plentiful
raw material supply and is one of the few seeing ex-China projects built. For 2017-18,
however, everything is fine, provided Chinese aluminium production continues to grow. Over
the past few months Chinese alumina restarts have underperformed aluminium owing to
enhanced environmental regulation, enabling a strong price trajectory. Meanwhile, unlike
aluminium, ex-China output is below levels seen two years ago. While US restarts may ease
this in time, we are still looking at an alumina market in deficit for 2017 and 2018.
In our view, the Chinese import arbitrage is very telling for the global alumina market. For
January through September 2016, as aluminium production growth stalled, this was shut.
From September onwards however the Chinese price has been trending, almost reluctantly
dragging international pricing higher such that the arbitrage remains open. The alumina
market has seen many positive elements, but recent stability suggests prices are topping out.
Alumina demand has been aided by the aluminium production growth we are now seeing.
We model 5.5mt of additional alumina demand this year. With ex-China output having fallen
1.6% YoY in 2016 and set to fall a further 2.2% this year, we now have a 4.0mt deficit for
2017, but smaller for 2018. Of course, we do expect a Chinese domestic alumina production
response. China has plenty of capacity, and the economics should now be encouraging
restarts. However, these restarts are being hindered by more stringent environmental
checks, something not uncommon in China at the present time. Given domestic production
makes up over 95% of Chinas needs, the response here is crucial.
Longer term, we reiterate that aluminas structural problem is being the only commodity
where there is significant ex-China supply still coming to market, mainly from those currently
short alumina in India and the Middle East. This becomes pertinent from mid-2018 onwards,
and we forecast pricing of <$240/t FOB Australia by the end of the decade.
Fig 154 Bauxite prices have done very little over the Fig 155 We see a recovery in Chinese bauxite imports
past year, while the alumina constraint is clear over the coming years, given global supply growth
80 Bauxite and Alumina (RHS) prices, $/t CFR China 500 Chinese bauxite imports
80
75
450
70 3
70 5
400 60 25
65 Atlantic
4
50
60 350 5 Malaysia
40 16 3
49 24 India
55 5
300 3
30 3 8 Indonesia
50 5
8 5 Australia
250 20 9
35
45
10 20 21
14 16
40 200
Jun 12
Jun 14
Jun 15
Sep 15
Sep 12
Dec 12
Mar 13
Jun 13
Sep 13
Dec 13
Mar 14
Sep 14
Dec 14
Dec 15
Mar 16
Jun 16
Mar 15
Sep 16
Dec 16
0
2013 2014 2015 2016f 2020f
Source: C&M, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
Fig 156 Macquarie global alumina market balance

million tonnes 2014 2015 2016 2017F 2018F 2019F 2020F 2021F
World Consumption 106.8 112.8 116.0 121.6 125.8 130.9 135.4 135.5
% Change Y-o-Y 7.1% 5.5% 2.9% 4.7% 3.5% 4.0% 3.4% 0.1%
of which: China 56.6 61.6 64.0 68.7 71.2 75.4 78.1 78.1
World Production 106.6 112.9 113.3 117.6 124.0 131.9 139.2 141.9
% Change Y-o-Y 3.1% 5.9% 0.4% 3.7% 5.5% 6.4% 5.5% 1.9%
of which: China 51.7 57.5 58.9 64.3 67.5 70.7 73.6 74.0
Global balance -0.3 0.1 -2.7 -4.0 -1.8 1.0 3.8 6.4
Source: CRU, IAI, Macquarie Research, January 2017
19 January 2017 58
For bauxite however, despite the ongoing shifts in the supply side, pricing remains dull. While
decent growth in traded volumes is likely over the coming years, this seems set to be driven
by excess supply growth more than anything else. While the return of Indonesian supply
following the easing of export restrictions is not likely to shake up the market too much, this
says more about how comfortable bauxite buyers feel with the incumbent supply side. We
expect medium-term bauxite pricing of $45-55/t CFR China, with the low end tested in 2019.
Prior to the 2014 Indonesian ban 10% of global supply, almost all in China relied on
Indonesian bauxite. China has successfully sourced bauxite alternatives, and has had less
reliance on Indonesian investment, but Hongqiao has ramped up a 1mt alumina refinery in
Indonesia over 2016 which is now shipping at full capacity.
The new regulation is said to have a minimum alumina content of 42% for any bauxite
shipments. On average Indonesian material just about meets this, however, quite frankly it
just isnt as desirable in todays market, particularly as its main purchaser prior to the ban,
Hongqiao, now has alternative bauxite supply from Guinea and alumina from Indonesia it can
rely on. Most likely, any Indonesian material would fight with Malaysian bauxite for market
share into China at the low end of the quality spectrum. Impact on bauxite/alumina pricing
would be limited in this case, but in the medium term it may well cause reviews of Rio Tinto
and Alcoas plans to sell large bauxite quantities into China. Our base case assumes 70mt of
Chinese imports in 2020 from ~50mt currently.
Fig 157 Ex-China is the large delta in the global Fig 158 and while the current market is tight, we
alumina balance see production growth in India and the Middle East
Alumina supply-demand balance Changes in alumina output, 2020f vs. 2016 (kt)
12 6000
China Ex-China Global 5000
10
4000
8
3000
6 2000
million tonnes
4 1000
2 0
-1000
0
-2000
-2
-3000
Japan
Other Asia
Europe
Middle East
Africa
Oceania
India
CIS
-4
South America
North America
-6
2016E
2017F
2018F
2019F
2020F
2021F
2013
2014
2015
Source: CRU, IAI, Macquarie Research, January 2017 Source: IAI, Macquarie Research, January 2017
Fig 159 The alumina price move has been led by China Fig 160 though imports remain a very small part of
domestic, opening the import arbitrage overall Chinese alumina consumption
$/t 20 Alumina supply to China, quarterly
Chinese alumina import arbitrage
50 18
40 16 Imports
30 Production
14
20
Million tonnes
12
10
10
0
8
-10
6
-20
-30 4
-40 2
-50 0
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Jul-13
Jul-14
Jul-15
Jul-16
Mar-13
Mar-14
Mar-15
Mar-16
Sep-13
Nov-13
Sep-14
Nov-14
Sep-15
Nov-15
Sep-16
Nov-16
May-13
May-14
May-15
May-16
Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3
09 09 10 10 11 11 12 12 13 13 14 14 15 15 16 16
Source: S&P Global, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
19 January 2017 59
Steel
Starting the drive down a long road to sustainable improvement
Steel has always been cyclical, but this market is now emerging as the most amplified play on
global industrial cycles. This comes from being a market operating well below its potential
capacity, where producers are under perennial pressure to make the most of any cyclical
opportunity. On the upside, the early part of 2017 looks set to continue with a steel-friendly
industrial recovery. In the short term we still feel there is some price upside post Chinese
New Year. And yet major challenges lie ahead, not least an expected slip back to YoY
demand decline into H2 2017 and 2018. Meanwhile, the steel industry remains at the
forefront of the global protectionist drive, further delaying the much needed capacity reset.
Steelmakers did their best to offset it, but the Q4 16 raw material price rises resulted in Q4
margin compression. However, their inability to align raw material costs and finished goods
prices will likely benefit them in the current quarter, as steel prices gain further (given they lag
the cycle) and raw materials continue their journey back to cost curve normality. Over the
longer term, steel margins remain hampered by semi-permanent low capacity utilisation and
we dont expect a sustained recovery to longer cycle norms for several years.
Clearly Chinese steel demand performed better than we expected over 2016, rising 2.9%
YoY to 710mt. The upgrade in demand has largely been driven by residential property, but as
the economic recovery has become broader-based as evidenced by the recovery in flat-
product intensive sectors such as machinery and autos in our recent proprietary surveys,
other areas also contributed. Meanwhile, there was a clear drop-off in Chinese steel exports
in 2H16. This has been driven partially by stronger steel demand and prices in China, but we
have long been writing that we believed ex-China developing markets had already been
saturated by Chinese steel exports since last year. For 2017 we expect Chinese steel
exports to decline again, from 106mt to 99mt, but then we expect exports to stabilise around
those levels from 2018 onwards.
For 2017 we forecast Chinese steel consumption to rise by 1.5%, despite the higher base in
2016. We continue to look for steel consumption in China to make a strong start to 2017 on
follow-through spending in infrastructure and property post Chinese New Year, and the
ongoing demand strength in consumer-related sectors. Beyond 1H17, we continue to expect
Chinese steel consumption to resume the declining trend seen over 2014/15. We believe the
current infrastructure boom has resulted in front-loading of the five-year plan and hence
infrastructure steel consumption will decline from 2H17. In property we continue to worry that
the slowing pace of urbanisation combined with the demographic profile does not support a
structural recovery in housing construction, and we expect floor space construction volume to
resume its downtrend from 2H17.
Fig 161 Steel prices continue to benefit from the Fig 162 We are past the peak of steel overcapacity as
expectation of healthy global reflation Chinese closures continue, but it remains high
800
Weekly steel prices 300 Global steel overcapacity
Chinese HRC 250

700 US HRC
German HRC
200
million tonnes
600
150
$/tonne
500 100
50
400
0
300
-50
200 -100
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018f
2020f
Jul-14
Jul-15
Jul-16
Jan-14
May-14
Mar-14
Jan-15
May-15
Mar-15
Jan-16
May-16
Jan-17
Sep-14
Nov-14
Sep-15
Nov-15
Mar-16
Sep-16
Nov-16
Source: CRU, Macquarie Research, January 2017 Source: worldsteel, Macquarie Research, January 2017
19 January 2017 60
With the recent closure of illegal induction furnaces, the government-led capacity clampdown
in China is a hot topic again. For 2016~2020 as a whole, Chinas central government targets
closure of 100~150mtpa of steelmaking capacity, though some reports suggest this timeline
may be reduced to three years, and the last two years will be focused on promoting industry
consolidation. This would leave a 40~90mtpa of capacity to be undertaken over the coming
two years. This will certainly help to reduce the global capacity overhang, but a global
utilisation level where steelmakers have sustainable pricing power looks many years away.
Fig 163 Global steel consumption has been Fig 164 Our recent surveys have shown a distinct
recovering strongly, helped by a low base pick-up in Chinese orders in HRC-heavy sectors
15%
Global steel consumption
10%
YoY change
5%
0%
-5%
-10%
2011 2012 2013 2014 2015 2016
Source: worldsteel, Macquarie Research, January 2017 Source: Macquarie China Steel Sector Survey, January 2017
Fig 165 Steels next challenge comes from the middle Fig 166 China still has a long way to go to catch the US
of this year, when we expect a negative demand trend in steel capital stock, but is catching up quickly
mn tonnes
China residential real estate steel demand
180
Steel capital stock, kg/capita
16,000
USA
160
14,000 China
140
12,000
120
10,000
100
8,000
80
6,000
60
4,000
40
20 2,000
0 -
1950
1954
1958
1962
1966
1970
1974
1978
1982
1986
1990
1994
1998
2002
2006
2010
2014
2030f
2018f
2022f
2026f
2034f
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017F
2018F
2019F
2020F
2021F
Source: CISA, Macquarie Research, January 2017 Source: worldsteel, Customs Data, Macquarie Research, January 2017
Fig 167 Global steel supply and demand

million tonnes 2013 2014 2015 2016 2017f 2018f 2019f 2020f 2021f
China 762 738 691 710 721 704 698 692 685
ex-China 884 924 907 901 913 927 941 956 970
World Consumption 1646 1662 1598 1612 1633 1631 1639 1648 1654
% Change YoY 6.5% 1.0% -3.9% 0.8% 1.4% -0.2% 0.5% 0.6% 0.4%
World Production 1650 1667 1621 1618 1643 1639 1634 1638 1636
% Change YoY 5.8% 1.0% -2.8% -0.2% 1.6% -0.3% -0.3% 0.2% -0.1%
China share 50% 49% 50% 50% 50% 49% 49% 49% 48%
Balance 4 5 23 6 10 8 -4 -10 -18
Source: worldsteel, Customs Statistics, Macquarie Research, January 2017
19 January 2017 61
Iron ore
How long can prices hold up amid abundant supply?
Spot iron-ore prices surged in 4Q16 to average $70.2/t, reaching highs not seen since late
2014. Supply has clearly responded quickly to these high prices, and while demand has been
stronger than we expected prior to 2H16, we believe supply is now running ahead of demand
as evidenced by surging iron ore import volumes, rising port inventory and rising iron ore
inventory at steel mills, which in days cover terms is at its highest since September 2015.
Chinese steel demand in 2016 performed better than we expected at the start of the year,
and we recently updated our steel demand model to give 2016 steel consumption growth of
2.9% YoY. For 2017 we forecast Chinese steel consumption to rise by 1.5%, despite the
higher base in 2016, on follow through spending in infrastructure and property and ongoing
demand strength in consumer related sectors. The key risks to our forecasts come from a
more rapid slowdown in Chinas property market, or an earlier tightening of macro policy. Into
2H17 however we continue to expect Chinese steel consumption to resume its declining trend
seen over 2014/15. We believe the current infrastructure boom has resulted in front loading of
the five year plan and hence infrastructure steel consumption will decline from 2H17. In
property we worry that the slowing pace of urbanisation combined with the demographic
profile does not support a structurally sustainable recovery in construction.
Fig 168 Iron ore price rebound has lasted much longer Fig 169 Resulting in a clear supply response outside
than many expected of the larger suppliers
62% Fe 65% Fe 000 tonnes Chinese iron ore imports by country for non-majors
180
58% Fe 52% Fe 25,000 India Canada Peru
160 Iran Mauritania Chile
20,000 Ukraine Sierra Leone Others

TSI iron ore prices, US$/t
140
120
15,000
100
10,000
80
60 5,000
40
0
Sep-14
Sep-15
Sep-16
Nov-14
Nov-15
Nov-16
Jan-14
May-14
Jan-15
May-15
Jan-16
May-16
Mar-14
Jul-14
Mar-15
Jul-15
Mar-16
Jul-16
20
Oct-12
Oct-13
Oct-14
Oct-15
Oct-16
Apr-12
Apr-13
Apr-14
Apr-15
Apr-16
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jan-17
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
Source: TSI, Mysteel, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
Fig 170 Iron ore port inventory has risen strongly Fig 171 Steel mill ore inventory levels also higher
mn tonnes
Iron ore port stock at 31 ports
120 40
Mill iron ore inventory, days of use
110 35
100 30
90 25
80 20
70 15
60 10
Apr-12
Oct-12
Apr-13
Oct-13
Apr-14
Oct-14
Apr-15
Oct-15
Apr-16
Oct-16
Jan-12
Jan-13
Jan-14
Jan-15
Jan-16
Jul-12
Jul-13
Jul-14
Jul-15
Jul-16
Dec-16
Oct-13
Dec-13
Oct-14
Dec-14
Oct-15
Dec-15
Oct-16
Apr-13
Apr-16
Jun-13
Aug-13
Apr-14
Aug-14
Apr-15
Aug-15
Aug-16
Jun-14
Jun-15
Jun-16
Feb-13
Feb-14
Feb-15
Feb-16
Source: Port Data, Mysteel, Macquarie Research, January 2017 Source: Mysteel, Macquarie Research, January 2017
19 January 2017 62
In terms of iron ore supply, 2016 seaborne volumes came in 70mt ahead of our forecasts
made at the start of the year, as iron ore supply has yet again proven itself to be highly price
elastic. Domestic Chinese iron ore mines also saw a strong recovery in output in 2H16 in
response to the higher prices, with operating rates according to SMM and Mysteel back at
levels not seen since early 2015.
For 2017 however we expect seaborne supply will see little increase. We forecast only a 19mt
increase in supply from the big four producers, mainly coming from Vales new S11D project,
and a further 21mt from the mid-tier producers, as with 2016 mainly led by the continuing
ramp up in Roy Hill. Nonetheless, with our steel demand forecast giving a 17mt increase in
global iron we still see a declining need for iron ore from the higher cost seaborne suppliers
who have returned to the market in 2016, together with a decline in iron ore needed from
domestic Chinese mines, and this implies that iron ore prices should fall.
Based on our average price forecast for 2017 of $54/t, higher cost supply sources should see
a decline in exports, but if prices stayed at say $70/t, we believe this could incentivise at least
a further 30mt from the smaller seaborne suppliers. Pulling together our seaborne supply
forecasts and Chinese demand forecasts means we see a net need for domestic Chinese
iron ore output of only 185mt in 2017, down from 220mt in 2016. Chinese mine costs in US$
terms are down around 10% YoY given the recent depreciation in the RMB, and the resource
tax cut seen since July 1st. According to SMMs most recent domestic mine cost curve, they
see 175mt of domestic iron ore output at $50/t, rising to 217mt at $55/t, and should the price
be at $60/t through 2017 they forecast domestic mine supply would be 254mt.
2018 meanwhile continues to look challenging for the iron ore market given the final leg
higher in seaborne supply expected from Vale (S11D), BHP Billiton (as their rail repairs
complete by end of 2017) and potentially the restart of the Samarco operations in Brazil.
Fig 172 The 2018 market balance remains challenging Fig 173 precipitating a need for ex-major supply to
as steel production heads negative drop sharply
mt Annual change in iron ore demand Seaborne supply required ex-majors, mtpa
70 250
Annual change in seaborne iron ore supply
60
50 200
40
30 150
20
10
100
0
-10
50
-20
-30
0
-40
1Q15
1Q13
3Q13
1Q14
3Q14
3Q15
1Q16
3Q16
1Q17F
3Q17F
1Q18F
3Q18F
1Q19F
3Q19F
1Q20F
3Q20F
1Q21F
3Q21F
2016F 2017F 2018F 2019F 2020F 2021F
Source: Customs Data, Macquarie Research, January 2017 Source: Customs Data, Macquarie Research, January 2017
Fig 174 Iron ore market supply-demand

million tonnes 2013 2014 2015 2016 2017F 2018F 2019F 2020F 2021F
ex-China demand 392 424 429 420 427 432 425 426 425
China demand 1246 1261 1222 1236 1243 1208 1191 1177 1155
Total iron ore demand (China + ex-China) 1639 1685 1650 1657 1670 1640 1616 1603 1580
YoY 162 46 -35 6 14 -30 -24 -13 -23
Seaborne iron ore supply 2013 2014 2015F 2016F 2017F 2018F 2019F 2020F 2021F
Total majors 811 959 1044 1068 1087 1135 1143 1145 1145
YoY 71 148 85 24 19 47 8 2 0
Chinese domestic 409 323 255 220 183 133 137 144 142
Non-major seaborne 454 445 383 418 423 413 423 417 401
YoY delta (displacement actual) 90 -102 -119 -18 -6 -77 -32 -15 -23
Source: Customs Statistics, worldsteel, Macquarie Research, January 2017
19 January 2017 63
Manganese
What goes up..
If there is still doubt that the 2016 commodity rally was fundamentally driven, you need look
no further than manganese. This is a market without any financial contract which might cause
distortion, and yet still matched or exceeded steel making raw material peers in terms of price
action, firstly in ore and more recently in manganese alloys. There were simple reasons for
this - manganese saw the most severe cuts to existing supply of all bulk commodities at the
bottom of the cycle, while steel production came in ahead of expectations.
2017 started with manganese prices at a level second only to 2008s peak, both in USD and
Rand terms. Manganese ore prices ripped almost 400% higher in 2016 on strong supply cuts,
taking mine output down ~10% YoY on our estimates, the largest cuts of any commodity we
cover. Add in a recovery in Chinese (and global) steel production rates plus strong growth in
200-series stainless production (with high manganese content) and a squeeze higher was
likely, even if the amplified nature of the move was a surprise.
However, high prices have a tendency to cure high prices. Such a position, trading over
100% out of the cost curve, was never likely to last and since the start of this year 44% Mn
ore prices have dropped 15%. In the near term, the seasonal slowdown in Chinese steel
production is the key driver, with mills adequately stocked through Chinese New Year.
Meanwhile, pricing at the levels will tend to attract a supply response. To be clear, current
pricing and that we expect over the first half of this year is still decent, however the sequential
decline represents a return to price normality.
Fig 175 The dramatic manganese ore price rise has Fig 176 Ferromanganese prices have finally reacted in
started to reverse recent months to the ore gains
Mn ore spot prices Manganese ore versus ferromanganese
10.0
160 pricing (indexed 2009=100)
9.0 44% CIF China
37% FOB SAF 140 FeMn
8.0
7.0 120 Mn ore
US$/dmtu
6.0 100
5.0 80
4.0 60
3.0 40
2.0
20
1.0
2013 2014 2015 2016 2017
0
2009 2010 2011 2012 2013 2014 2015 2016
Source: Metal Bulletin, Macquarie Research, January 2017 Source: S&P Global, Macquarie Research, January 2017
Fig 177 South African mine supply is no longer Fig 178 Chinese imports of ore have been stagnant
growing, even for low grade material over recent years
South African Mn ore production by grade China's Mn ore imports by country
16,000 18
40-45% Mn 30-40% Mn >45% Mn ROW MMR
14,000 16
BRA GAB
m tonnes gross weight
'000t annualised GW
12,000 14 ZAF AUS

12
10,000
10
8,000
8
6,000 6
4,000 4
2,000 2
0
0
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Jan-12 Nov-12 Sep-13 Jul-14 May-15 Mar-16
Source: DTI, Macquarie Research, January 2017 Source: Customs data, Macquarie Research, January 2017
19 January 2017 64
Fig 179 200-series stainless steel continues to be a Fig 180 We expect manganese ore pricing to return to
driver of demand growth a lower norm, with 2018 hurt by negative steel demand
Mn usage in stainless steel output Macquarie manganese ore price forecast - USD
USD/dmtu vs. ZAR ZAR/dmtu
1,600
8 80
1,400 200S 300S & 400S US$/dmtu
ZAR/dmtu (RHS)
1,200 7 70
'000 tonnes Mn
1,000 6 60
800
5 50
600
400 4 40
200
3 30
0
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018F
2020F
2 20
2017F
2018F
2019F
2009
2010
2011
2012
2013
2014
2015
2016
Source: ISSF, IMnI, Macquarie Research, January 2017 Source: Customs data, Macquarie Research, January 2017
On the supply side, we do expect a moderate reaction from flexible tonnes, most notably
Chinese domestic material. However, should the spot price falls be aggressive, a restart at
the Chinese-owned Woodie Woodie operation in Australia (~4% of global mine capacity)
would be unlikely. There are still structural problems with supply, with all recent growth in
South Africa being lower grade material and, even then, output struggling despite the high
Rand price. Indeed, over the past four years Chinese manganese ore imports have
essentially been flat, with supply from key countries like Australia and Gabon in decline.
However, the bigger question is whether a sustained supply response is actually needed.
Manganese demand will always be tied to wider steel cycles, and our expectation remains
that global steel production will return to negative YoY growth in H2 2017. As such, we would
expect pricing to be at a level which at least keeps supply offline or even attempts to force
further supply cuts. Given our balanced market expectation, we expect prices to return to a
cost curve normality level of $3.5-4/mtu CIF China.
One positive area for manganese demand remains stainless steel, where 2016 saw a
complete reversal of 2015s demand slump. Our latest stainless steel production forecasts
still see a healthy ~3% compound growth rate to 2020. However, we see this growth being
driven by 300- and 400-series stainless steel, typically containing around 1% manganese,
rather than 200-series stainless steel, which tends to contain 615% manganese.
Fig 181 Global manganese ore market balance (m tonnes Mn contained, unless stated)
2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
Crude steel production, mt 1,624 1,579 1,572 1,596 1,590 1,585 1,587 1,584
% Change y-o-y 1% -3% -0% 2% -0% -0% 0% -0%
Stainless steel production, mt 42.9 42.2 45.7 47.0 48.3 49.5 50.8 52.1
% Change y-o-y 9% -2% 8% 3% 3% 2% 3% 2%
Mn ore consumption 18.5 17.9 18.1 18.3 18.3 18.3 18.4 18.4
% Change y-o-y 1% -3% 1% 2% -0% 0% 0% 0%
-Mn alloys for crude steel 15.9 15.4 15.3 15.5 15.4 15.3 15.3 15.2
-EMM for stainless steel 1.7 1.6 1.8 1.9 1.9 1.9 2.0 2.0
-Other 0.9 0.9 1.0 1.0 1.1 1.1 1.2 1.2
Mn ore mine production 19.8 19.7 17.8 18.4 18.5 18.4 18.2 18.1
% Change y-o-y 4% -0% -10% 3% 0% -1% -1% -1%
Global Mn ore balance 1.3 1.8 -0.3 0.1 0.1 0.0 -0.2 -0.3
Source: IMnI, worldsteel, ISSF, CRU, Macquarie Research, January 2017
19 January 2017 65
Metallurgical Coal
Prices come tumbling down, but where will they find a floor?
Spot met coal prices have endured a rollercoaster ride over the last twelve months, from lows
of only $74/t in February 2016 to highs above $310/t in November, and back down below
$180/t again by mid-January. The main driver of the surge in prices was Chinese production
cuts forced by the government under the 276 days policy, with met coal supply essentially
being a casualty of the governments efforts to support thermal coal prices. Buyers in China
became panicked over potential shortages of met coal in October, but with the government
having suspended the 276 days policy since November, supply has returned quickly.
On the seaborne market meanwhile supply disruptions over 2H16, which coincided with
production cuts in China and added to buyers fears of shortages, have now eased.
Additionally while rainfall levels in Australia have been significant supply disruption from
weather impacts appears less than feared, while the high prices over 2H16 have provided a
strong margin incentive for miners to lift output, bringing more cargoes into the market from
marginal suppliers in the US and Mongolia especially.
Seaborne spot prices are now trading at a significant discount to the 1Q17 HCC contract
price of $285/t. As such there is a clear incentive for steel mills to defer cargoes into 2Q on
expectation of lower prices, as we wrote about recently. If this becomes widespread, it may
lead to miners offering more spot cargoes in the short term, further pressuring prices.
Fig 182 Coking coal prices have been very volatile Fig 183 Quarterly HCC contract prices a lag to spot
Met coal spot price benchmarks Premium hard coking coal settlements
350
350
Premium HCC
330
300
315
300 Low Vol PCI

$/t, FOB Australia
285
285
$/t, FOB Australia
250
Semi-soft
235
250 200
225
225
225
210
209
200
200
150
172
170
200
165
152
145
143
100 120
120
119
117
110
150
93
93
89
84
81
50
100 0
4Q10
1Q13
2Q15
1Q17
2Q10
3Q10
1Q11
2Q11
3Q11
4Q11
1Q12
2Q12
3Q12
4Q12
2Q13
3Q13
4Q13
1Q14
2Q14
3Q14
4Q14
1Q15
3Q15
4Q15
1Q16
2Q16
3Q16
4Q16
50
Jul 10 Jul 11 Jul 12 Jul 13 Jul 14 Jul 15 Jul 16
Source: Platts, Macquarie Research, January 2017 Source: TEX Report, Macquarie Research, January 2017
Fig 184 Coking plants inventory has jumped Fig 185 Steel mill inventory also on the rise
24 170
Independent coking plants' coking coal inventory, days of use Steel mills coking plants' coking coal inventory, kt
22 160
150
20
140
18
130
16
120
14
110
12
100
10 90
8 80
Oct-12
Oct-13
Oct-14
Oct-15
Oct-16
Oct-16
Oct-16
Nov-16
Nov-16
Dec-16
Dec-16
Dec-16
Apr-12
Jul-12
Apr-13
Jul-13
Apr-14
Jul-14
Jan-15
Apr-15
Jul-15
Apr-16
Jul-16
Jan-17
Jul-16
Jul-16
Jul-16
Aug-16
Aug-16
Sep-16
Sep-16
Jan-12
Jan-13
Jan-14
Jan-16
Jan-17
Source: Mysteel, Macquarie Research, January 2017 Source: Mysteel, Macquarie Research, January 2017
19 January 2017 66
Seaborne met coal prices have now fallen to a level which is starting to attract interest from
Chinese steel mills given that seaborne met coal prices have now fallen below Chinese
domestic prices for the first time since 2Q16, making imports more attractive. Additionally
Chinese contract prices for HCC are equivalent to $175/t FOB, which should provide
something of a floor for seaborne spot prices near term as Chinese buying interest should
become very active if prices fall below that level.
Beyond Chinese New Year, the key to the future direction of met coal prices is likely to be the
direction of thermal coal prices in China and their consequent influence on whether the
Chinese government looks to bring back the 276 days policy for some mines, or leaves the
policy suspended as it is currently. In November the Chinese government postponed the 276
days policy until the end of heating season (late March/early April depending on the
weather). Chinese government policy is very focused on thermal coal prices, with the impact
on met coal being much less of a focus for policymakers.
As Chinese policy making becomes clearer post CNY, mills could begin to become nervous
again about potential production constraints in China, which may then result in a further
increase in buying appetite. Nonetheless met coal prices clearly remain well above marginal
cost, and supply is responding accordingly in marginal producing countries such as Mongolia
and the USA. Overall wed thus expect to see prices continue in a downward trajectory over
the medium term, albeit with somewhat less volatility in met coal markets in the year ahead
than has been seen over the past twelve months.
Fig 186 Coking coal supply finally recovered in 4Q Fig 187 Mongolia has been a key source of met coal
China saleable coking coal production rate kt
Mtpa China coking coal import from other countries
8000
750 China coking coal impor from Mongolia
7000
700 6000
650 5000
4000
600
3000
550 2000
1000
500
0
450
01/2014
03/2014
05/2014
07/2014
09/2014
11/2014
01/2015
03/2015
05/2015
07/2015
09/2015
11/2015
01/2016
03/2016
05/2016
07/2016
09/2016
11/2016
2012 2013 2014 2015 2016
Source: Sxcoal, Macquarie Research, January 2017 Source: Customs data, Macquarie Research, January 2017
Fig 188 Seaborne met coal supply and demand

Met coal demand (mt) 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
Europe 56 54 48 50 50 50 51 51
India 47 51 50 56 58 61 63 65
Japan 55 55 55 55 55 52 52 52
Korea and Taiwan 43 44 44 46 47 47 47 47
Other 30 30 30 31 31 31 31 31
Total Ex-China 232 233 227 237 242 242 244 247
China 62 46 50 52 51 51 48 46
Total 294 280 277 290 293 293 293 293
YoY 2% -5% -1% 5% 1% 0% 0% 0%
Met coal supply (mt)

Australia 186 185 186 189 189 189 189 189
Canada 30 27 27 29 31 31 31 31
USA 51 38 32 37 35 35 35 35
Other 26 29 32 35 38 38 38 38
Total 294 280 277 290 293 293 293 293
Source: Customs data, Macquarie Research, January 2017
19 January 2017 67
Gold
Medium-term outlook strong, but near-term concerns
The gold market hasnt entirely moved on from obsessing with US monetary policy, but it now
has US fiscal policy to think about too. The dollar connects the two, and while 2017 has
begun much like 2016, with dollar bullishness being questioned and gold moving higher, we
think the US currency will have another leg higher, keeping gold under pressure. But we are
fundamentally bullish, expecting large price gains in 2H and 2018, with a host of reasons -
political, economic and fundamental - why gold should remain in demand.
Nothing suffers mood swings like the gold market. For much of 2016 optimism abounded as
the price rallied hard on the largest ETF inflows and highest net spec length on record. This
was brought crashing to a halt by the US election, after which it suffered five weekly price falls
in a row, taking it from over $1,300/oz to just above $1,120/oz. ETF holdings slumped by 219t
in November/December, meaning a full-year total of 532t, behind 2009s 644t. Futures spec
net length shed 700t from this peak. But as we start 2017, things are looking OK again, with
the price heading back above $1,200/oz.
Fig 189 Gold price, $/oz, and high/low Fig 190 Gold investment, tonnes v gold price, $/oz
Source: Bloomberg, Macquarie Research, January 2017 Source: CFTC, Company data, Macquarie Research, January 2017
These swings reflect a great deal of uncertainty about the health of the global and in
particular US economies. The seismic events that drove gold higher - a Chinese hard landing,
European and US political upsets - didnt derail the world economy as expected. In particular
the last of those, Trump, was soon interpreted as a positive for the US, as investors
anticipated more fiscal easing. This gave the Federal Reserve the green light to raise interest
rates, as it duly did in December, and confidence to increase its forecast for anticipated rate
hikes this year to three from two. This pushed the dollar higher and saw real interest rate
expectations (as shown by US inflation-linked bonds, whose price fell - see fig 185 overleaf)
shift higher. Add in a slump in Indian demand, and by December we were puzzled why the
gold price was so high.
The start of 2017 has brought a reality check as investors fret over whether the economic
situation has really changed or whether Trump will really be so positive. The dollar and US
real rates have given up much of their gains. So far the bounce has been quite modest. But it
naturally invokes memories of what happened at the start of 2016. Then the Fed had just
raised rates for the only other time since the GFC and promised four more, sending gold to a
post GFC low. But soon it became clear this was not going to be achieved.
We think it will be different this year. The world and US economies are in far better shape
than they were at the start of 2016, lowering the chance of an immediate setback. If there is,
the US Congress will presumably be more amenable to a fiscal stimulus to keep things going,
even if it is not the done deal it seemed in December. This points to the dollar rally having a
bit more to go, and we cannot rule out a border adjustment tax meaning rather more than a
bit. We expect golds rally will reverse.
19 January 2017 68
Fig 191 Gold , $/oz v TIPS bonds, price Fig 192 India gold imports, tonnes/month
180
160
140
120 3m moving
average
100
80
60
40
20
0
2011 2012 2013 2014 2015 2016
Source: Barclays, Macrobond, Macquarie Research, January 2017 Source: Macrobond, Macquarie Research, January 2017
But only temporarily. We cannot be bearish medium-term, and while timing is difficult we
expect strong price gains in 2H 17 and 2018 on firming investment.
In the US we see inflationary pressures accelerating, but are sceptical (pace Decembers
hawkishness) that the Fed will be able to entirely offset this thus meaning a fall in real
interest rates. If the Fed does, we see an increased potential for a US recession in 2018.
Globally we see a number of gold-friendly high impact, high risk political and economic
issues, such as rising protectionism and a string of elections in Europe.
Price gains could be limited by weak physical markets, however.
We had previously forecast a sharp recovery in Indian demand in 2017 after a poor
2016. The demonetisation of two banknotes in November has temporarily stopped that
(gold imports initially surged but fell back sharply in December (fig 186)), and while this
will soon sort itself out, it underlines the ongoing challenge gold faces in India. In
particular underlying gold demand will be hit by tighter restrictions on sales and the
possibility that more money enters the formal banking system forever.
Chinese demand also fell in 2016: GFMS, the research group, estimate jewellery and
retail investment fell 15% YoY in 2016, while PBoC purchases have tailed off of late.
Gold imports were also down about 20%. Of course the gold price was higher, and by
year-end the decline in jewellery demand was abating, though the trade is not
optimistic for next year and the government seems keen to moderate imports.
Finally, central bank demand in 2016 suffered due to Venezuelas disposals, but was
buoyed by strong Russian purchases. We expect a modest improvement this year.
Fig 193 Gold supply & demand, tonnes

Tonnes 2010 2011 2012 2013 2014 2015 2016 2017F 2018F 2019F 2020F 2021F
Jewellery 2,034 2,022 1,998 2,469 2,243 2,166 1,948 2,150 2,300 2,425 2,500 2,465
Industrial 476 468 426 418 399 347 332 326 325 324 323 322
Producer Dehedging 106 -18 40 39 -105 24 -85 -50 -15 0 0 1
Investment 1,613 1,756 1,633 907 945 990 1,547 1,140 1,220 1,155 1,105 1,013
Coins 287 326 304 379 251 263 250 240 240 230 230 220
Bar Hoarding 944 1,245 1,050 1,408 851 851 765 800 780 775 775 743
ETFs 382 185 279 -880 -157 -124 532 100 200 150 100 50
Stock moves/unknown 177 -136 -69 220 341 384 358 318 -105 -154 -58 109
Total Demand 4,406 4,093 4,018 3,955 3,823 3,863 4,100 3,834 3,825 3,850 3,900 3,910
Mine Production 2,739 2,838 2,861 3,061 3,131 3,179 3,100 2,984 2,850 2,800 2,775 2,760
Central bank sales -77 -457 -544 -409 -466 -482 -250 -325 -325 -300 -275 -250
Gold Scrap 1,744 1,705 1,701 1,303 1,158 1,166 1,250 1,225 1,300 1,350 1,400 1,400
Total Supply 4,406 4,086 4,018 3,955 3,823 3,863 4,100 3,834 3,825 3,850 3,900 3,910
Source: GFMS, Macquarie Research, January 2017
19 January 2017 69
Silver
Reflation OK, inflation good
After enjoying a re-rating against gold between April to July, taking the gold/silver ratio from
over 80 to below its 20-year average around 67, the silver price then largely tracked its bigger
cousin (but with more volatility), falling from nearly $21/oz at the start of August to just $15.80
by December. In 2017 so far it has recovered, again in line with gold, keeping the gold/silver
ratio stable around 71.
Fig 194 Silver price, $/oz and gold/silver ratio Fig 195 Solar panel production and installations, GW
120
Production Installations
100
80
60
40
20
0
2007 2009 2011 2013 2015E 2017E 2019E
Source: Bloomberg, Macquarie Research, January 2017 Source: Company data, Macquarie Research, January 2017
We expect gold to continue to be the main driver of silver prices in 2017. We are bullish on
gold medium-term on rising inflation expectations, though expect a weaker 1H, on renewed
dollar strength and short-term US economic outperformance. Given such a backdrop it is hard
not to be favourable to silver too - a global relation should not be as bad for silver as it is for
gold given its large industrial base, and any signs of a global inflation should benefit silver just
as much. As such we expect silver will be flat in 1H, outperforming gold, and rally in 2H. The
scenario in which we would be less favourable silver is if golds gains are on the back of
renewed geopolitical tensions, a clear possibility given the new US Administration and
European/Asian political issues.
On silvers fundamentals, one concern we have had is that 2017 looks set to be a much
worse year for solar panel installations, as China in particular sees an incentive-related
pullback following a bumper 2016. However, as fig 189 shows, our global renewables team
forecast that production of solar panels will remain robust, with last years surge met from
inventory. Theoretically it is production that matters for silver, though installations probably
drive manufacturers outlook and purchases of silver powder. Metals Focus, a research
group, estimate silver solar usage will decline marginally. They are, however, optimistic on
other aspects of silver industrial demand, such as automotive, and our expectations for higher
global industrial production growth also bodes well.
Metals Focus also expect silver jewellery demand to rise by 3%, compared to a fall in 2016,
foreseeing a positive performance in India. This might come as a surprise given the ongoing
demonetisation and fears about gold, but while India silver imports in 2016 were roughly half
their (gargantuan) level of 2015, December (unlike gold) saw a strong performance and it
seems plausible silvers use in India is less concentrated on the black money sectors the
government is concerned about.
19 January 2017 70
Platinum & Palladium

Slowing car market, more recycling main headwinds
An improving global economy offers upside risk to PGMs, but we think the car sales market
will slow, which coupled with more recycling would mean a weak year.
What a wretched performer platinum has been over the last ten years. In early January the
price in US dollars was 25% down from its 2007 average ($988/oz, compared to $1,317/oz)
and of course has fallen even more from highs reached in 2008 and 2011. Adjusted for
inflation it is 35% lower. It is true that the platinum price in South Africa rand has risen by 45%
over that period, but this is to ignore higher South African inflation. Take that into account and
you find it has fallen 17% (though interestingly as recently as mid-2016 it was still slightly
higher, and this real rand price has clearly been much more stable than the dollar or nominal
rand price).
Fig 196 Ten years of platinum misery, 2006 = 100 Fig 197 Diesel on the wane, % share in Europe
60
58
56
54
52
50
48
2011 2012 2013 2014 2015 2016 v
Source: Bloomberg, Macquarie Research, January 2017 Source: National data, Macquarie Research, January 2017
Why has platinum performed so badly? Over the same period gold has gained 73%, silver
26%. One clue comes from the best performing precious metal, palladium, which has risen by
114%. Platinums cheaper sister metal was the cause of the first of the three demand shocks
it faced over this period - palladium first replaced it almost entirely in gasoline car catalysts,
and then partially in diesels. The second shock was the global financial crisis, and the
double-dip recession in Europe, its heartland. The current, and potentially the most
challenging of them all, is the decline of diesel.
European diesel cars still provide 20% of total platinum demand. But the share in Europe is
falling (fig 191). We estimate it was 49.9% in 2016, down 2.1% points on 2015 and the first
time since 2009 it has not commanded a majority share. The effect of this is to render the
impressive recovery in European car sales - up 7% in 2016, and 23% since the 2011 nadir,
moot. Diesel sales rose just 1% last year, and are up only 13% since 2011. At least that is
positive - in 2017 if the European market grows at the 2-3% rate our research group expects
a similar market share decline or diesel will mean falling sales, and a hit to platinum demand
of something like 35 koz. This might not sound like much but repeated (and accumulating)
every year it soon adds up.
It is also arguable that platinum has suffered a fourth demand shock - declining Chinese
jewellery demand. 2017 is unlikely to see a big recovery there - the issues of lower luxury
consumption, declining gold jewellery demand, poor pricing and a lack of marketing remain in
place. But it does offer an upside risk. Chinas economy is enjoying a strong patch which
seems to be feeding into retail sales. Some mainland jewellers reported sales growth in 4Q
2016 for the first time that year and SGE turnover in platinum did accelerate in December.
19 January 2017 71
Fig 198 Chinas platinum imports remain low, though Fig 199 Palladium volatile, but heading upwards in
December likely to be higher (000 oz) 2016, $/oz - mostly tracking equities (MSCI emerging)
400 180
350 160
300 140
120
250
100
200
80
150
60
100 40
50 20
China imports SGE turnover
- -
2013 2014 2015 2016
Source: Bloomberg, Macquarie Research, January 2017 Source: Bloomberg, Macquarie Research, January 2017
Things have been much better for palladium, with the price beginning 2017 at a high. After
plunging lower at the start of 2016, it trended higher throughout the year, though with much
volatility. These swings seemed to be driven by global economic confidence, as shown by
the long-running tight correlation with global emerging markets equities (fig 193).
From 3Q the price began to clearly outperform equities, however, something we attribute to
the surge higher in global car sales (fig 194), led by China. We had expected Chinese car
sales to look good on a YoY basis mid-year because they were poor in 2015 - what we hadnt
foreseen was they would still look good in 4Q (when 2015 sales were already growing
rapidly). Furthermore for palladium in the past we have noted a disconnect between
estimated Chinese palladium demand and imports - now (if we include Hong Kong, where two
large shiploads came from Russia in September/October) they have caught up and even
exceeded it (fig 195).
Whats not to like? The 2017 outlook. Some of that Chinese demand was surely brought
forward, and US and European sales look toppy. We expect global car sales, which rose at a
three-year high of 4.8% in 2016, to grow by only 2%. Palladium on balance will continue to
benefit from shifts away from diesel and tighter emissions standards, and sales do have some
upside risk - in China if the economy remains strong, and in the US if a Trump Administration
does boost growth through a fiscal stimulus. But the threat of trade wars and political risks
cannot be entirely discounted, and EVs continue to nibble away at demand.
Fig 200 Global car sales - whoosh - but how long can Fig 201 Chinas palladium imports finally match up to
it last? Monthly sales, annual rate, and SA, vehicles its requirements, 000 oz (rolling 12m)
2,500 HK pd imports
China pd imports
2,000 Pd required for car production
Plus jewellery
Plus others
1,500
1,000
500
0
2001 2003 2005 2007 2009 2011 2013 2015
19 January 2017 72
If car demand in both markets is going to be weaker, recycling is set to be higher - and it
could be a lot higher. In the last few years we have been surprised by the subdued volumes
in this area, explained by refiners as a function of low prices both for PGMs (encouraging
hoarding by collectors) and steel (so reducing the incentive to scrap the car or the catalyst
casing). The price of both has since risen substantially, and volumes will follow: a simple
assessment of how much platinum and palladium went into cars in the late 1990s/early 2000s
that are now being scrapped suggests if the market does turn it could do so quite
substantially.
Finally we need to discuss investment flows in palladium. Palladium suffered another year of
huge outflows in 2016, over 600k, and yet the price gained. Obviously stock liquidation is a
fact in a deficit market, but normally this has been seen in undocumented above-ground
stocks, not ETFs. If this implies those above-ground stocks are drying up that would be very
bullish indeed, and suggests when car sales pick up again, palladium could rally hard.
To sum up, in platinum fundamentally the situation looks poor due to the ongoing drag from
diesel. A lot of this must, it is true, be in the price, and platinum is likely to be dragged higher
by gold, so we dont expect much downside. But substantial price gains seem some years off
- when supply is being actively cut and emissions legislation in heavy-duty helps demand.
In palladium, we are bearish near term. The price is high and global car sales the dominant
demand source are clearly set for a weaker period. Palladium can be fickle, and while it has
coped with hefty investor liquidation so far, we dont think it will if demand fades.
Fig 202 Platinum supply-demand balance, 000 ozs
Supply 2012 2013 2014 2015 2016E 2017F 2018F 2019F 2020F 2021F
South Africa & Zimbabw e 4,441 4,692 3,986 5,058 4,888 4,832 4,666 4,659 4,691 4,660
( of which Mine stocks) 112 (200) 556 182 150 0 0 0 0 0
Russia (inc mining stocks) 800 736 700 680 735 719 722 733 730 775
North America 310 300 303 304 342 328 315 318 325 340
Others 125 163 156 149 150 150 150 140 140 140
Disruption allowance 0 0 0 0 (49) (192) (193) (193) (194) (194)
Total Mine Supply 5,676 5,891 5,145 6,191 6,067 5,837 5,661 5,657 5,692 5,721
Autocat Recycling 1,120 1,205 1,282 1,127 1,184 1,307 1,431 1,546 1,600 1,633
Dem and
Net autocatalyst 1,963 1,787 1,838 2,140 2,148 2,080 1,937 1,886 1,822 1,856
Net Jew ellery 1,888 2,238 2,135 2,255 1,901 2,051 2,102 2,154 2,205 2,208
Other industrial 1,489 1,583 1,728 1,760 1,901 1,830 1,856 1,860 1,908 1,971
Total Dem and 5,340 5,608 5,701 6,155 5,949 5,961 5,895 5,900 5,935 6,034
Market balance 336 283 (556) 36 117 (125) (234) (242) (243) (313)
Investment 450 871 277 360 452 0 (200) (150) (100) (50)
Surplus/Deficit (114) (588) (833) (324) (335) (125) (34) (92) (143) (263)
Source: Johnson Matthey, SFA Oxford, Thomson Reuters GFMS, Macquarie Research, January 2017
Fig 203 Palladium supply-demand balance, 000 ozs

Supply 2012 2013 2014 2015 2016E 2017F 2018F 2019F 2020F 2021F
South Africa & Zimbabw e 2,585 2,760 2,443 3,028 2,895 2,963 2,959 2,939 2,934 2,890
( of which Mine stocks) (80) (6) 238 50 40 0 0 0 0 0
Russia 2,627 2,581 2,660 2,464 2,491 2,520 2,520 2,500 2,500 2,550
North America 836 882 902 945 958 938 953 964 963 1,075
Others 165 150 150 140 126 130 130 140 150 150
Disruption allowance 0 0 0 0 (34) (136) (138) (140) (140) (136)
Total Mine Supply 6,213 6,373 6,155 6,576 6,436 6,414 6,424 6,403 6,407 6,529
Russian Stockpile sales 260 100 100 0 0 0 0 0 0 0
Total Supply 6,473 6,473 6,255 6,576 6,436 6,414 6,424 6,403 6,407 6,529
Autocat Recycling 1,660 1,860 2,189 1,939 2,065 2,276 2,424 2,559 2,690 2,835
Dem and
Net autocatalyst 5,013 5,191 5,311 5,716 5,681 5,333 5,301 5,217 5,011 4,929
Net Jew ellery 150 130 140 152 170 188 191 194 198 201
Other industrial 1,881 1,654 1,527 1,564 1,564 1,523 1,486 1,457 1,410 1,391
Total Dem and 7,044 6,975 6,978 7,432 7,415 7,044 6,978 6,868 6,619 6,520
Market balance (571) (502) (723) (856) (979) (630) (555) (465) (211) 9
Investment 470 24 915 (700) (615) (350) (250) (150) (100) (50)
Surplus/Deficit (1041) (526) (1638) (156) (364) (280) (305) (315) (111) 59
Source: Johnson Matthey, SFA Oxford, Thomson Reuters GFMS, Macquarie Research, January 2017
19 January 2017 73
Diamonds
After the rally, theres the after-rally
Downstream View Improving, mainly in the US: 2017 is gearing up to be an interesting
year in the diamond market. Expectations for global growth in jewellery demand have finally
started to improve. The US is solidifying its global leadership position, now having delivered
12 straight months of above GDP growth, including a strong 12% YoY increase in November
2016. Even China, which has been a weak spot since 2014, is finally starting to rally albeit off
a low base. Retailers such as Richemont, Chow Tai Fook, and Luk Fook have all reported
surprisingly strong results over the past 2 quarters from their mainland Chinese operations,
adding much needed optimism in the sector which has stagnated since 2014. A return to
growth in China could be a major turbo boost for diamond demand as shown by the relative
size of China vs the US jewellery sectors in the exhibit below.
The rise of synthetic diamonds continues to be a key concern for the diamond market, at least
theoretically. While the market has been focused on this issue for much of the last 20 years,
we believe the tone of concern is noticeably changing. To date, the size of the disclosed
synthetic share of the diamond market is at most 1% or 2%, in our view. But several new
synthetic diamond producers are gaining publicity using the marketing platform that synthetics
are more environmentally sustainable, more ethical, and cost effective. Further, the increase
in undisclosed synthetic diamonds being marketed as natural diamonds is becoming a key
concern for diamond traders and retailers.
Fig 204 US jewellery & watch sales YoY% (seasonally adjusted)
14.0% 12.4%
12.0%
10.0%
8.0% 6.7%
5.2% 5.5% 5.7%
6.0% 5.6%
4.4% 4.4%
3.9%
3.5% 3.2% 3.3%
3.1% 3.4%
4.0% 2.5% 2.7% 3.0%
1.8% 1.6% 1.5%
1.2%
2.0% 1.2% 0.9%
0.9% 0.9%
0.7% 0.3%
0.0%
-0.1%
-0.5% -0.2%
-0.5%
-2.0% -1.2% -0.8%-0.8%
-1.4%
-4.0%
NOV '14
SEP '15
NOV '15
NOV '16
SEP '14
SEP '16
MAY '14
JUL '14
MAY '15
JUL '15
MAY '16
JUL '16
JAN '16
JAN '14
MAR '14
JAN '15
MAR '15
MAR '16
Source: Customs Statistics, Macquarie Research, January 2017
Midstream View Back in business, but with Headwinds: In the midstream of the
diamond pipeline, diamond cutters and polishers returned to market in 2016 with a
vengeance, following the inventory destock of 2015 that lead to near 20% decline in rough
diamond prices. Sales volumes at DeBeers were up a whopping 76% YoY to $5.6B, back to
near bull market levels and we estimate total sales by DeBeers and Alrosa exceeded
production in 2016 by 15M carats. Nonetheless we still believe inventories held by the major
producers remain above average (with 10-15M carats of excess inventories remaining).
In the short term, we are increasingly concerned about the effects of the demonetization
policy in India having a material impact on the diamond cutting/polishing industry there. As
much as 80% of the worlds rough diamonds by volume are cut and polished in India, focused
mainly on the lower end of the quality spectrum. We are hearing increasingly by producers
that Indian buyers, who often conduct business in cash, are stepping away from the market.
DeBeers first sight of 2017, conducted during January 17-20th, will be an important
indicator of the extent of reticence by Indian buyers.
19 January 2017 74
Fig 205 China plateaus, but is now a major jewellery player
Source: Macquarie Research, US BEA, China NBS January 2017
Upstream View Wall of supply upon us: On the supply side, there are increasing
concerns that 2017 will be a year of oversupply in the rough diamond market. We estimate a
total of 18M carats of annualized production from new mines and expansions are coming
online during the year, equivalent to nearly 15% production growth. In most commodities this
would signal a weak outlook for pricing. In the case of the diamond sector where the
dominant producers have a history of managing the supply-demand balance, we are less
concerned. We also note that nearly 10M carats of the planned 18M carats of new supply is
slated to come from Rio Tintos Argyle mine, where average per carat values are only about
$30-40/ct, significantly less than the ~$120/carat global average for gem quality stones, and
therefore not in direct competition with many of the mid and large cap producers.
The overall theme of global scarcity remains a key long-term positive for the diamond industry
which we believe has been overlooked of late. In Alrosas recent reserves and resource
update, the company reported a depletion of 77M carats compared with the 2016 update, for
instance.
As a result, we continue to forecast muted pricing (0% to -5%) for rough and polished
diamonds in the short term (H1/17). By mid-year, when we hope to have clarity on producer
discipline, and the ability of the midstream to absorb new supply, we think the stage will be
set for meaningful price upside (5-10% per annum) through the end of the decade.
Fig 206 New projects add max 39M carats, but Fig 207 Cautious Price View in Short-term, But Long-
production peaks 2018E-21E (K cts) term Upside Is Evident
300
45,000
Argyle
250
40,000 Ekati Forecast
35,000 Bunder 200
30,000 Severalmaz
150
Udachny
25,000
Finsch
20,000 100
Cullinan
15,000
Liqhobong 50
10,000 Renard
0
5,000 Gahcho Kue
Feb-14
Feb-06
Feb-08
Feb-10
Feb-12
Feb-16
Feb-18
Feb-20
Oct-06
Jun-07
Oct-08
Jun-09
Oct-10
Jun-11
Oct-12
Jun-13
Oct-14
Jun-15
Oct-16
Jun-17
Oct-18
Jun-19
Oct-20
Jun-21
Botuobinskaya
-
Grib Rough Diamonds Polished Diamonds Rough Forecast Polished Forecast
Source: Company Data, Macquarie Research, January 2017 Source: Bloomberg, WWW Int, Macquarie Research January 2017
19 January 2017 75
Global Crude Oil

From Binaries to Balancing in 2017 and 2018
2016 was a year that saw a plethora of improbable binary events and the oil market was no
exception. OPEC defied expectations and not only managed to coordinate the first
production cut of the decade, but also corralled a significant group of Non-OPEC producers to
agree to reduce supply by a combined ~1.8 MBD. If the OPEC/Non-OPEC group can
execute on the deal, the global supply-demand balance will move into deficit for the first time
since early 2014 (Fig. 202 below). We believe this will support WTI prices in the mid-$50/bbl.
range in 2017, with upside towards $60/bbl. With a robust shale response mounting and an
eventual reversal of cuts expected, we believe the bill for 2017 relief will come due in 2018,
where we see inventories again building.
Fig 208 Deficits in 17 yield to renewed surplus in 18 Fig 209 Grades indicate early signs of compliance
Global S/D Balances [K BPD] Brent - Dubai Spread Brent - Urals Spread
$4.5 $2.0
2,002
1,737
$4.0 $1.7
Brent Urals $/bbl

Brent Dubai $/bbl
922 791 $3.5
322 $1.4
115
$3.0
$1.1
$2.5
-595
$2.0 $0.8
1Q12
3Q12
1Q13
3Q13
1Q14
3Q14
1Q15
3Q15
1Q16
3Q16
1Q17
3Q17
1Q18
3Q18
$1.5 $0.5
Oct-16 Nov-16 Dec-16 Jan-17
Source: IEA, Macquarie Capital (USA), January 2017 Source: Bloomberg, Macquarie Capital (USA), January 2017
2017 Focus on deal compliance and extension, this time its different
While anticipated production cuts have supported price, OPEC quota compliance has been
historically poor. In that context, we expect that the market views a 70-80% compliance rate
as reasonably successful. Recognizing this concern, OPEC has agreed to a newly
established third party monitoring system, although details of this process and how it would
potentially incorporate Non-OPEC contributors are less clear. While we view effective
monitoring as a step in the right direction for both achieving the cuts and maintaining
cooperation, we remain sceptical of a hometown referees calls.
We expect the incentive for cooperation among participants will be at its peak at the onset of
the deal and market indicators are currently corroborating reports of US refiners receiving
lower oil allocations. In measuring compliance through market indicators, we focus on two of
the most important contributing party grades or the Brent Dubai and Brent Ural spreads in
reference to Saudi Arabia and Russia. Both spreads are now indicating a reduction in supply
through tightening differentials, yet we will expect a stronger confirmation to be demonstrated
in crude loadings late January to early February.
Watch the hands - potential for deal slippage persists
Despite our early optimism, succuss in a cartel can often sow the seeds of cheating. With
most OPEC producers estimated to maintain underwater budgets even at $60/bbl, we feel its
important to keep an eye on possible means to remain compliant yet potentially push more
volumes out the door. OPEC quotas typically have excluded condensate as theyre
considered part of NGL volumes, and thus could be a means of revenue expansion under the
current regime. Additionally, OPEC producers could potentially draw down inventories without
violating the production agreement. Finally, in evaluating other risks to the deal, we feel a
constant friction in the growth allowance to Iraq and Iran could move out of balance and
second, what has become normal variations in production from Nigeria and Libya could be
enough to swing balances for better or worse.
19 January 2017 76
While the OPEC deal is set for six months, we feel the market is already imbedding a high
probability of deal extension. The corollary to the dont fight the Fed today is dont fight the
Saudis, premised on the potential for a 18 Saudi Aramco IPO, which would allow KSA to
better tap its balance sheet/improve its liquidity position and should benefit from stable oil
price expectations. That said, deal extension is not a forgone conclusion, in our view, and
believe another round of gamesmanship will occur in June 2017.
The US Shale Response: Riyadh, we have a problem
We estimate the 2014-2015 KSA policy of letting crude prices float may have removed $1trn
in energy capital spending and delayed or removed significant production volumes. Yet,
OPECs recent success in coordinating production cuts and supporting global oil prices will
not occur in a vacuum and could reverse the prior benefits in a rather short period. Add
higher liquidity from stronger prices and an infusion of capital from more globally oriented
producers, and we anticipate a sharp response from tight oil in. At our current WTI price
forecast of $56 in 2017 and $62 in 2018 we expect 1.7 MBD of U.S. onshore oil growth by YE
18 relative to YE 16 levels.
Fig 210 Shale response driven by the Permian Fig 211 Per rig productivity remains in upward trend
Source: EIA, IHS, Macquarie Capital (USA), Jan-17 Note: Incl. inland waters Source: IHS, Baker Hughes, Macquarie Capital (USA), January 2017
In the potential for a large production response from the US, we see significant risk of OPEC
reverting back towards dysfunction from the current spirit of coordination, leading to an
eventual reversal from 2017 production cuts. Outside of shale, we are seeing early signs of a
response to improved prices from some of the oil markets most vulnerable barrels, as
conventional onshore declines moderate. All told, these OPEC reductions may ultimately
prove bearish longer-term.
Fig 212 Early signs of stabilization emerging in high cost, conventional non-OPEC oil
Source: IHS, Bloomberg, Colombia Ministry of Mines and Energy, Macquarie Capital (USA), January 2017
19 January 2017 77
US Natural Gas
Tightness likely to persist across summer 2017
After moving to an uncomfortably bullish posture on natural gas ahead of winter, given stalling
supply and rising real demand, we continue to see support for natural gas prices through
summer 2017. With strong underlying gas demand growth likely offsetting price-related
impacts to power demand (reverse coal to gas switching) healthy production growth will be
required to achieve an acceptable level of storage at the end of the 2017 injection season.
A harder path to supply growth improves skew for bulls
While we are only slightly trimming our gas production growth forecast for summer 2017, this
comes with significantly higher uncertainty. Given announced and further potential delays to
large Marcellus/Utica pipeline projects, the gas market must look outside of the Northeast for
supply growth in 2017. Associated gas growth will help; with improved crude oil prices, we
expect a significant tight oil response, and with that, up to ~2+ BCFD of exit/exit associated
gas growth in 2017, similar to the annualized rate experienced from 2011-2015. However,
this impact will not be fully felt across summer 2017 and the uncertain timing and limited
visibility of this production are likely to blunt its market impact in forward looking balances.
Fig 213 Associated gas only down marginally from peak and headed for sharp re-growth
Source: IHS, Macquarie Capital (USA), January 2017
Outside of the Marcellus/Utica and tight oil plays, we see a mounting response to higher
prices in the Haynesville/Cotton Valley, where producers are revisiting the once-hyped play
with aggressive fracs that push the bounds of stimulation intensity. We think these new
completions, coupled with increased activity from a relatively stable production base can drive
meaningful growth by YE 17. Unfortunately for supply-oriented bears, we expect this growth
to proceed at a far more moderate pace than the market has come to expect from the
Marcellus/Utica, where per rig productivity in core areas can greatly exceed these plays given
faster drilling times.
Rounding out the shale gas landscape, we see potential for still significant y/y growth from the
Marcellus/Utica in summer 2017 (on the order of ~2 BCFD), but this growth will need to be
driven by a variety of sources and improved prices. Further, as the Northeast appears to be
producing somewhat below capacity today despite firm cash prices across the region, risk
surrounds the historically reliable If you build it, they will come heuristic for Marcellus/Utica
supply.
19 January 2017 78
Fig 214 Shale gas supply flat in Appalachia, while Haynesville poised for re-growth after recent stabilization
Source: IHS, Bentek, Macquarie Capital (USA), January 2017
Demand growth ahead of schedule, tightening summer balances

Likewise, weather excluded, the demand side continues to roll in quite favorably for bulls,
given potentially faster than anticipated LNG start-ups and throughput pushing at capacity,
higher coal prices that will mitigate reversal of coal to gas switching at higher y/y prices, and
steady industrial demand growth. All told, we expect end of October storage levels of 3.6
TCF in 2017, indicating a tight gas market and supporting prices across summer. The rate of
prodn growth heading into Winter 17/18 and significantly expanded NE takeaway could
weigh on 18 balances, but first the market must confront concerns around sourcing adequate
supply to meet rising demand in 17.
Fig 215 Summer 2017 balances appear tight even assuming significant y/y supply growth
SUPPLY / DEMAND and END OF SEASON STORAGE

Daily [BCFD] Win 14/15 Sum 15 Win 15/16 Sum 16 Win 16/17 Sum 17
YOY Supply Growth(Decline)
Production 5.58 2.70 1.00 (0.90) (1.30) 1.90
Canadian net imports 0.10 0.36 (0.30) 0.53 0.20 0.20
LNG imports 0.14 0.09 0.00 0.08 0.05 0.00
Total Supply incr/(decr) 5.82 3.15 0.70 (0.29) (1.05) 2.10
YOY Demand Growth/(Decline)

Pow er - Coal Plant Retirements 1.40 0.58 0.25 0.50 0.60 0.50
Core Industrial 0.02 (0.40) 0.10 0.42 0.50 0.40
Res/Com (2.20) (0.58) (4.50) (0.22) 4.00 0.00
Mexican exports 0.36 0.92 1.60 0.69 0.75 0.75
LNG exports 0.00 0.00 0.15 0.57 1.40 1.30
Coal to gas sw itching 0.00 4.15 1.40 1.26 (2.70) (1.50)
Total demand incr/(decr) (0.42) 4.67 (1.00) 3.22 4.55 1.45
YOY Indirect & special demand

Other / w eather / balancing 0.83 (2.03) (1.34) 0.70 1.00
Nuclear 0.00 0.00 0.00 0.00 0.00 0.00
Hydro plant runs 0.00 0.00 0.00 0.00 0.00 0.40
Total Indirect & special incr/(decr) 0.83 0.00 (2.03) (1.34) 0.70 1.40
Total YOY Storage Incr/(Decr) 5.42 (1.52) 3.73 (4.85) (4.90) 2.05
Storage incr/(decr) remaining (2,109) 2,548 (1,542) 1,495 (2,297) 1,934

End of season storage 1,462 4,010 2,468 3,963 1,666 3,600
Source: EIA, Bentek, Macquarie Capital (USA), January 2017
19 January 2017 79
Thermal Coal
Will 276 days return or will prices settle at an acceptable level?
Thermal coal prices have finally eased since November on the back of the numerous
changes to the 276 days policy seen since September. Spot thermal coal prices in
Qinhuangdao have fallen from highs in November of over 730RMB/t, to just under 600RMB/t
in mid-January, as domestic mine supply has clearly picked up and inventory is rising.
In terms of the outlook for prices in 2017, all eyes remain focused on what happens next
regarding Chinese policy. If prices continue to slide, we expect that some form of output
restrictions will come back with the 276 days policy, which in November was suspended
until the end of 1Q, being reapplied to some mines. Beijing policymakers now essentially
provide a floor to global coal prices, with a Beijing put underwriting the market.
Indeed the government recently outlined in a policy memorandum their targeted thermal coal
price range based around the annual contract price of RMB535/t. Supply interventions now
only appear likely if prices are above RMB600/t or below RMB470/t. With the clear threat to
supply either side of this range, it is possible that just the governments pronouncements on
the market may be enough to prevent prices trading outside this range. RMB535/t equates to
around $70/t FOB Newcastle, and our confidence in this Beijing put supporting global coal
markets is one of the key reasons for our recent coal price upgrades.
Fig 216 Thermal coal prices have had an extremely Fig 217 Supply has clearly responded after policy
volatile last twelve months loosening
China thermal coal production run rate
Rmb/t USD
1000 China QHD 5500kcal coal price-LHS 160 Btpa
2.6
900 Newcastle FOB-RHS
140
800 2.4
120
700
100 2.2
600
500 80 2.0
400 60
1.8
300
40
200 1.6
100 20
1.4
0 0
Sep 11
Sep 12
Sep 13
Sep 14
Sep 15
Sep 16
May 11
May 12
Jan 11
Jan 12
Jan 13
May 13
Jan 14
May 14
Jan 15
May 15
Jan 16
May 16
Jan 17
1.2
2012 2013 2014 2015 2016
Source: Platts, Sxcoal, Macquarie Research, January 2017 Source: Sxcoal, Bloomberg, Macquarie Research, January 2017
Fig 218 Domestic coal shortages led to imports Fig 219 Inventory has been rising as a result
Mt China monthly coal (inc. lignite) imports

Plant plus port thermal coal stocks - China
40 Days
45 2013 2014
35
2015 2016
40
30
25 35
20 30
15
25
10 2015 2014 2013 2016
20
5
0 15
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Source: China Customs, Macquarie Research, January 2017 Source: Sxcoal, Macquarie Research, January 2017
19 January 2017 80
Outside of the near-term production controls, China recently issued its 13th five year plan
(2016-2020) for the coal industry. The plan outlines a capacity closure target of 800mtpa over
the five year period, while at the same time it approves the addition of 500mtpa advanced
new capacity. The number of coal mines operating in the country is targeted to be reduced
from 9700 at the end of 2015 to 6000 by the end of 2020.
Interpreting the policy led us to calculate a possible net increase of 268mtpa in operating coal
mine capacity through 2020, bringing total operating mine capacity to 4.16Bn tonnes by then.
This means no reduction to Chinas real effective supply capability for coal is contrary to the
headline numbers. The five year plan gives a coal production target of 3.9Bn tonnes for 2020,
versus its forecast of 4.1bn tonnes coal consumption.
With production specifically being outlined to be below demand, this leaves a net requirement
for imports of 200mtpa by 2020. An import change in the policy wording compared with the
previous five year plan shows more conservatism towards coal imports, as the new policy is
said to encourage high quality coal imports, reinforce coking coal imports and strictly control
low quality coal imports, while in the previous five year plan the wording was more general to
encourage imports. China imported 229mt of coal in the first eleven months of 2016, so if
the government wants to control imports to no more than 200mt it would mean a reduced
demand for seaborne coal from China compared to recent years, and indeed it would be the
lowest import number since 2010.
The seaborne market meanwhile has reacted quickly to the changes in prices and Chinas
import coal demand last year. Chinas trade position was that of being a net exporter of coal
in 1Q16, with 5.5mt combined exports of coal, lignite & coke, the highest volume since 1Q11.
Exports however fell off from 2Q on supply cuts, but still 2016 exports were the highest since
2010. Coal imports meanwhile surged, reaching 27mt in November 16, their highest since
Dec 14. Total 2016 imports were 256mt versus 204mt in 2015, though this is still below the
291mt seen in 2014.
Seaborne thermal coal supply saw little change last year from the major supply countries,
though Indonesian exports did pick up in 2H16, offsetting some of the decline seen in 1H16.
Outside of Indonesia, Russia and Colombia saw higher export volumes in response to the
higher margin incentives, but elsewhere seaborne coal supply has been slow to respond.
While the concept of a Beijing put leads us to be less pessimistic on the potential floor price
for thermal coal in the medium term, the structural headwinds remain in terms of declining
western world demand, and ongoing disappointment in the much hoped for import demand
growth in India, where Coal India continues to do a decent job of lifting domestic coal supply
to meet demand.
Fig 220 Thermal coal market supply-demand

Dem and (m t) 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F
EU-28 126 116 90 85 81 77 73 70
Other Atlantic 66 74 73 76 79 82 85 88
Korea 100 101 103 109 113 114 114 114
Japan 140 144 142 142 142 142 142 142
India 151 164 146 146 141 136 131 131
Other Pacif ic 146 149 156 164 173 181 185 187
Total Ex-China 729 749 710 722 729 732 730 732
China 201 130 146 146 134 126 122 115
Total 930 879 856 868 863 858 853 847
YoY 1% -5% -3% 1% -1% -1% -1% -1%
Supply (m t) 2014 2015 2016F 2017F 2018F 2019F 2020F 2021F

Australia 201 202 200 202 204 204 204 204
Indonesia 407 363 355 355 345 335 325 315
South Af rica 75 76 74 76 77 78 79 80
Colombia 75 80 86 89 92 95 98 101
Russia 106 106 112 114 114 114 114 114
USA 31 24 14 18 16 16 16 16
Other 25 14 15 14 15 16 17 17
Total 919 866 856 868 863 858 853 847
Source: Customs Statistics, Macquarie Research, January 2017
19 January 2017 81
Uranium
Theres low, and theres too low
2017 has gotten off to a decent start for 2016's worst performing commodity, with
expectations of supply adjustments seeing prices up >10% from the start of the year.
However, when the initial base was so low, that only equates to a $2/lb move in the spot
price. To put things in context, uranium pricing is currently trading 50% of where it was 40
years ago in nominal terms never mind adjusting for inflation. There is no other commodity
for which this is true. Essentially, this has put uranium in the situation where many peer
commodities were at this time last year, trading too far into the cost curve for pricing to be
sustainable.
The uranium price was hammered in 2016 by weak spot demand from China and the US, and
a distinct lack of mine supply cuts. This leaves the market as one of the few trading deep into
the cost curve, with ~25% cash negative at the present time. Even with weak fundamentals,
we think that from current levels there is more upside risk than downside, from mine
shutdowns certainly, but perhaps more importantly given confidence has been restored in the
US, the worlds largest uranium consumer. With the closure of a large number of nuclear
power plants announced earlier in 2016 on economic grounds, legislative actions in New York
and Illinois keeping some of these open will provide both more optimism and spot market
demand into 2017.
Fig 221 The spot price of uranium has started to turn Fig 222 Chinese import prices have also fallen, but
after a disastrous 2016 they continue to pay above the spot price
$/lb U3O8 spot price China natural uranium imports and unit price
tU $/lb
70 6,000 160
Imports
Realised Price (rhs) 140
60 5,000
Spot Price (rhs)
120
50 4,000
100
40 3,000 80
30 60
2,000
40
20
1,000
20
10
0 0
2009 2010 2011 2012 2013 2014 2015 2016 2017 Jan 08 Jan 10 Jan 12 Jan 14 Jan 16
Source: UxC, Bloomberg, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
Fig 223 We are past the period of peak Chinese Fig 224 while inventories at major consumers still
stocking, but the market still needs some build sit at four years of global mine output
China uranium balance
tU
Estimated inventories at four major consumers
25,000 Domestic output ,000 tU Years
Net imports enriched
China
Net imports natural 300 6.0
20,000
Actual demand Japan
250 EU 5.0
15,000
US
200 Years global mine output 4.0
10,000
150 3.0
5,000
100 2.0
0
50 1.0
-5,000 - 0.0
2017F
2018F
2019F
2020F
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2008 2009 2010 2011 2012 2013 2014 2015 2016
Source: China Customs, NBS, Macquarie Research, January 2017 Source: EIA, WNA, Macquarie Research, January 2017
19 January 2017 82
Just as with other commodities last year, supply cuts are to be expected at current price
levels. Last year saw Cameco idle its Rabbit Lake operations. More recently,
Kazatomproms announcement of a 10% cut to Kazakhstans uranium output this year has
piqued some interest. If this 2,000tU cut is invoked, this will be the first YoY fall in output from
the worlds largest uranium supply country in recent history. While production growth has
been tailing off in recent years following a surge at the end of the last decade, trend has still
been an addition of 500tUpa. Interestingly, this cut also comes when Kazakh production is
more competitive on a global basis following the devaluation of the Tenge, suggesting
Kazatomprom are willing to show industry leadership in supply adjustment.
However, there is still some supply growth coming through as a partial offset, with the tail end
of Camecos Cigar Lake ramp-up now just about complete, the ongoing ramp-up at the
2,000tUpa Husab mine in Namibia and an expected production recovery at Rios Rossing
operation. Overall however, we still expect 2017 to mark the lowest level for uranium supply
since 2010, helped by the steady decline in secondary material. Even with reactor
requirement growth of 2%, this still leaves us with a current year surplus of ~3,800tU in the
current year. Admittedly, this is the smallest surplus since 2008, and we expect the market to
move back towards a balanced situation by 2020. However, this is still a net inventory build.
As such, Chinese stocking continues to be very important for the uranium market. As figure
217 shows, we are now past peak Chinese rate of inventory build, however the market
remains dependent on this to absorb the surplus. And unusually, uranium is a market where
the Chinese buyers on average pay more than the prevailing spot price, though the average
customs value has been falling in recent times. On top of the Chinese stocks of ~100ktU,
there is ~50ktU in each of Europe, Japan and the US. Given this level of global inventory, the
need for incentive pricing for new uranium mine projects could be decades away.
Fig 225 The low price is now getting a supply Fig 226 despite the fact that producers there are still
reaction, with Kazakh supply set to drop relatively competitive on a global basis
Kazakhstan's natural uranium output Uranium cash cost curve

tU 45
25,000
40
Australia
20,000 35 Canada
30 China
15,000
$/lb U3O8
25 Kazakhstan
20 Spot price Namibia
10,000
15 Niger
Other
5,000 10
Russia
5
US
0 0 Uzbekistan
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017f
0 5 10 15 20 25 30 35 40 45 50 55
Cumulative Volume (kt U)
Source: Kazatomprom, Macquarie Research, January 2017 Source: Industry Data, Macquarie Research, January 2017
Fig 227 Global uranium market balance (tonnes U)

Tonnes U 2013 2014 2015 2016 2017F 2018F 2019F 2020F 2021F
Supply
Total Primary Supply 59,534 55,967 60,604 61,437 58,817 59,298 59,686 60,076 63,000
Total Secondary Supply 20,975 17,650 16,450 15,550 15,200 14,650 14,400 14,150 13,900
Total 80,509 73,617 77,054 76,987 74,017 73,948 74,086 74,226 76,900
% change YoY 2% -9% 5% 0% -4% 0% 0% 0% 4%
Reactor Requirements
China 5,678 5,435 5,589 6,488 7,809 8,965 9,956 10,782 11,608
Ex-China 56,075 58,319 60,695 62,643 62,410 61,674 62,277 63,458 63,922
Total 61,754 63,754 66,283 69,131 70,220 70,639 72,234 74,240 75,530
% Change YoY -2% 3% 4% 4% 2% 1% 2% 3% 2%
Balance 18,756 9,863 10,771 7,856 3,797 3,308 1,852 -14 1,370
Source: Macquarie Research, WNA, Industry data, Trade Statistics, January 2017
19 January 2017 83
Crude Palm Oil

Tight CPO inventory, vegetable oil complex should support prices
CPO prices strengthened in 4Q16 to end the year strongly at US$713/T (+36% YoY), driven
by: (i) lower palm oil inventory. Malaysias Dec16 palm oil inventory fell by 37% YoY to 1.66m
T, the lowest for December since 2010, due to weak production and robust export demand;
(ii) soybean complex firmed up in 4Q16 as uncertainties in South American crop prospects
offset the impact of record US crop; (iii) soybean oils value share remained stable at 40%;
and (iv) Indonesia remains committed to its biodiesel plan (solid renewal of biodiesel tender).
Overall, these events were broadly in line with our expectations, as highlighted here.
Fig 228 USDA expects further tightening of vegetable Fig 229 which should, in our view, support the
oil complex in 2017E recovery in vegetable oil prices
(US$/T)
16.0% 1,500
14.0% 1,300
12.0%
1,100
10.0%
900
8.0%
6.0% 700
4.0% 500
2.0%
300
Jan-11
Jul-11
Jan-12
Jul-12
Jan-13
Jul-13
Jan-14
Jul-14
Jan-15
Jul-15
Jan-16
Jul-16
0.0%
2010 2011 2012 2013 2014 2015 2016E 2017E
Veg. Oil Stocks-usage 5 yr average
CPO (US$/T) SBO (US$/T)
Source: USDA, Macquarie Research, January 2017 Source: Bloomberg, Macquarie Research, January 2017
Moving into 2017, we forecast the CPO stock-to-usage ratio to remain tight at 12.5% (versus
5-year historical average of 19%), as lower starting inventory and robust demand growth
should more than offset recovery in production.
Overall, we forecast the CPO prices to average US$675/T in 2017E (+7% YoY). Our CPO
price forecasting starts with the soybean complex. The tightening stock-usage ratios expected
by USDA for the broader vegetable oil market and soybean oil itself should help sustain the
soybean oils value share at 40%, eliciting soybean oil price forecasts of US$755/T for 2017E.
We then pencil in a discount of US$80/T for CPO prices (to soybean oil), a narrower discount
to historical average of US$120/T, taking into consideration a tighter palm oil inventory.
Elsewhere, we continue to omit the energy complex from our price setting framework. The
increase in CPO prices have outpaced recovery in the crude oil price. At over US$200/T
premium to gas oil, the CPO prices are too high to drive discretionary blending, we think.
Fig 230 We expect 2017E CPO stock-usage ratio to stay low on reduced starting inventory and robust demand
growth, which more than offsets a recovery in production
CPO S&D fundamentals (mT) 2011 2012 2013 2014 2015 2016f 2017f
Supply 50.5 53.8 56.6 59.9 62.6 60.3 65.7
YoY% Change 10.2% 6.5% 5.1% 5.9% 4.4% -3.7% 9.0%
Demand 48.7 52.5 57.7 59.4 60.9 63.6 66.9
YoY% Change 4.7% 7.7% 10.0% 2.9% 2.6% 4.4% 5.2%
Balance 1.8 1.4 -1.1 0.5 1.7 -3.3 -1.2
Stock to use, % 20.2% 22.4% 18.2% 19.6% 21.1% 15.0% 12.5%
Source: USDA, Oil World, MPOB, GAPKI, Macquarie Research, January 2017
19 January 2017 84
Fig 231 Weaker 2016 CPO production and healthy Fig 232 has led to large inventory drawdown;
exports demand Malaysias Dec16 inventory was 37% below average
('000 T) ('000 T)
2,500 2,500
2,000
2,000
1,500
1,500
1,000
1,000
500
500
0
-
Jan-14
Apr-14
Jul-14
Oct-14
Jan-15
Apr-15
Jul-15
Oct-15
Jan-16
Apr-16
Jul-16
Oct-16
Malaysian CPO production Malaysian CPO exports Month's avg. inventory (2011-15) Most recent month's inventory
Source: MPOB, Macquarie Research, January 2017 Source: MPOB, Macquarie Research, January 2017
Fig 233 Low CPO inventory, tightened global Fig 234 should support CPO prices; we forecast
vegetable oil SU ratio, and stable SBOs value share... CPO price to average US$675/T in 2017E (+7%)
(US$/T)
55% 1200
1100
50%
1000
45% 900
800
40%
700
35% 600
500
30%
400
25% 300
1074 929 755 739 557 633 675 766
Jan-06
Jul-06
Jan-07
Jul-07
Jan-08
Jul-08
Jan-09
Jul-09
Jan-10
Jul-10
Jan-11
Jul-11
Jan-12
Jul-12
Jan-13
Jul-13
Jan-14
Jul-14
Jan-15
Jul-15
Jan-16
Jul-16
Jan-17
200
2011 2012 2013 2014 2015 2016 2017E LT
SBO Value Share Average
Source: Bloomberg, Macquarie Research, January 2017 Source: Bloomberg, Macquarie Research, January 2017
Fig 235 That said, upside to CPO price is somewhat Fig 236 and lower discretionary blending, given low
capped by tightened discount to SBO oil prices
(US$/T) (US$/T)
100 400
300
0
200
-100 100
0
-200
-100
-300 -200
-300
-400 -400
-500
-500
Jan-06
Jul-06
Jan-07
Jul-07
Jan-08
Jul-08
Jan-09
Jul-09
Jan-10
Jul-10
Jan-11
Jul-11
Jan-12
Jul-12
Jan-13
Jul-13
Jan-14
Jul-14
Jan-15
Jul-15
Jan-16
Jul-16
Jan-17
-600
Jan-11
Jul-11
Jan-12
Jul-12
Jan-13
Jul-13
Jan-14
Jul-14
Jan-15
Jul-15
Jan-16
Jul-16
CPO - SBO Diff (Spot) Avg +1 STD -1 STD

Gas oil - CPO
Source: Bloomberg, MPOB, Macquarie Research, January 2017 Source: Bloomberg, Macquarie Research, January 2017
19 January 2017 85
Milk Powder
Powder prices buoyed by supply contraction
Milk powder prices have trended up over the year, with WMP adding around 55% and WMP
up 39%. WMP has performed more strongly, and now traded around the long-run average,
and we expect a continued divergence between the two driven by relative supply changes
and stocks held in the market.
The recovery in WMP has been underpinned by supply constraints in NZ as total milk
production reduces by around 5% and mix is shifted into more favourable product streams.
Fonterra has also attempted to run down inventories to lower its working capital. Over 2016,
NZs WMP exports were down around 3% and Chinese demand recovered modestly, but
remains a significant way from the import volume peak in 2014.
The recovery has been relatively fragile, driven somewhat by weather issues impacting
supply and on-farm economics starting to improve. More elastic regions saw demand rise
materially on the back of soft prices, and we would expect this to slow a little with prices back
up. This has seen WMP prices soften around 8% over January and SMP prices remain flat.
Fig 237 More elastic demand expected to soften Fig 238 NZ WMP exports YTD down 3%
Rolling 12m WMP price Cumulative

800,000 exports (MT) US$/MT 0 1,600,000 export volume
inverted (MT)
1,400,000
700,000 1,000
1,200,000
600,000 2,000
1,000,000
500,000 3,000
800,000
400,000 4,000 600,000
300,000 5,000 400,000
200,000
200,000 6,000
Oct-08
Oct-15
Feb-11
Dec-09
Nov-12
Dec-16
Aug-07
Sep-11
Apr-12
Aug-14
Jan-07
Jun-13
Jan-14
May-09
May-16
Mar-08
Mar-15
Jul-10
0
Exports to price sensitive regions Price 2014 2015 2016
Source: GTIS, Macquarie Research, January 2017 Source: GTIS, GDT, Macquarie Research, January 2017
While China continues to show a recovery, the rest of Asia has been broadly flat, and other
price-sensitive regions have been seeing lower import volumes, attributable to higher prices
and also challenges created by weaker economic conditions (oil linked).
WMP prices have continued their divergence from SMP, which reflects underlying
fundamentals of each market including improved S&D for WMP and higher inventories
elsewhere in the world for SMP. Historically, these products have been priced relatively tightly
due to the underlying protein and fat components (including by-products) but the differential
has remained around US$500/MT but as wide as US$1,000/MT in December 2016. The
divergence in the two powder prices has been normalised by higher fat pricing (butter and
AMF), which has seen relative returns generally on par, but less favourable than cheese.
Stocks of SMP globally have increased on the back of higher milk output being stored as
skim, and these remain elevated. There is currently around 525,000MT of SMP inventory in
the US and EU, and a large amount of this sits in intervention stocks, with the EU increasing
the ceiling to ~350,000MT in mid-2016. The European Commission struggled to find attractive
bids for its SMP stock, with around 22,000MT offered for sale recently (6% of stock) and only
40MT was sold at a minimum price of EUR2,151/MT. Another round of bidding for 20,000MT
of SMP is currently being run. However, Brussels is clearly trying to not cause any further
weakness in the market through selling at too low prices.
WMP output from NZ is likely to remain tight, with the current season likely to remain soft (-
5% milk collection) as well as more favourable commodity and downstream product returns
seeing optimisation away from bulk WMP production. The strong demand for fat products
including butter is underpinning the returns from the SMP stream, which could continue to
weigh on SMP prices.
19 January 2017 86
Fig 239 Futures have flat outlook for milk powders Fig 240 SMP stocks remain elevated
US$/MT '000 MT
5,500 600
5,000
500
4,500
4,000 400
3,500
3,000 300
2,500
200
2,000
1,500 100
1,000
Jul-15
Jul-14
Jul-16
Jul-17
Jan-14
Jan-15
Jan-16
Jan-17
Apr-14
Oct-14
Apr-15
Oct-15
Apr-16
Oct-16
Apr-17
Oct-17
0
Sep-14
Nov-14
Sep-15
Nov-15
Sep-16
Jan-14
Jan-15
Jan-16
May-14
May-15
May-16
Mar-14
Mar-15
Mar-16
Jul-14
Jul-15
Jul-16
WMP WMP futures
SMP SMP futures USA EU
Source: GDT, Bloomberg, Macquarie Research, January 2017 Source: EC, USDA, January 2017
Global export volumes of WMP are down around 1% for the 12 months to November 2016.
NZ exports are down by around 3% and Argentina down 8% due to weaker milk production,
while the EU has increased by 1%. Uruguay has seen significant increases in exports, with
demand from other regions with Latin America with domestic supply conditions challenging.
Australias exports are up by 2% despite lower milk intake with the sell down of some
accumulated inventory. China has increased import demand by around 8% in the last 12
months, while rest of Asia is flat at +1%, Africa +2% and Middle East -11%. Interestingly,
Russia has imported a more WMP, which could be reconstituted into domestic production.
Fig 241 WMP trade flows

Exporter volumes to key regions (12m to November 2016)
New Zealand EU-28 Argentina Uruguay Australia Total
Importer '000MT chg pcp '000MT chg pcp '000MT chg pcp '000MT chg pcp '000MT chg pcp '000MT chg pcp
China 370 3% 15 146% 0 -33% 3 469% 11 122% 401 8%

Rest of Asia 350 -2% 31 56% 0 - 3 172% 42 -1% 425 1%
Africa 298 5% 125 -11% 39 45% 22 -9% 0 -98% 485 2%
Middle East 176 -16% 127 -5% 2 100% 1 142% 6 -11% 313 -11%
Latin America 72 -29% 38 3% 65 -35% 92 46% 3 -29% 271 -11%
Russia 6 2928% 0 - 10 4080% 3 503% 0 - 19 1852%
ROTW 70 -4% 35 13% 0 62% 0 -54% 4 -15% 110 0%
World 1,343 -3% 371 1% 117 -8% 125 38% 67 2% 2,023 -1%
SMP export volumes are also down by 1% for the 12 months to November, with NZ
increasing exports by around 26% as the product stream (due to butter pricing) has been
more favourable to WMP. The United States has increased exports by around 3% while
Australia is down by 6%. Interestingly, despite a 6% increase in SMP production over the
period, the EU has seen exports decline by 15%, with product being sold into intervention
stocks as opposed to export customers. On the import side, most regions are flat to down,
excluding Middle East up 10%.
Fig 242 SMP trade flows

Exporter volumes to key regions (12m to November 2016)
EU-28 United States New Zealand Australia Belarus Total
Importer '000MT chg pcp '000MT chg pcp '000MT chg pcp '000MT chg pcp '000MT chg pcp '000MT chg pcp
China 50 1% 18 -40% 128 6% 16 -16% 0 - 213 -3%

Rest of Asia 179 -17% 209 12% 218 15% 123 -13% 0 - 729 -1%
Africa 195 -18% 11 3% 63 216% 4 61% 0 - 274 0%
Middle East 84 8% 4 11% 51 43% 20 -26% 0 - 159 10%
Latin America 27 -35% 297 3% 17 69% 0 -85% 0 - 342 0%
Russia 0 -79% 0 -100% 0 - 0 - 104 -4% 104 -6%
ROTW 46 -18% 27 -7% 42 18% 4 -60% 5 -31% 124 -10%
World 582 -15% 566 3% 519 26% 169 -17% 109 -6% 1,945 -1%
19 January 2017 87
Nitrogen
Riding the energy price uplift, but structural concerns remain
In our last Commodities Compendium we noted that nitrogen pricing had hit a level that was
too low to be sustainable, particularly given energy prices were trending higher. Since then
we have certainly had a turnaround in prices arguably too strong of one. 2017 is unlikely to
be a year of strong global nitrogen demand, while current pricing is attracting marginal supply
back into the market, most notably in Chinese urea. Thus the current urea rally is unlikely to
be sustained, particularly now Chinese anthracite pricing is following coal lower. Global
ammonia prices have reacted strongly to higher gas prices, with the notable exception of the
US. Even given the combination of domestic production ramp-ups and the threat of
protectionism, the US import price looks too low compared to other global benchmarks.
Longer term, the challenges to industrial nitrogen from increased efficiency of use and
displacement through by-product alternatives are growing more pertinent.
From $180/t at the start of Q4, FOB Black Sea urea prices are now above $240/t. While
some of this may be down to increased demand in key import regions, the majority is down to
the rising industry cost structure. At <$200/t, Chinese urea exports simply werent economic
into most global markets, hence the October-November export volumes trending below the
five year range. And that was before anthracite started to rise. Anthracite pricing had lagged
the Chinese-led thermal coal price gains, to the extent that it was even trading at a heating
value discount at some points. Since the double relaxation of Chinese coal supply in Q4,
thermal coal prices have corrected strongly by anthracite has only just started to turn. Given
thermal coal needs to fall another RMB50/t to meet the Chinese governments price range
midpoint, we can expect anthracite to follow. This may not pull the rug out from under urea
prices, but it does mean the path of least resistance into mid-year is likely to be lower.
One upside risk to the medium-term urea price might come from Chinese supply-side reform.
While there has been little mentioned about this for nitrogen, given the coal-intensive nature
of production we would be surprised if it was not targeted, either for winter curtailments or
wider shutdowns over the coming years. China already has a policy in place to limit nitrogen
application growth, and with structural industry overcapacity the scenario feels much the
same as that of the steel industry, where we have already seen government-led cuts.
The ammonia focus this year is all on the US, with six major plants currently in the ramp-up
phase, with Incitec Pivot reporting their plant is operating above 75% of capacity already.
Around 2.5mt of new capacity is set to be added this year, which will have a knock on effect
on imports, particularly as a ~2% planting shift from corn to soy means less nitrogen intensity.
There is concern that US protectionism could spread to ammonia, however this is more likely
to have impact on 2018 import volumes once full domestic capacity additions are ramped.
Even with the capacity gains, we see the current US price as trading too far below
international comparisons to be sustainable given there still is a need for some imports.
While much of this arbitrage may come from international prices settling down after their
recent $100/t rally, we do think at current Henry Hub prices US ammonia needs to rise.
Fig 243 Urea prices finally followed coal higher Fig 244 while ammonia has risen to a lesser extent
US$/t Urea prilled prices US$/t Ammonia prices
750
360
650
310 550
260 450
350
210
250
160
150
Sep-14
Sep-15
Sep-16
May-14
May-15
May-16
Jan-14
Jan-15
Jan-16
Jan-17
Sep-14
Sep-15
Sep-16
Jan-14
May-14
Jan-15
May-15
Jan-16
May-16
Jan-17
FOB Black Sea FOB China FOB Black Sea CFR Tampa
Source: CRU, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
19 January 2017 88
Fig 245 Anthracite pricing lagged thermal coal on the Fig 246 Gas prices have picked up recently,
way up, and is lagging again on the way down supporting the global nitrogen cost structure
Anthracite versus thermal coal prices Reference gas prices

1,200 (RMB/t) 18
Japan LNG import price
1,000 16
Russian gas at German
14
border
800 12 Henry Hub
$/MMBTU
10
600
8
6
400
4
200 Thermal coal 5500kcal 2
Average fertilizer anthracite price 0
Mar 14
Mar 15
Mar 16
Sep 14
Nov 14
Sep 15
Nov 15
Sep 16
Nov 16
May 14
May 15
May 16
Jul 14
Jul 15
Jul 16
Jan 14
Jan 15
Jan 16
0
2013 2014 2015 2016 2017
Source: Bloomberg, Macquarie Research, January 2017 Source: Macrobond, Macquarie Research, January 2017
Fig 247 2017 is the big year for US ammonia Fig 248 hence we expect US imports to continue
production growth, with ramp-ups happening now their decline
New US Ammonia capacity, kt mT US imports of ammonia

4,500 7
Imports from Trinidad
4,000 6 and Tobago
3,500
5
3,000
2,500 4
2,000
3
1,500
1,000 2
500
1
0
2016 2017 2018 -
Iowa Fertilisers CF Industries Incitec/Dyno Nobel Others 2010 2011 2012 2013 2014 2015 2016 2017f 2018f
Source: CRU, Macquarie Research, January 2017 Source: Customs Data, CRU, Macquarie Research, January 2017
Fig 249 As we suspected, Chinese exports just didnt Fig 250 The 2017 recovery in nitrogen prices should
work at a urea price <$200/t FOB fade with raw material costs
'000t Chinese exports of urea Nitrogen price forecast

US$/t
2,500
400
375 Ammonia FOB Black Sea
2,000 350 Urea prilled, FOB Black Sea
325
1,500 300
385
275
1,000 250
225
273
254
245
240
234
200
230
230
230
230
230
500
226
225
220
220
199
175
- 150
2015
LT
2016f
2017f
2018f
2019f
2020f
2021f
5 Yr Range 5 Yr Avg 2016 2015
Source: China Customs, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
19 January 2017 89
Potash
The market share battle is maturing
The potash market is one we have long held structural concerns about, and we are still in the
middle innings of the necessary market adjustment. Quite justifiably for a market which is
seeing a market share battle into the key consumer, prices are trading at a level to cause
pressure on supply. However, unlike others (such as uranium) where spot prices are
unsustainably low following a 2016 collapse, for potash spot prices have been relatively
stable post the mid-year contract settlement. We are not bearish on potash pricing from
todays levels, and it finally seems the key market players have moved towards acceptance of
the wider problems to be faced. However, any recovery will be slow and will need a boost
from wider agricultural markets, which may be some time off yet.
The last six months have been extremely dull from a potash pricing perspective. After the
Chinese contract settlement at $219/t CFR, spot prices into South East Asia have barely
moved. Meanwhile, customs values into China from key supply countries, which was the best
indicator that the contract system was being heavily discounted in H1 2016, have also been
extremely flat. While in a wider sense potash remains a laggard among commodity markets,
it does appear to have found its cost curve normality, something many others are likely to be
seeking throughout 2017.
There are some reasons to see potash fundamentals improving. Global farmer margins are
finally improving, such that the multi-year potash destock may be coming to an end. This is
perhaps particularly pertinent in the US, where domestic prices are rising and 2016 saw year-
end inventories at the lowest level in the past five years. At the same time, Canadian exports
to the US were down 16% YoY as supply was diverted into other markets. We would expect
some recovery in this volume over 2017. Meanwhile, Uralkali reported another full year
production fall for 2016, while Chinese potash imports are on the rise again after a very weak
mid-2016 as buyers awaited a contract settlement, with Canpotex and BPC reporting full
order books through Q1 2017.
However, it could be argued that higher Chinese imports at flat customs values confirm the
potash market is oversupplied, with excess material being cleared into China, hence the very
high Chinese inventory volumes. Meanwhile, the full extent of ramp-ups at K+S Legacy and
Eurochem are yet to be felt. Moreover, looking at the breakdown of Chinese imports shows
the market share battle is still underway, with a very different make-up of supply volumes
each quarter. 2015 was the year of Belaruskalis push into China, H1 2016 saw Uralkali enter
the battle while Q3 2016 saw Canpotex fight back. More recently, Israeli material has been
pushing back into China after a hiatus. This highlights the ongoing problem of overcapacity in
the potash industry. While supply discipline has become the norm for major producers and
consolidation is set to progress, there is still a long road to travel before strong price recovery.
Fig 251 Since the mid-2016 contract settlement, Asian Fig 252 as have customs values into China from all
spot prices have been flat as a pancake key suppliers, as the market share battle matures
Chinese contract vs SE Asian spot price Average customs value of China's potash imports
US$/t
360
650 Contract CFR China
340
600 CFR SE Asia
550 320
500
300
450
400 280
Global average
350 260 Russia
300 Belarus
240 Canada
250
200 220
Jun-12
Feb-14
Dec-09
01/2015
02/2015
03/2015
04/2015
05/2015
06/2015
07/2015
08/2015
09/2015
10/2015
11/2015
12/2015
01/2016
02/2016
03/2016
04/2016
05/2016
06/2016
07/2016
08/2016
09/2016
10/2016
11/2016
Sep-13
Dec-14
Jul-14
May-15
May-10
Mar-11
Aug-11
Apr-13
Mar-16
Aug-16
Jan-12
Nov-12
Jan-17
Oct-10
Oct-15
Source: CRU, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
19 January 2017 90
Fig 253 Chinese imports have recovered strongly Fig 254 Looking at quarterly import data shows the
after the contract price fall transient nature of supply into China
Chinese imports of potash Proportion of Chinese MOP imports by source

'000t KCl
Russia Belarus Canada Israel/Jordan Other
1,400 100%
6% 5% 9% 10% 7% 8%
8% 6% 6% 7% 4% 6%
1,200 90% 8% 17%
11% 17% 12% 9%
20% 16%
80% 30% 35% 11%
1,000 24% 37%
70% 16% 13%
28% 21%
800 35%
60% 23% 27% 41%
12% 15%
600 50% 23% 26% 27% 13% 17%
15%
400 40% 27% 24%
25% 6%
30% 36% 17%
18%
200 46%
20% 38% 38% 42%
37% 32%
- 27% 23% 27% 23%
10% 19%
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 15%
0%
Q1 14
Q2 14
Q3 14
Q4 14
Q1 15
Q2 15
Q3 15
Q4 15
Q1 16
Q2 16
Q3 16
Q4 16
5 Yr Range 5 Yr Avg 2015 2016
Source: China Customs, Macquarie Research, January 2017 Source: China Customs, Macquarie Research, January 2017
Fig 255 Canadian exports have recovered recently, Fig 256 while the downtrend in Uralkali production
but are still well down YTD, particularly to the US continues
Canada potash exports ('000t KCl) Uralkali potash production

2,100
12.5
1,900
1,700 12
1,500
million tonnes
11.5
1,300
1,100
11
900
700 10.5
500
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 10
2014 2015 2016
2015 2014 2013 2016
Source: Customs Data, Macquarie Research, January 2017 Source: Uralkali, Macquarie Research, January 2017
Fig 257 Potash inventories are low in the US, but now Fig 258 We expect only a slow and steady potash
very high in China recovery as industry consolidation takes hold
Year-end potash inventories Standard potash price forecast

mT KCl US$/t
FOB Vancouver
325
10
8.8
7.9 300
8
275
6
4.5
250
303
4 3.1 3.3 3.1 3.1

282
280
2.6
262
1.9 225
1.8
250
250
240
2
230
230
230
225
220
220
215
200
0
2012 2013 2014 2015 2016 175
1Q16
2Q16
3Q16
4Q16
2015
LT
1Q17f
2Q17f
3Q17f
4Q17f
2018f
2019f
2020f
2021f
USA China
Source: China Customs, Bloomberg, Macquarie Research, January 2017 Source: CRU, Macquarie Research, January 2017
19 January 2017 91
Important disclosures:
Recommendation definitions Volatility index definition* Financial definitions
Macquarie - Australia/New Zealand This is calculated from the volatility of historical All "Adjusted" data items have had the following
Outperform return >3% in excess of benchmark return price movements. adjustments made:
Neutral return within 3% of benchmark return Added back: goodwill amortisation, provision for
Underperform return >3% below benchmark return Very highhighest risk Stock should be catastrophe reserves, IFRS derivatives & hedging,
expected to move up or down 60100% in a year IFRS impairments & IFRS interest expense
Benchmark return is determined by long term nominal investors should be aware this stock is highly Excluded: non recurring items, asset revals, property
GDP growth plus 12 month forward market dividend speculative. revals, appraisal value uplift, preference dividends &
yield minority interests
Macquarie Asia/Europe High stock should be expected to move up or
Outperform expected return >+10% down at least 4060% in a year investors should EPS = adjusted net profit / efpowa*
Neutral expected return from -10% to +10% be aware this stock could be speculative. ROA = adjusted ebit / average total assets
Underperform expected return <-10% ROA Banks/Insurance = adjusted net profit /average
Medium stock should be expected to move up total assets
Macquarie South Africa or down at least 3040% in a year. ROE = adjusted net profit / average shareholders funds
Outperform expected return >+10% Gross cashflow = adjusted net profit + depreciation
Neutral expected return from -10% to +10% Lowmedium stock should be expected to *equivalent fully paid ordinary weighted average
Underperform expected return <-10% move up or down at least 2530% in a year. number of shares
Macquarie - Canada
Outperform return >5% in excess of benchmark return Low stock should be expected to move up or All Reported numbers for Australian/NZ listed stocks
Neutral return within 5% of benchmark return down at least 1525% in a year. are modelled under IFRS (International Financial
Underperform return >5% below benchmark return * Applicable to Asia/Australian/NZ/Canada stocks Reporting Standards).
only
Macquarie - USA
Outperform (Buy) return >5% in excess of Russell Recommendations 12 months
3000 index return Note: Quant recommendations may differ from
Neutral (Hold) return within 5% of Russell 3000 index Fundamental Analyst recommendations
return
Underperform (Sell) return >5% below Russell 3000
index return
Recommendation proportions For quarter ending 31 December 2016

AU/NZ Asia RSA USA CA EUR
Outperform 57.53% 50.72% 45.57% 42.28% 60.58% 52.79% (for global coverage by Macquarie, 8.71% of stocks followed are investment banking clients )
Neutral 33.90% 33.97% 43.04% 50.11% 37.23% 35.62% (for global coverage by Macquarie, 8.05% of stocks followed are investment banking clients)
Underperform 8.56% 15.30% 11.39% 7.61% 2.19% 11.59% (for global coverage by Macquarie, 4.63% of stocks followed are investment banking clients )
Company-specific disclosures:
Vikas Dwivedi invests from time to time in oil and gas commodities and / or related derivative and futures products consistent with published views.
Important disclosure information regarding the subject companies covered in this report is available at www.macquarie.com/research/disclosures.
Analyst certification:
We hereby certify that all of the views expressed in this report accurately reflect our personal views about the subject company or companies and its or
their securities. We also certify that no part of our compensation was, is or will be, directly or indirectly, related to the specific recommendations or
views expressed in this report. The Analysts responsible for preparing this report receive compensation from Macquarie that is based upon various
factors including Macquarie Group Ltd total revenues, a portion of which are generated by Macquarie Groups Investment Banking activities.
General disclaimers:
Macquarie Securities (Australia) Ltd; Macquarie Capital (Europe) Ltd; Macquarie Capital Markets Canada Ltd; Macquarie Capital Markets North America
Ltd; Macquarie Capital (USA) Inc; Macquarie Capital Limited and Macquarie Capital Limited, Taiwan Securities Branch; Macquarie Capital Securities
(Singapore) Pte Ltd; Macquarie Securities (NZ) Ltd; Macquarie Equities South Africa (Pty) Ltd; Macquarie Capital Securities (India) Pvt Ltd; Macquarie
Capital Securities (Malaysia) Sdn Bhd; Macquarie Securities Korea Limited and Macquarie Securities (Thailand) Ltd are not authorized deposit-taking
institutions for the purposes of the Banking Act 1959 (Commonwealth of Australia), and their obligations do not represent deposits or other liabilities of
Macquarie Bank Limited ABN 46 008 583 542 (MBL) or MGL. MBL does not guarantee or otherwise provide assurance in respect of the obligations of
any of the above mentioned entities. MGL provides a guarantee to the Monetary Authority of Singapore in respect of the obligations and liabilities of
Macquarie Capital Securities (Singapore) Pte Ltd for up to SGD 35 million. This research has been prepared for the general use of the wholesale
clients of the Macquarie Group and must not be copied, either in whole or in part, or distributed to any other person. If you are not the intended recipient
you must not use or disclose the information in this research in any way. If you received it in error, please tell us immediately by return e-mail and delete
the document. We do not guarantee the integrity of any e-mails or attached files and are not responsible for any changes made to them by any other
person. MGL has established and implemented a conflicts policy at group level (which may be revised and updated from time to time) (the "Conflicts
Policy") pursuant to regulatory requirements (including the FCA Rules) which sets out how we must seek to identify and manage all material conflicts of
interest. Nothing in this research shall be construed as a solicitation to buy or sell any security or product, or to engage in or refrain from engaging in any
transaction. In preparing this research, we did not take into account your investment objectives, financial situation or particular needs. Macquarie
salespeople, traders and other professionals may provide oral or written market commentary or trading strategies to our clients that reflect opinions
which are contrary to the opinions expressed in this research. Macquarie Research produces a variety of research products including, but not limited to,
fundamental analysis, macro-economic analysis, quantitative analysis, and trade ideas. Recommendations contained in one type of research product
may differ from recommendations contained in other types of research, whether as a result of differing time horizons, methodologies, or otherwise.
Before making an investment decision on the basis of this research, you need to consider, with or without the assistance of an adviser, whether the
advice is appropriate in light of your particular investment needs, objectives and financial circumstances. There are risks involved in securities trading.
The price of securities can and does fluctuate, and an individual security may even become valueless. International investors are reminded of the
additional risks inherent in international investments, such as currency fluctuations and international stock market or economic conditions, which may
adversely affect the value of the investment. This research is based on information obtained from sources believed to be reliable but we do not make
any representation or warranty that it is accurate, complete or up to date. We accept no obligation to correct or update the information or opinions in it.
Opinions expressed are subject to change without notice. No member of the Macquarie Group accepts any liability whatsoever for any direct, indirect,
consequential or other loss arising from any use of this research and/or further communication in relation to this research. Clients should contact
analysts at, and execute transactions through, a Macquarie Group entity in their home jurisdiction unless governing law permits otherwise. The date and
timestamp for above share price and market cap is the closed price of the price date. #CLOSE is the final price at which the security is traded in the
relevant exchange on the date indicated.
Country-specific disclaimers:
Australia: In Australia, research is issued and distributed by Macquarie Securities (Australia) Ltd (AFSL No. 238947), a participating organisation of the
Australian Securities Exchange. New Zealand: In New Zealand, research is issued and distributed by Macquarie Securities (NZ) Ltd, a NZX Firm.
Canada: In Canada, research is prepared, approved and distributed by Macquarie Capital Markets Canada Ltd, a participating organisation of the
Toronto Stock Exchange, TSX Venture Exchange & Montral Exchange. Macquarie Capital Markets North America Ltd., which is a registered broker-
dealer and member of FINRA, accepts responsibility for the contents of reports issued by Macquarie Capital Markets Canada Ltd in the United States
19 January 2017 92
and sent to US persons. Any US person wishing to effect transactions in the securities described in the reports issued by Macquarie Capital Markets
Canada Ltd should do so with Macquarie Capital Markets North America Ltd. The Research Distribution Policy of Macquarie Capital Markets Canada
Ltd is to allow all clients that are entitled to have equal access to our research. United Kingdom: In the United Kingdom, research is issued and
distributed by Macquarie Capital (Europe) Ltd, which is authorised and regulated by the Financial Conduct Authority (No. 193905). Germany: In
Germany, this research is issued and/or distributed by Macquarie Capital (Europe) Limited, Niederlassung Deutschland, which is authorised and
regulated by the UK Financial Conduct Authority (No. 193905). and in Germany by BaFin. France: In France, research is issued and distributed by
Macquarie Capital (Europe) Ltd, which is authorised and regulated in the United Kingdom by the Financial Conduct Authority (No. 193905). Hong Kong
& Mainland China: In Hong Kong, research is issued and distributed by Macquarie Capital Limited, which is licensed and regulated by the Securities
and Futures Commission. In Mainland China, Macquarie Securities (Australia) Limited Shanghai Representative Office only engages in non-business
operational activities excluding issuing and distributing research. Only non-A share research is distributed into Mainland China by Macquarie Capital
Limited. Japan: In Japan, research is Issued and distributed by Macquarie Capital Securities (Japan) Limited, a member of the Tokyo Stock Exchange,
Inc. and Osaka Exchange, Inc. (Financial Instruments Firm, Kanto Financial Bureau (kin-sho) No. 231, a member of Japan Securities Dealers
Association). India: In India, research is issued and distributed by Macquarie Capital Securities (India) Pvt. Ltd. (CIN: U65920MH1995PTC090696), 92,
Level 9, 2 North Avenue, Maker Maxity, Bandra Kurla Complex, Bandra (East), Mumbai 400 051, India, which is a SEBI registered Research Analyst
having registration no. INH000000545. Malaysia: In Malaysia, research is issued and distributed by Macquarie Capital Securities (Malaysia) Sdn. Bhd.
(Company registration number: 463469-W) which is a Participating Organisation of Bursa Malaysia Berhad and a holder of Capital Markets Services
License issued by the Securities Commission. Taiwan: In Taiwan, research is issued and distributed by Macquarie Capital Limited, Taiwan Securities
Branch, which is licensed and regulated by the Financial Supervisory Commission. No portion of the report may be reproduced or quoted by the press
or any other person without authorisation from Macquarie. Nothing in this research shall be construed as a solicitation to buy or sell any security or
product. The recipient of this report shall not engage in any activities which may give rise to potential conflicts of interest to the report. Research
Associate(s) in this report who are registered as Clerks only assist in the preparation of research and are not engaged in writing the research. Thailand:
In Thailand, research is produced, issued and distributed by Macquarie Securities (Thailand) Ltd. Macquarie Securities (Thailand) Ltd. is a licensed
securities company that is authorized by the Ministry of Finance, regulated by the Securities and Exchange Commission of Thailand and is an exchange
member of the Stock Exchange of Thailand. The Thai Institute of Directors Association has disclosed the Corporate Governance Report of Thai Listed
Companies made pursuant to the policy of the Securities and Exchange Commission of Thailand. Macquarie Securities (Thailand) Ltd does not endorse
the result of the Corporate Governance Report of Thai Listed Companies but this Report can be accessed at: http://www.thai-
iod.com/en/publications.asp?type=4. South Korea: In South Korea, unless otherwise stated, research is prepared, issued and distributed by Macquarie
Securities Korea Limited, which is regulated by the Financial Supervisory Services. Information on analysts in MSKL is disclosed at
http://dis.kofia.or.kr/websquare/index.jsp?w2xPath=/wq/fundMgr/DISFundMgrAnalystStut.xml&divisionId=MDIS03002001000000&serviceId=SDIS03002
001000. South Africa: In South Africa, research is issued and distributed by Macquarie Equities South Africa (Pty) Ltd, a member of the JSE Limited.
Singapore: In Singapore, research is issued and distributed by Macquarie Capital Securities (Singapore) Pte Ltd (Company Registration Number:
198702912C), a Capital Markets Services license holder under the Securities and Futures Act to deal in securities and provide custodial services in
Singapore. Pursuant to the Financial Advisers (Amendment) Regulations 2005, Macquarie Capital Securities (Singapore) Pte Ltd is exempt from
complying with sections 25, 27 and 36 of the Financial Advisers Act. All Singapore-based recipients of research produced by Macquarie Capital
(Europe) Limited, Macquarie Capital Markets Canada Ltd, Macquarie Equities South Africa (Pty) Ltd and Macquarie Capital (USA) Inc. represent and
warrant that they are institutional investors as defined in the Securities and Futures Act. United States: In the United States, research is issued and
distributed by Macquarie Capital (USA) Inc., which is a registered broker-dealer and member of FINRA. Macquarie Capital (USA) Inc, accepts
responsibility for the content of each research report prepared by one of its non-US affiliates when the research report is distributed in the United States
by Macquarie Capital (USA) Inc. Macquarie Capital (USA) Inc.s affiliates analysts are not registered as research analysts with FINRA, may not be
associated persons of Macquarie Capital (USA) Inc., and therefore may not be subject to FINRA rule restrictions on communications with a subject
company, public appearances, and trading securities held by a research analyst account. Information regarding futures is provided for reference
purposes only and is not a solicitation for purchases or sales of futures. Any persons receiving this report directly from Macquarie Capital (USA) Inc. and
wishing to effect a transaction in any security described herein should do so with Macquarie Capital (USA) Inc. Important disclosure information
regarding the subject companies covered in this report is available at www.macquarie.com/research/disclosures, or contact your registered
representative at 1-888-MAC-STOCK, or write to the Supervisory Analysts, Research Department, Macquarie Securities, 125 W.55th Street, New York,
NY 10019.
Macquarie Group
19 January 2017 93
EMEA Research
Heads of Equity Research Industrials Commodities & Precious Metals
Peter Redhead (Global Head) (852) 3922 4836 Capital Goods Colin Hamilton (Global) (44 20) 3037 4061
Shai Hill (Europe) (44 20) 3037 4232 Rowan Goeller (Johannesburg) (27 11) 583 2131 Jim Lennon (London) (44 20) 3037 4271
George Brits (South Africa) (27 11) 583 2223 Christian Breitsprecher (Frankfurt) (49 69) 50957 8014 Matthew Turner (London) (44 20) 3037 4340
Alternative Energy & Utilities Transportation & Infrastructure Stefan Ljubisavljevic (London) (44 20) 3037 4247
Rowan Goeller (Johannesburg) (27 11) 583 2131 Vivienne Lloyd (London) (44 20) 3037 4530
Gurpreet Gujral (London) (44 20) 3037 2249 Lynn Zhao (Shanghai) (86 21) 2412 9035
Dominic Nash (London) (44 20) 3037 4239 Materials Ian Roper (Singapore) (65) 6601 0698
Jose Ruiz (London) (44 20) 3037 1912 Chemicals/Containers, Packaging/Paper & European Macro Group
Peter Crampton (London) (44 20) 3037 4559
Forest Products, Construction Materials
Dilip Kejriwal (London) (44 20) 3037 2252 Economics & Strategy
Rowan Goeller (Johannesburg) (27 11) 583 2131
Consumer Danny Moran (New York) (1 212) 231 0698 Peter Eadon-Clarke (Global) (813) 3512 7850
Cooley May (Houston, US) (1 212) 231 2586 Matthew Turner (London) (44 20) 3037 4340
Sreedhar Mahamkali (London) (44 20) 3037 4016 Elna Moolman (Johannesburg) (27 11) 583 2570
Andreas Inderst (London) (44 20) 3037 2629 Global Metals & Mining
Daniele Gianera (London) (44 20) 3037 2517 Alon Olsha (London) (44 20) 3037 2637 Strategy
Matthew Webb (London) (44 20) 3037 2367 Patrick Morton (London) (44 20) 3037 2014 George Brits (South Africa) (27 11) 583 2223
Toby McCullagh (London) (44 20) 3037 2649 Gerhard Engelbrecht (Johannesburg) (27 11) 583 2407 Quantitative
Natasha Moolman (Johannesburg) (27 21) 813 2774 James Oberholzer (Johannesburg) (27 11) 583 2367 Gurvinder Brar (Global) (91 97) 8055 5902
Qaqambile Dwayi (Johannesburg) (27 11) 583 2229 Yatish Chowthee (Johannesburg) (27 11) 583 2208 Giuliano De Rossi (London) (44 20) 3037 1997
Energy Hayden Bairstow (Perth) (618) 9224 0838 Steve Gao (London) (44 20) 3037 2765
Pharmaceuticals Jakub Kolodziej (London) (44 20) 3037 2016
Iain Reid (London) (44 20) 3037 2119
Josiah Rudolph (Johannesburg) (27 11) 583 2210
David Farrell (London) (44 20) 3037 4465 Ada Li (Johannesburg) (27 11) 583 2181
Giacomo Romeo (London) (44 20) 3037 4445
Kate Sloan (London) (44 20) 3037 4453
Real Estate Find our research at
Aditya Suresh (Hong Kong) (852) 3922 1265 Property Trusts & Developers Macquarie: www.macquarieresearch.com/ideas/
Gerhard Engelbrecht (Johannesburg) (27 11) 583 2407 Nazeem Samsodien (Cape Town) (27 21) 813 2771 Thomson: www.thomson.com/financial
Financials Mahir Hamdulay (Cape Town) (27 21) 813 2705 Reuters: www.knowledge.reuters.com
TMET Bloomberg: MAC GO
Banks Factset: http://www.factset.com/home.aspx
Edward Firth (London) (44 20) 3037 4077 Guy Peddy (London) (44 20) 3037 4509 CapitalIQ www.capitaliq.com
Piers Brown (London) (44 20) 3037 4044 Tim Nollen (New York) (1 212) 231 0635
Contact macresearch@macquarie.com for access
Vardhman Jain (London) (44 20) 3037 2587 Mark Murphy (London) (44 20) 3037 5498 requests.
Namita Samtani (London) (44 20) 3037 2197 Giasone Salati (London) (44 20) 3037 2670
Sven Thordsen (Johannesburg) (27 11) 583 2218 Bob Liao (London) (44 20) 3037 2868
Diversified Financials Andrew DeGasperi (New York) (1 212) 231 0649 Email addresses
Neil Welch (London) (44 20) 3037 4272 George Brits (South Africa) (27 11) 583 2223 FirstName.Surname@macquarie.com
Arun Melmane (London) (44 20) 3037 4221 Zintle Gantsho (Johannesburg) (27 11) 583 2107 eg. peter.redhead@macquarie.com
Hugh Miller (New York) (1 212) 231 2323 Richard Majoor (Johannesburg) (27 11 583 2225
Insurance
Andy Hughes (London) (44 20) 3037 2256
Philip Kett (London) (44 20) 3037 2505
Equities ETF Trading South Africa Equity Syndication

Stevan Vrcelj (Global Head) (612) 8232 5999 Robin Quinnell (London) (44 20) 3023 8885 Franco Lorenzani (Johannesburg) (27 11) 583 2014
Mark Duncan (Global Head) (852) 3922 5888 US Sales Trading South Africa Sales
Dipesh Patel (London) (44 20) 3037 4978
JT Cacciabaudo (New York) (1 212) 231 6381 Roland Wood (Cape Town) (27 21) 813 2611
Sarah-Jane Wagg (Johannesburg) (27 11) 583 2000
Mike Gray (New York) (1 212) 231 2555 Sven Forssman (Johannesburg) (27 11) 583 2369
European Execution Services Chris Reale (New York) (1 212) 231 2555 Ed Southey (Johannesburg) (27 11) 583 2026
Andrew Stancliffe (London) (44 20) 3037 4784 EU Cash Sales Khensani Mokoena (Johannesburg) (27 11) 583 2016
Robert Tappin (London) (44 20) 3037 4827 Atish Jogi (Johannesburg) (27 11) 583 2252
Wayne Drayton (London) (44 20) 3037 4980 Sam Bygott-Webb (London) (44 20) 3037 4767
Jesse Ushewokunze (New York) (1 212) 231 2504
Ryan McSorley (London) (44 20) 3037 8424 Richard Alderman (London) (44 20) 3037 4875
Matthew Hanley (London) (44 20) 3037 4949 Gaelle Jarrousse (London) (44 20) 3037 4960 South Africa Sales Trading
David Dunne (London) (44 20) 3037 4750 John Gilbert (London) (44 20) 3037 4933 Harry Ioannou (Johannesburg) (27 11) 583 2015
Jon Knapman (London) (44 20) 3037 1704 Awais Khan (London) (44 20) 3037 4967 Welcome Plessie (Johannesburg) (27 11) 583 2058
KC O'Rourke (London) (44 20) 3037 4910 Daniel Sorger (London) (44 20) 3037 4903 Martin Hughes (Johannesburg) (27 11) 583 2019
Danny Want (London) (44 20) 3037 4847 Jonathan Mathews (London) (44 20) 3037 4869 Marcello Damilano (Johannesburg) (27 11) 583 2018
Peter Homan (London) (44 20) 3037 4740 Iain Whiteley (London) (44 20) 3037 4771 Commodity Hedge Fund Sales
Gregorio Esmelian (London) (44 20) 3037 4973 Rupert Woolfenden (London) (44 20) 3037 4603
Chris Looney (New York) (1 212) 231 0836
Will Rogers (London) (44 20) 3037 1721 Paras Amlani (London) (44 20) 3037 4846
Iain Lindsay (London) (44 20) 3037 4825
Robert Lenihan (London) (44 20) 3037 4908 Chris Wilson (London) (44 20) 3037 4925
Guy Keller (Singapore) (65) 6601 0303
Adam Brown (London) (44 20) 3037 4836 Toby Ingram (London) (44 20) 3037 4957
Cybelle Dib (London) (44 20) 3037 4839 Clara Twamley (London) (44 20) 3037 4832 Commodity Corporate Sales
Holger Hoepfner (Geneva) (41 22) 818 7777 Nael Noueiri (London) (44 20) 3037 4913
Richard Bateson (London) (44 20) 3037 4821
David Hemming (London) (44 20) 3037 4909 Christian Schmuck (New York) (1 212) 231 8007 Rohan Khurana (Singapore) (65) 6601 0308
Nick Bryan (London) (44 20) 3037 4768 Arjan Dorrestijn (New York) (1 212) 231 8054 Commodity Investor Products
Chris Charbonnier (London) (44 20) 3037 4760 Chris Carr (New York) (1 212) 231 6398
Arun Assumall (London) (44 20) 3037 4953
Marc Crome (London) (44 20) 3037 4778 Doug Stone (New York) (1 212) 231 2606
Catherine Littlefield (New York) (1 212) 231 6348
George Sampson (London) (44 20) 3037 1732 Jan Halaska (Boston) (1 617) 598 2503
Bahar Ghaffari (London) (44 20) 3037 4969
ETF Sales
Bachir Binebine (London) (44 20) 3037 4680
Jessica Lana (London) (44 20) 3037 4808
This publication was disseminated on 19 January 2017 at 23:21 UTC.

Commodities Compendium 190117 RX e 263687

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Commodities Compendium 190117 RX e 263687

Hochgeladen von

Copyright:

Verfügbare Formate

GLOBAL

Cash flow, caution and (a lack of) capex are widespread

Our key commodity recommendations

Thermal Coal Nickel Zinc

Source: Macquarie Research, January 2017

Outperformers on a 12- to 24-month basis from current levels

Zinc The raw material constraint story becomes ever clearer.

Cobalt A consistent deficit expected, and major country risk remains.

Uranium Prices are finally invoking a supply response.

Potential for medium-term raw material constraints

Commodities for which the longer-term challenges are underappreciated

Aluminium Industry cost deflation is an ongoing headwind.

Nitrogen - Increased efficiency of use and displacement by alternatives present problems.

Metals & Bulks: Passing through the sweet spot 16

Energy: Co-ordination, compliance and caution 24

Agricultural: Waiting for a weather kick-start 26

It would be a stretch to call the recovery overwhelming. Of those 50 countries in our

Manufacturing, 3m average IP, 3m average

Fig 8 US IP growth by type, % YoY Fig 9 China IP growth by type, % YoY

Fig 10 Manufacturing PMIs vs IP/manufacturing

Fig 12 10yr bond yields, % Fig 13 US 10 year yield, decomposition, %

Fig 23 Macquaries real GDP forecasts

Iron ore Iron Ore +142%

(40%) (20%) 0% 20% 40% 60% 80% 100%

Fig 27 Daily property sales cooling Fig 28 Inflation is clearly picking up

60 Total Residential YoY % PPI CPI

3,000 2012 2013 2014 2015 2016

Metals and Bulks Executive Summary

Fig 33 Macquarie metals and bulk commodity price forecasts

Iron ore - Australian fines c/mtu fob 81 86 88 79 71 70 77 65 69 84 83 86

40% 33% 31%

70,000 Proportion of supply losing money - cash basis

LME Price vs 10%Ni RKEF costs, $/tonne HRC price premium

19000 100% US over China

7000 LME Price 10%Ni RKEF Costs 0%

Global steel overcapacity - set to improve Aluminium stocks on exchange

LME CME SHFE

Energy Executive Summary

Agricultural Executive Summary

MacPI index, 1997-2000=100 China's soybean imports

Brazil Argentina USA Other

Fertilisers - Quarterly & Annual Average Price Forecasts

Source: CRU, Macquarie Research, January 2017

Dairy Executive Summary

Fig 57 but a little ahead of that implied by feed

3,500 2,000 400

WMP WMP average

US EU NZ '000 MT, Chinese WMP demand still

China - WMP imports Russia - Cheese imports

Supply side contraction continues

Fig 60 EU farmgate milk prices seeing strong

Germany France Ireland

US milk production continues to track up

NZ production impacted by unfavourable weather over peak period

14% Change pcp Rolling 12m

Dairy industry still tough in Australia

8.0 stepped and

Argentinian dairy industry facing challenges

15% 4.50 35.00

-20% 0.00 0.00

ARS$/lt EUR/100lt (RHS)

Milk collection growth for key exporters slowing to 1%

Key importer demand growth modest