Mining is risky. With a commodity output, they operate in an environment of perfect competition. It’s capital intensive and they don’t need to contend with depreciation alone, there’s also depletion. Pit walls can fall, people can get injured, and governments can penalize mines at will (nationalization, permitting, fines).
In a previous post I looked at Franco Nevada’s negative retained earnings and meager returns on capital invested. Given historic performance I thought it would be interesting to look at implied returns from prospective investments and the sensitivity of project failure.
Let’s say that Franco invests $250M in eight projects and the expected post-tax return is 4%. Well, what needs to go wrong to obliterate the return of the portfolio.
Well, this would do it:
1 project fails immediately
1 project fails at year eight
1 project fails at year ten
1 project ends two years early
What are the odds that this could happen? That’s an interesting question. It’s certainly higher than 5%. I imagine 50% is closer to reality. There’s also the impacts with grade issues and ramp up.
What is the purpose of a corporation? Let’s not get into the
stakeholder vs. shareholder debate. At the core, a corporation’s goal is to
increase its shareholder’s wealth. Otherwise, why start the corporation? I’d only start a company if I believed this investment would yield more
value (return) than a risk free asset. I’d want to be compensated for this risk as
well, hopefully earning a materially higher return than a riskless asset.
A corporation’s purpose, while simple, is easily convoluted
by countless factors. The owners of the corporation may not understand it in themselves.
I’m not Stanley Druckenmiller but I’m
no finance/business slouch either. After a couple
thousand hours of studying/working in finance, I
can honestly say that I’m only starting
to build a picture of what I consider to
be a “good” business. Why is this the case?
There’s so much noise.
A company’s price and fundamentals are so
detached. Who am I to say that a
company with zero earnings is a bad company? The
FANGs, they don’t earn much and yet
they are the best performing stocks in
the past decade. The fact that growth stocks don’t
trade on fundamentals makes it easier for
non-growth stocks to trade on the same basis. The market doesn’t hold them to account, don’t they see
I think Warren Buffet
has it right. The Berkshire Hathaway owner’s manual states that their goal “is
to maximize Berkshire’s average annual rate of gain in intrinsic business value
on a per-share basis.” This makes sense. If
I’ve got a dollar, I want to invest it in the area that’s going to
give me the most dollars in return. Berkshire
defines intrinsic value as the present value of future cash flows. They trade money. Trade 1 dollar for 1.08 dollars. Makes sense.
A $100 million investment with a $100 million NPV is subjecting
earned dollars to zero return. Now that doesn’t
make sense. Who would do that?
Well, as far as I can tell
Franco Nevada does this, and a whole-heck
of the mining industry.
Why do I say this? Well let’s take a look at the financial statements
for FNV for year end 2017.
Ok, so I’m a shareholder. How much money have I pumped into this company?
Wow! 6.5 Billion in total, the value has steadily marched higher over the past decade. Must be some good returns to justify this. Well, FNV has generated $836M in net income of the past ten years. That’s a pretty measly 1.88% on the contributed capital. Hmmmm.
But, but, they pay a
Yeah, they do. But it’s more than their net income.
So what does this really mean. Well FNV is paying out more than their making. Is this good business? Let’s keep issuing capital, take on some debt, and pay some dividends. More people see the dividend track record and subscribe. That’s an interesting cycle… sounds familiar. What’s the name for that?
Well, if you were a
shareholder between 2007 and 2016 it hasn’t
been good business. Earnings haven’t increased and the return on your equity
has been pretty lousy.
But who cares! The stock price is going up.
And a big reason for
that is steeper valuations from a PE perspective (note the chart above takes
out PE ratios over 100). In an environment where your competition is miners with a destructive track record for
capital allocation, you’re the best looking
girl at the bar.
Now I’m not saying
that the people that run FNV are bad miners. They just aren’t deploying capital
in a fashion that produces high returns. But
this begs the question, why are new players (triple flag, orion, osisko,
sandstorm) entering the business? With competition, prospective returns surely
aren’t going to go higher. The narrow margin of error may get pushed lower.
There’s also the question about why people own these stocks. I mean, it’s precious metals exposure and it’s certainly outperforming Barrick but common. Return on equity less than a LT treasuries. These are not risk free assets.
Yesterday Newcrest announced that they were increasing their stake in Solgold (owner of Cascabel property in Ecuador (85%)). This will take Newcrest’s ownership of Solgold to 15.33%. Newcrest had previously purchaed shares of Solgold in 2016 and 2017. What makes this interesting is that BHP has also shown interest in the property, purchased 100,000 shares in October. Newcrest will not be the larget shareholder in the company, followed by DGR global (11.24%) and BHP (11.18%).
So what is going on here? Well SolGold has an enterprise value of US$810M so this company is not cheap. Cascabel must be one hot ticket. Yup sure is...
Resource update in November showed 10,900,000 tonnes of copper and 23.2M oz of gold. Wow. Amazing. The resource contains 2.95B tonnes at 0.52% Cu. Perhaps even more interesting, there’s a high-grade core of 420M tonnes at 1.47%Cu; almsot $100/tonne rock at $3/lb Cu.
The November 2018 investor presentation highlights the fact that modern exploration activities have allowed the for the discovery of the property. Drill results contain some of the best porphyry copper gold intervals ever recorded. For example. Hole 12, 1560m at .59% Cu and 0.54 g/t.
The image above (from November 2018 investor presentation) shows the gargantuan extent of the deposit. Based on the geochem data it looks like there are a bunch more targets as well.
Given Newcrest’s experience with block caving, it’s not surprising how interested they are with the property. Look at the benchmarking!
Solgold also owns multiple subsidiaries with a 3,200 km2 land package.
Well. Solgold is pretty amazing. That’s what’s going on with Solgold.
What is the state of mining in Armenia? How should foreign investors view this country? I’ll spend some time reviewing some recent developments in the country and formulate a view.
Gained independence in 1991
Had been governed by Serzh Sargsyan since 2007 as part of the right-wing republican party
Serzh Sargsyan was reelected for a fourth term in April 2018, peaceful protests started which were concerned about what was starting to look like an indefinite rule
Serzh steps down and the leader of the Civil Contract Party is elected (Nikol Pashinyan)
It would be impossible to look at Armenia without investigating the state of affairs with Lydian. Lydian received approval to build the project (Amulsar) in 2014 but construction has been impacted by local protests.
In August, Armenia’s inspectorate for Natural Protecion and Mineral Resources suggested that the environmental assessment be re-evaluated.
The inspectorate stated that new ecological factors should be considered for the property. Specifically, there are new sightings of red plants and animal species.
But get this, so the head (Artur Grigoryan) of the Environmental and Mining Inspection agency directs Lydian to refrain from any mining activities until the ministry can conclude if these new “red plants” are actually at the site. So Grigoryan sends in his team. The team concludes that these new organisms cannot live at this site and are not there. Lydian appeals the original directive, but the appeal is heard by Grigoryan. Obviously, he rejects the appeal which is predicated on information from his own ministry. So now Lydian has challenged Grigoryan’s position through an administrative court. The court has accepted the appeal which suspends Grigoryan’s edict. Great! It doesn’t matter though because the place is still blockaded.
It’s pretty amazing that a country with 16.8% unemployment is challenging industry and preventing the creation of hundreds of jobs.
Seems to me that Armenia is not a jurisdiction you want to be developing a project in.
In general, mining companies have a poor track record of delivering projects on time or on budget. Today, for the first time, I heard about the McNulty Curve; a graph which predicts the level of pain that companies will endure when ramping up a project.
Some previous studies on project cost overruns highlighted a few key reasons for process plant failure (defined as major cost over-run or inability to achieve design capacity):
insufficient effort was devoted to understanding process chemistry
insufficient continuous pilot-scale testing was conducted
the plant lacked parallel process streams and/or in-line spare equipment units
the design incorporated sequential unit operations that either were first-of-a-kind or the largest ever built or both
Terry McNulty advanced this concept and derived ramp up curves for various types of processes. He defined the types as follows:
The owners relied on mature technology.
Standard types of equipment were selected.
Thorough pilot-scale testing was done on potentially risky unit operations.
If the technology was licensed, the project was one of the first licensees.
Some equipment was a prototype in size or application.
Pilot-scale testing was incomplete or was conducted on non-representative samples.
Process conditions were unusually severe or corrosive.
Non-innovative parts of the flowsheet received inadequate attention
There was very limited pilot-scale testing and important steps were ignored.
Feed characteristics such as mineralogy were poorly understood.
During process development, product quality received little attention.
There were serious design flaws.
Engineering, design, and construction were on a “fast track” with inadequate planning to offset added risk.
If continuous tests were run, they were only to make the product.
Equipment was downsized or design criteria were compromised to reduce cost overruns.
The flowsheet was unusually complex with prototype equipment in two or more unit operations.
Process chemistry was poorly understood.
McNulty highlights a few other factors that were correlated with project over runs, some of which are particularly relevant for the mining industry.
Corporate management had a promotional or overly aggressive attitude.
The owners had very little day-to-day engineering input.
Driving forces underlying the project were ill-conceived.
The ore receiving and preparation areas received little attention.
Translation of the testwork to design criteria was flawed.
Sources: Most of this information was from a paper called Minimization of Delays in Plant Startups. It can be found here: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.126.1359&rep=rep1&type=pdf#page=119
Positive asymmetric returns
are those with limited downside and much higher upside. This article looks at
the prospects for a gold developer whose share price has been battered. Orca
Gold is developing the Block 14 gold property located in Sudan. Their share
price is down 40% YTD even after significantly advancing their flagship
In early November, Orca
released the details of an updated feasibility study which describes an open
pit operation producing 167K oz per annum. Reserves contain 80 million tonnes
grading 1.11 g/t Au for 2.85M ounces. The property is expected to produce gold
at an all-in sustaining cost of $789/oz. The after-tax NPV of the project is
estimated to total $403M; at a 5% discount rate.
If you look at the chart below
you will notice that the share price did not budge after the release of the
study (early November) and the stock continued its downward march. So what’s
the deal? The enterprise value of Orca is $37M and their 70% stake in Block 14
is worth US$282M. They are trading at 13% of the NPV5% of the property. This is
a massive discount. The implied discount rate of the property is approaching
50%. So, the market is pricing in a serious chance that this asset does not
make it to production.
This is interesting. The
discount rate assumed for this property is egregiously high. Maybe the worst is
already priced into the stock price?
So why is the market pricing
in such a steep discount? Well, Sudan -the location of their asset- is still
designated by the United States as a State Sponsor of Terrorism. In 2017 the
U.S removed some of the economic sanctions for the country but it is unclear if
financial institutions could lend money to Orca to fund the project (the
government has a carried interest in the property). I think that as long as
Sudan is on this list, Orca’s share price will suffer.
This is where it gets
interesting. Sudan is progressing talks with the US to be removed from the
terrorism sponsor list1 (which would remove the associated
sanctions). If this were to occur, it’s reasonable to assume that Orca would be
re-rated to a much higher level.
The project is probably still
riskier than the typical gold mine. It’s located in the desert, away from
infrastructure, needs a well field, and Sudan does not have an established
That being said, and as shown
in the chart below, Orca is amongst the most discounted juniors out there with
a near-term development project. This investment is not for the faint of heart.
If talks with Sudan do not progress than the share price could easily continue
to suffer. A successful resolution, however, could result in 1X share price
appreciation. This is asymmetry.
There are numerous reasons why a process plant may miss
recovery targets. Gold can be more encapsulated in insoluble material, other elements
can consume important reagents -limiting availability for gold-, and sometimes
the ore is characterized as “Preg Robbing.” Refractory ore is an issue and has
been for a while. Notable mines with refractory ore are Prestea, Tarkwa, and
What does this mean?
Sometime ores contain carbon. This carbon absorbs gold
particles which would have otherwise been absorbed onto the “reagent” carbon.
In “The Chemistry of Gold Extraction,” that author states that as little as
0.1% carbon may produce preg-robbing properties.
Carbonaceous ores can be separated into two categories:
mildly and highly carbonaceous.
Mildly Carbonaceous: contain small quantities of
organic carbon, typically less than 1%.
Highly Carbonaceous: contain carbon >1% and
contains carbon that has a strong gold absorbing tendency. This material can
reduce gold recoveries to below 80%.
Ways to deal with preg-robbing:
Use a CIL
process rather than CIP: a carbon in leach flowsheet completes the leach
and absorption process in the same location simultaneously. In completing this
process simultaneously, there is less time for the organic carbon to absorb the
hydrocarbons can passivate the
surface of carbon, decreasing the propensity to absorb gold.
Treatment: Aqueous solutions of chlorine have strong oxidizing
capabilities. It’s not well understood why this process works. Between a pH of 3
and 5 hypochlorous species are predominant and are thought to passivate the
Roasting: Burn it off!
John O.Marsden, I. H. (n.d.). The Chemistry of
Gold Extraction. SME.
The valuation of mining companies often starts with a
calculation of Net Asset Value (NAV). NAV assessment is a cumulative discounted
cash flow analysis of a company’s portfolio of assets (mines) minus associated
corporate and other overheads. Enterprise value (value of equity + debt – cash)
can be compared to NAV. If enterprise value is higher than NAV then the company
trades at a premium. If enterprise value is less than the NAV, than the company
trades at a discount.
Generally, NAV is calculated using a 5% discount rate but
there are often occurrences where this value will be increased (geopolitical risk,
asset risk, permitting). The fact that NAV is generally calculated at 5% is
very impactful. If Franco Nevada trades at a premium to the 5%NAV than that
means that the market is pricing in a lower discount rate. A 2% discounted cash
flow analysis is probably a value that brings Franco closer to its market price.
Conversely, if Alio Gold is trading at a large discount to their 5%NAV then the
market is implying that something closer to a 10% discount rate could be
Sometimes the market can be become irrational and punish projects/companies. A discount/premium to NAV analysis can be a useful way to identify “value” plays in the space.
The banks have teams of analysts that produce DCF models for
each of the mining companies. I do not have this luxury. I do have, however,
published technical reports, and market prices. Almost all technical reports
include a summary of NPV and most of the time it utilizes a 5% discount rate.
This is great. For single asset companies, I now have a view of the company’s
NAV (ignoring adjustments).
Single asset companies are the most interesting as they provide the most direct view of the company’s NAV; less noise. The chart below plots project NPV as disclosed in technical reports versus company enterprise value. These are recently published technical reports that contain over 1M oz of gold. You’ll notice that the highest value is 50%. Minera IRL Limited’s Ollachea property commands the smallest discount while a group of companies are currently trading at only 10-20% of their project’s NPV.
It’s an interesting mental exercise, hypothesizing why some
companies are undervalued relative to their projects. A few ideas:
Time/Stage: technical reports usually display
the NPV as of the time that a construction decision is made, not the current
time period. If we assume that the standard PEA is 6 years from a construction decision
and use an 8% discount rate, the discount value is 58% when compared to the
published NPV. It shouldn’t be a surprise that Valentine Lake and Mt. Todd are
at such a large discount as they are at the PEA stage. Conversely, the highest
ranked project (Ollachea) is currently in construction.
Jurisdiction: This goes without saying and
explains why a project like Block 14, located in the Sudan, is not generating
much share value for Orca Gold.
So how do we use this information? A value investor could
look at this and see if the market is mispricing any assets. We could also look
at the development path of an asset and predict a share price appreciation.
A useful example would be Valentine Lake. Marathon released a PEA on the property at the end of October which showed an after tax NPV5% close to $500M. The company currently trades at an enterprise value of $80M. A pretty strong discount but probably warranted given the project’s state. Their recent investor presentation shows that a PFS will be published in 2019 and a FS in 2020. They are having exploration success, the project looks to be improving. Obviously, there’s a due diligence component to this investment but there is a hypothesis for a re-rating over the next two years. If their discount jumps from a measly 15% to a more justified 30% in the next two years that’d provide a pretty strong return.
Seems like Valentine Lake could be an interesting opportunity.
Orezone gold corporation is a Canadian listed explorer with an enterprise value of US$51.8M. YTD they are down about 16%. They’re primary property is Bombore, located in Burkina Faso. They own 90% of the project with the government carrying a 10% free interest. Let’s see what we have here.
In July Orezone released a Feasibility study.
Ok. Lots of gold. Grade is very poor.
What does the reserve look like?
Hopefully this has a null strip ratio. Nope. 1.68:1.
Recoveries aren’t special. Get pretty low at lower grades
(80%). Large grind size (125um) and low cyanide consumption should make this a
pretty cheap process plant. 12,000 tpd.