In a recent Sprott Money podcast, Eric Sprott discussed the merits of investing in high-cost producers during gold bull markets (https://www.sprottmoney.com/Blog/when-you-contemplate-negative-interest-rates-i-would-rather-own-gold.html). The intuition is simple, a high-cost producer may only produce $100/oz of cash at a $1,200/oz gold price. When prices jump to $1,300/oz, net income effectively doubles; why wouldn’t the share price jump with it? The math does not lie, but I thought it would be worthwhile to investigate the validity of this statement in the context of current market performance.
YTD, gold prices are up
approximately 5% or $60/oz. That is a sizeable amount, enough to produce a
material increase in profits for high-cost gold producers.
Let us take a look at Detour Gold
as an example. In 2018, the company produced 610,000 oz of gold at an
all-in-sustaining cost of $1,158/oz (https://www.detourgold.com/investors/news/press-release-details/2019/Detour-Gold-Reports-Fourth-Quarter-and-Year-End-2018-Financial-Results/default.aspx).
With this production, cost, and a realized price of $1,268/oz, they produced
adjusted net earnings of US$64.2M. At today’s gold prices, using last year’s
production costs, we could expect earnings to increase by 65%. This statement
is a simplification of reality as production costs are inherently variable as
is production. The underlying notion, however, that low-cost producers have
more leverage to the gold price is true (a small increase in gold prices can
make these miners much more profitable).
So given the increase in gold
prices, my hypothesis is that we should see a positive correlation between
production costs and 2019 returns. Well, let us see what the data shows.
This YTD price return tornado
shows some promising results. I see high-cost producers like Detour in the
green, but there are many similar peers that are in the red.
This chart is much more useful.
It plots 2019 YTD price returns against 2018 AISC per oz. The relationship,
however, is the exact opposite of what I would expect. The lower the cost the
producer, in general, the better they have done this year.
Before you start challenging this
chart, I will acknowledge a few caveats. First, this is the GDX peer group. The
silver price has been hammered this year, and some poor performing companies
(Hecla) have a large exposure to the silver metal. There are also notable
idiosyncratic factors that have resulted in poor (St. Barbara Feasibility study
disappointment) and better than average performance.
The underlying premise, however,
that high-cost producer will outperform in a rising gold environment appears to
be less certain than the underlying logic would propose.
So how should you think about
this information in the context of your investment framework? To start, the
data suggests that jumping into high-cost producers does not make sense at the
moment. Perhaps the gold price needs to consolidate before market participants
rerate these companies. The data suggest that less risky, lower cost producers,
may be the first to benefit (in terms of stock price) during a bull market.
In this case, Kirkland Lake
stands out as the obvious choice; being the lowest cost producer in the GDX
Intuitively I’ve known that the gold mining industry doesn’t allocate capital well. I’ve touched on some of the return performance of royalty companies in the past. Their low return royalty model looks angelic compared to other actors. As I browse through company balance sheets I note the scars of the past (dramatically negative retained earnings). I’ve never really looked at this information in much detail. I know Kinross made some bad bets, I know Barrick lost its track with Pascua Lama. The full magnitude of the capital destruction, however, was way worse than I thought it would be.
Take a look at this snapshot. In 2013, this group -which is by no means comprehensive- wrote off over $24B in 2013.
Over this same time period, the group destroyed $70B worth of capital. Barrick alone tallied $30B.
If we look how much was written off, compared to these
company’s current enterprise value the data shows the Kinross leads the pack.
Between 2005 and 2018 the company wrote off $12B, which is 50% more than the
Interesting stuff. Terrifying really. There are, however, a handful of companies that did pretty well during this time period. All of the streamers kept write-down % under 10%. Kirkland Lake was one of the only companies that survived the period without an issue.
“Pit Optimization” is a mysterious catch-all term used by mining engineers, financiers, and geologists. You input a bunch of economic variables into various Black-Box algorithms and with a click of a button, and 20-30 minutes, out spits a pit shell that is “optimized.” From this shell, you can tell the world how much metal is economic and then run some mine plans to figure out the economic reality for the project. My interest in digging into this topic a bit deeper was spurred by the realization that experienced industry leaders were treating the Nevada Net Proceeds Tax differently.
What is Whittle? What is Lerchs-Grossman?
Lerch-Grossman is the name of a modeling approach for solving open-pit optimization. It was developed in 1965 and implemented in the 1980’s by Jeff Whittle. The algorithm uses block dependencies to determine which blocks must be mined out as a group (i.e. if you mine G you must mine B, C, and D). The graphical relationship between blocks is determined by slope parameters (i.e. if you are using a very shallow slope than mining G may necessitate the mining of A through E).
Figure 1: Dependency Example (source: CONSTRUCTION ECONOMIC ORE BODY MODELS FOR OPEN PIT OPTIMIZATION, D. Whittle)
I like to think about optimization in a backwards fashion. Flip the topography upside down and whatever falls out is the optimized pit. Instead of gravity, you have revenue, this is the force propelling the ore out of the ground. In the opposite direction, let’s call it friction, we have costs; forces that go against the revenue generated by the ore. Like any physics equation, the fundamental forces need to be correct and well thought out.
With this explanation in mind, let’s think about a tricky situation. Below you’ll see excerpts from two technical reports. Both open-pit mines located in Nevada. One accounts for the State’s Net Proceeds Tax, one does not. So who is right here?
Well, the calculation of the net proceeds tax looks like you can deduct just about every expense that occurs during mining so the tax is more of a net profit royalty. Thinking about the flipped topo, the net proceeds tax only applies to blocks that actually generate a profit. So, as my beautiful the drawing shows, the tax applies after the blocks have “fallen out.” The tax applies to a component of the profit, so there is no way that there can be enough force to push the blocks back up. Therefore, I just can’t see why the Net Proceeds Tax would be included in the determination of the ultimate resource. Looks like Mine 2 has it right.
In playing devil’s advocate, I wonder if the tax should be used in the determination of interim pit phases? Imagine you have two phases, one high profit, one low profit. Does this change your thinking? It still shouldn’t matter. As long as the second phase is above breakeven, it should be mined.
Ok, so for now, I think the case is closed. NNPT and NPI royalties
should not be included as cost factors in open pit optimization.
I was thinking about investment frameworks today. I’ve never explicitly laid out my own process. If I had to I’d say the following points are the most important in my framework (mining related):
The project has to score well (decent grade, IRR, preferably in a good jurisdiction, tried and tested mining method, simple process). There can be complexity but the simpler the better.
Value has to be a concern. There’s so much uncertainty in junior investing. We need to get 3-10x payoffs on the winners to offset the inevitable losses.
Momentum. This one is tougher. I’m not sure where I stand on this. Revaluations happen fast.
Increasing base. This can mean production, resource, reserve. Not picky.
Preservation of ownership percentage. I get that dilution is going to occur. I’d just like to get under as many warrants and options as possible.
I was looking at some valuations today, single asset development projects.
One that jumped out at me was New Polaris. I’d not heard of Carnac resources until today. They released a technical report on New Polaris in March.
Valuation: They have an enterprise value of ~$9 million. The post-tax NPV5% is ~$210 million; works out to a EV/NPV value of less than 5%. Well this is interesting.
Now obviously production, if ever, is far out. Within the comparable peer group, however, these guys are cheap.
This is an example of the asymmetric returns that I like. What’s the worst that can happen? You lose 100%. What’s the upside here?
Now I’m not a chartist but there looks like there could be a tide change.
On the warrant and option from there’s not too much that is below the current stock price of $0.065/share.
So, this project is cheap. Has some momentum indicators and there’s not a lot of drag on the option/warrant front.
They have ~2M in cash and spent about the same last year. Not the worst cash balance out there but not the best.
Carnac has a mixed bag of miscellaneous projects. The only one that really stands out is New Polaris. ~10 g/t, historic mine, challenging infrastructure, challenging metallurgy (probably). Does not check the box as a simple project.
Overall, I think this is pretty interesting. Will watch.
I think it’s fair to say that we’ve all grown accustomed to
miners overpromising and underdelivering. Whenever a new project is coming
online there is a palpable uneasiness that is detectable in analysts,
management, and investors. Will the project hit the guidance throughput? Mining
I don’t want to group all projects into the underdeliver bucket but there is certainly a large percentage that not only fail to deliver on ramp-up projections but also in ultimate capacity. The chart below shows actual vs. projected mining rates for a large project that was recently constructed. Now I get that management can be overly optimistic and ramp-ups can be challenging. You’ll notice that the year 1 tonnage is way off of the projection but, by year 2 and 3, the company has closed much of the gap. What I’m more interested in is the levelling off of production and the inability to ever achieve expectations regarding ultimate mining rates.
Management will often highlight specific variables that are
performing worse than expected. Shovel downs are decreasing availability.
Pioneering is impacting productivity. A large weather storm reduced operating
time. These are all possibilities. What I’m
interested in, however, is why production falls short of budget when all variables are as expected.
But how can this be?
How can production be less than expected when all inputs are as expected? Well,
I think that the companies contracted out
to perform mining studies underestimate volatility’s role in determining production
Let’s look at a simple example:
The capacity table shows one shovel that is paired with three trucks. You’ll notice that truck and shovel capacity is
fairly balanced, both around 25K tons mined per shift. You could expand the time period
to encompass a year and this is how most
feasibility studies determine equipment requirements. Now they would use variable
parameters for each of these factors based on the specific conditions in the
mine during the period (haul distance) and planned maintenance.
Ok, nothing wrong here, right?
These assumptions grossly underestimate the influence
randomness within each variable. Fluctuations in, say, availability have asymmetric
impacts on the production of the mine. -10% one day and +10% the next does not
average out to zero impact on the mine. It averages something less.
This histogram of shift-by-shift availability reflects the average that is shown in the table (77%). This histogram reflects the outcome of 730 simulations (1 year’s worth of shifts) of another probability density function that is based on actual data.
These availability values were used to calculate shift production
for an entire year. The chart below shows the mine’s production as a function
of shovel availability. You’ll notice
that tons mined per shift increases linearly as shovel availability is increased -the mine is shovel limited-. Once
availability exceeds 70%, however, the mine
is limited by truck capacity.
This is the root of
the issue. Mine production falls short when shovel availability is lower than
average and cannot sufficiently increase production when availability is higher
than average. This asymmetry makes it
effectively impossible for feasibility parameters to be achieved.
Average simulated production is barely over 20K tons per shift. A far cry (-18.5%) from the 25K tons per shift that would be predicted by the study. Coincidentally, this is the same delta between actual and planned tonnage in Y5 of the earlier chart.
Mines are fragile. They are hurt by volatility and cannot
make up production on the positive side because rates
are capped by a new bottleneck. It’s
my opinion that this static mindset, when
it comes to production scheduling, is the root cause of lots of the industry’s
So what do we take away from this very simplified example?
Well, it pays to have buffer
(excess) capacity. Miners with small equipment fleets are the most susceptible to
volatile operating parameters as they don’t
have other units to average out the shift-by-shift outcomes. In general, a
static approach to forecasting will overestimate long term production.
So I woke up this morning thinking about gold producer price
to NAV5% discounts/premiums. I’ve struggled to understand the intuition behind
being priced at a premium to NAV5% as this implies that the appropriate
discount rate for the stream of cash flows is less than 5%. How can this be
given that gold mining is inherently risky. Arguably more so than the typical
business in the S&P 500.
Well there are a couple of explanations for this pricing
Analyst NAV assessments underestimate the future
stream of cash flows
This is possible but I doubt this is the answer
as any future resource conversion is going to occur so far in to the future
that it would be discounted to oblivion
Analysts use a flat price deck when creating the
You could make the case that a nominal price
deck should be used but this change would be offset by the use of a nominal
discount rate; probably not the answer.
NAV5% is not the correct discount rate to be
used for specific gold producers.
I think this is the underlying logic driving the
price to NAV logic.
In the gold space, small companies typically trade at a
discount to their NAV. Generally, discount to NAV is negatively correlated with
size; smaller company, larger discount. The intuition behind this equilibrium
makes sense as larger companies are more diversified (operationally and
jurisdictionally). Senior producers generally trade at a premium to NAV and
streaming companies trade at an even higher premium.
Being priced at a premium is a challenging concept to understand.
At the extreme, this can imply that the purchase of the security will provide
cash flows that are less than the purchase price. This is very similar to
negative interest rates. Unlike negative interest rates, however, the pricing
of these securities is not manipulated by quantitative easing and central
banks. The “rational” investor is doing the pricing.
So how can a discount rate of zero (or negative) be
justified. The capital asset pricing model states that the equity risk premium
is the risk free rate + beta * equity risk premium. The current T-bill rate is
around 1% so this implies that the beta of these investments must be zero or
slightly negative. This train of thought led me to start looking into beta and
the correlation of gold to the broader economy.
So what is beta? I like to think of it as a leverage factor
for stock returns. A beta of 1.25 means that the stock will move 125% of the
move of the general market. Beta can also be thought of as “risk” factor. A
high beta implies high risk -higher volatility- and investors will require a
higher rate of return for an investment in the security.
If we look at the beta of monthly returns from 1973 to the end of 2018 we see that gold has exhibited a negative beta with respect to the S&P500. Conversely, copper, oil, and silver have positive betas. These relationships make sense as these other commodities are more related to economic growth and, in turn, the S&P500.
The beta values are misleading as they understate gold’s performance relative the S&P500 during times of instability.
Because, as you’ll see in this chart, gold generally hovers around a beta of zero during periods of relative calm and spikes in other occasions (2008).
chart that gold acts as insurance during calamity, increasing in value by more
than the S&P falls.
This post is a continuation of yesterday’s dive into the
relationship between company-wide grade and cash flow from operations. Spoiler
alert, there was no relationship. This fact isn’t surprising given the
multitude of other factors that impact a company’s cash flow from operations.
I decided to get a little more granular and look at this
relationship on an asset basis.
The chart shows that increased production correlates with lower AISC, an expected result given economies of scale. Large scale project require large capex and, as such, require low operating costs to provide attractive rates of return. This makes sense.
If we compare AISC for open pit operations vs. processed grades we see the trend that we were looking for! Increased grades results in lower AISC. Finally.
Similarly with UG operations, higher grade and lower opex.
The purpose of this analysis is to answer the question: “Is
I’ve heard this line
often and am interested in correlating mined grade to cash flow from operations
to see if there is any relationship.
I’ll start with a
broad assessment, ignoring the obvious impact of mine type (UG/OP).
The chart shows a couple of
interesting points. Pretium and Kirkland Lake are really in a league of their
own in terms of reserve grade for producers with >200K oz per annum of
The chart isn’t that useful with these higher reserve grade operators. Let’s take them out.
The chart, with a 5 g/t cap looks as follows:
Moreover, unfortunately, there really isn’t any relationship between reserve grade and CFO/oz (darn). This isn’t that surprising as reserve grades aren’t mined grades and there are countless other factors that impact the profitability of an operation. I was hoping that grade would be king-enough to prove some relationship.
This investigation will require further analysis. While we’re at it. Let’s look at 2018 production.
The chart shows that, as expected, more gold production = more CFO. It’s interesting to note which operators fall above or below the trend line. Kirkland Lake, Newcrest, and Sibanye stand out as a strong performers while Anglo Gold Ashanti is a laggard.
The last chart we’ll
look at is a comparison between 2018 CFO per oz produced vs. current enterprise
value per ounce. Now we wouldn’t expect CFO
to directly translate to enterprise value
think that this chart is interesting because it provides commentary regarding
how much reserve value is going to be captured by the company (or at least the
market’s perception of it). The shading of the dots reflects the comparison
between these two metrics.
The variation in this metric is very impressive. Take New Gold for example, in 2018 the CFO per oz was close $550 while the company’s enterprise value per reserve ounce is only $100/oz. As mentioned, there are innumerable other items to consider (asset quality, assets not in production, CFI, capital structure).
Not really sure how to conclude this. Relating grade to cash flow is going to be tricky.