A recent piece here on Seeking Alpha bore the title "2013 Third Quarter All-In Gold Mining Costs: Numbers Suggest Miners Are High Grading Their Mines". And in this article the author concludes as follows:
"What we think is going on is that miners are "high-grading" their mines to mine higher grade portions of the mine, which produces more gold at the same total costs - which ends up reducing the costs per ounce. This is exactly what we're seeing in the numbers as Q3FY13's total costs of $8.79 billion are very similar to Q1FY13's costs of $8.76 billion, even as the industry increases production."
In order to understand the significance of this assertion, it is worth to look closer at the meaning of the term "high-grading" as it is used in precious metal mining commentary. The link in the quote above provides some explanation, however, we prefer the explanations given in this article, mostly because it is non-commercial, more detailed and accompanied by illustrative examples.
In short, high-grading refers to the practice of mining selectively and targeting high-grade areas of a mine while leaving lower-grade ore untouched. In some cases, the lower-grade ore that is left behind by this practice cannot be mined economically anymore and therefore becomes waste.
The "numbers" that the author is referring to in the quote above are some cost metrics invented by the author himself (we have expressed our concern about this metric before in a different context). The author observes that costs (as defined in his self-defined metric) have fallen across the board for gold miners. Of course this is old news since the same observation can be made using commonly accepted metrics.
There is a myriad plus one possible explanations for costs falling in the precious metal mining sector (GDX), and we found it interesting to see the author honing in on the issue of high-grading in the referenced article. Unfortunately, the author failed to check his conclusions with first principles.
Since mining companies routinely provide data about head grades at their individual mines, this would have been easy to do. But not to worry, we have done the work for him and here are some results.
Barrick Gold - Falling Grades
The table below lists average grades mined at Barrick Gold's 21 mines which reported results for the past three years. The listed data includes Q3 grades, and also grades for the nine months ending September 30 for the years 2011, 2012 and 2013 plus the respective averages.
Comparing Q3 data we observe that 12 mines reported lower grades in 2013 and only 9 mines reported higher grades when compared to the same period 2012. Moreover, 13 mines reported grades lower than the three-year average, and only 8 mines reported higher grades.
Moving on to data for the 9 months ending September 30, we make very similar observations: 14 mines report lower grades for 2013 than for the same period in 2012; and 15 mines reported grades lower than the 3 year average.
Quite clearly, numbers for Barrick Gold are not giving any indication for high-grading.
Goldcorp - No Indication of High-Grading As Well
The table below lists head grades at Goldcorp's mines for the 5 past quarters. We included all 10 operations for which the whole data set was available.
Of the 10 mines considered in the table 7 mines reported lower grades in Q3/2013 when compared to the same quarter one year earlier; and 7 mines reported grades below the average over the past 5 quarters.
Again, we note that we are seeing the contrary of high-grading in the published data.
Newmont - High-Grading? Not!
The table below is similar in format as the table for Barrick Gold, showing mill and leach grades for the September quarters 2011, 2012 and 2013; as well as for the 9 months ending September 30; plus the average grades over the past three years. Nevada operations are added together which leaves us with data for 12 sites for our comparison.
Checking Q3 data we observe a grade reduction at 7 sites, one site with grades un-changed and 4 sites with higher grades when comparing to the same period in 2012. When comparing Q3/2013 data with the 3-year average for the same period we count 7 sites with lower grades as opposed to 5 sites with higher grades.
Turning to reported data for the 9 months ending September 30 we note reduced grades in 2013 on a year-on-year basis at 7 sites, and only 5 sites with higher grades. And when comparing 2013 data with the 3-year average we note an even split between sites with higher grades and sites with lower grades.
The Big Three in Summary
Our review included data from 43 mine sites operated by the big three gold miners. About two thirds of the mines reported lower grades year-on-year and also when compared to a longer-term average.
This data indicates the exact opposite of high-grading as diagnosed in the referenced article.
We also checked data from other gold miners and noticed the same trend.
A mine plan is typically based on a mineral reserve. A mineral reserve is the economically mineable part of a measured or indicated mineral resource. The keywords are "economically mineable" here. In order to show the economical viability a price must be assumed above which the company feels comfortable selling their metals and still make a profit.
This price assumption is crucial for the reserve definition and the mine plan based on the reserves. Ore grades vary throughout an ore body. The higher the price assumption, the greater the portion of the known ore body that can be mined economically. This price assumption is typically hidden in the foot notes of the reserve statements. Mining companies usually publish their reserve statements annually. Mine plans are directly connected to these yearly reserve updates.
Interestingly, price assumptions vary greatly between miners. Barrick Gold (ABX) for example, used a long-term price assumption of $1,500/oz for its latest reserve statement. Goldcorp (GG) assumed $1,350/oz and Newmont (NEM) assumed $1,400/oz for their calculations.
Let us assume that one of these companies has decided to lower the price assumption for the reserve statement due out shortly. This would lead to a reduction of reserve ounces because some parts of the ore bodies will now be considered non-economical to mine, and the mine plan will not include these portions of the ore body. In effect, the mine has been high-graded.
Will We See High-Grading in the Future? Absolutely.
Conservative miners have used price assumptions of $1,000/oz all along, and effectively ignored ore that could be mined had they lifted the threshold. Randgold (GOLD) comes to mind, or Yamana Gold (AUY). Both these companies, along with others, have been exceedingly conservative with their reserve statements. In times of high gold prices these conservative companies could be criticized for "under-stating" their reserve base, and effectively high-grading their mines. In times like we are experiencing right now, these miners can rightly boast their foresight.
We expect the less conservative gold miners to re-consider their price assumptions for the upcoming 2013 reserve statements. This will lead to reduced reserve bases for these miners, associated with write-downs. (We have already predicted the same for Barrick Gold here.) Based on these modified reserves we will then see the effects of the associated mine plans, quite probably using higher-grade ore than before.
High-grading sounds like a questionable practice at first sight. It is important to understand that the described practice is in fact something that happens all the time as part of the process of calculating reserves.
Changing the reserve grade and the associated mine plan is not an act that can be instigated on the spur of the moment. Rather, it comes at the end of a process that takes time. In our view high-grading is over-hyped and is in fact something that every mining house needs to do to some degree in order to adjust for volatility in the underlying metal price.
Rather than lamenting about high-grading, investors are advised to carefully check the upcoming reserve updates and understand the underlying implications. A conservative gold price assumption is part of a very good insurance against ongoing volatility in 2014. Ignore the commentators who suggest otherwise.