I have decided to write this article after reading how numerous contributors to Seeking Alpha, writing about investment in gold and silver companies, attach much importance to numbers for All Inclusive Sustainable Cost (AISC) per ounce produced. This despite a number of articles, such as by PWC, and two excellent notes, one a kitco.com article dated 4 February 2015 with the title as "The Real Cost of Mining Gold," the other "The Terrible Truth of Gold Mining Cost Reporting" published by Brent Cook's Exploration Insights, Turning Rock Into Money, pointing out the limitations to this measure.
This article will first give a background to how the AISC measure came about, the serious limitations to it and demonstrate why it is actually meaningless to use as an investment criterion. This article differs from the notes mentioned above by proposing adjustments that, should this be adopted by mining companies, will allow investors to gauge how these companies compare to what they promised they would do.
How AISC Came About
A cost standard proposed in 1996 by the Gold Institute was an early attempt at standardising cost reporting. The proposed standard for "Cash Operating Cost" was widely adopted throughout the global gold industry, and regulatory bodies such as the Toronto Stock Exchange and others recommended its use by listed companies. While the Gold Institute ceased operations at the end of 2002, the standard has remained in use until today and was also adopted with minor modifications by the Silver Institute in 2011. (Source: Seeking Alpha article: The Cost of Mining Gold: a 101, And a Critique.)
- Cash Operating Costs include: direct mining and milling costs, stripping and mine development costs, third party smelting and refining costs, transport costs, and by-product credits.
- Total Cash Costs include all the above costs plus royalties and production-related taxes.
- Total Production Costs include all the above costs plus: depletion & amortization and mine closure costs.
The reason why Cash Operating Cost per ounce of gold was widely adopted by the industry is because it purported to provide transparency to the economics of gold mining operations. However, it suffers from severe shortcomings as it excludes some expenses and capital items that drastically reduce a company's bottom line profitability.
In recognition of this shortcoming, the World Gold Council issued a "guidance note" on 27 June 2013 on All-in Sustaining Cost to incorporate costs relating to sustaining production, such as underground development and stripping costs and replacement of mining and other equipment to maintain production. The guidance note includes a table, reproduced below, that went into great detail what to include in the various measures and where in the financial statements these could be found.
This note will first give a background to how the AISC measure came about, the serious limitations to it and demonstrate why it is actually meaningless to use as an investment criterion. This article will differ from the notes mentioned above by proposing adjustments that, should this be adopted by mining companies, will allow investors to gauge how these companies compared to what they promised they would do.
The mining industry generally adopted All-in Sustaining Cost, with many starting to include All-in Cost in their statements as of 1 January 2014, as suggested by the World Gold Council.
It is of interest to note that one of the reasons given for adopting AISC is to show governments that mining companies are not as profitable as indicated by cash costs alone, which could ease pressure for higher royalties and taxes.
So what is the problem with AISC metrics?
Problem 1: The Definitions Do Not Include The Impact of:
- Income tax.
- Working capital (except for adjustments to inventory on a sales basis).
- All financing charges (including capitalised interest).
- Costs related to business combinations, asset acquisitions and asset disposals.
- Items needed to normalise earnings, for example impairments on non-current assets and one-time material severance charges.
In particular, income taxes and finance charges are relevant to the shareholder as they can have consequences for the amount of cash the company generates and will have available to reward shareholders through dividends, or share buybacks.
Problem 2: By-Product and Co-Product Revenue Deduction Makes Comparison Over Time Subject to External Factors Beyond Management Control.
When a company earns more than 80% of total revenue from a primary metal, the other metals are considered by-products. These can be deducted from the cash cost of production to get a net-cost of production into which the number of production units of the primary metal can be divided to get a unit production cost for that metal. If the primary metal accounts for less than 80% of revenue, then all metals are considered co-products, each carrying the cost of production in proportion to their contribution to total revenue.
Table 2 shows the effect of unit cost of production calculation for a mine that happens to generate exactly 80% of its revenue from gold, the balance from silver. The company could decide to report its costs in terms of by-the product calculation, or the co-product calculation.
The table above shows that under by-product accounting, the company may deduct the revenue received by selling the by-products (silver) from the operating costs, then divide by the units of gold produced. This calculation: (US$35 million - US$13.79 million)/50,000 would allow the company to report US$424 per ounce cash costs of production. It implies that the company has a margin of US$676/oz gold, which is at 50,000 oz x US$676 gives us the same amount of margin as for the conventional calculation.
So what is the problem with by-product deduction? Let's look at the same table which shows the same calculation, but after the silver price has risen for some reason 2.5 fold, whereas the gold price has remained the same. I know that this is unlikely, but want to exaggerate the calculation to emphasize the result.
As you can see, the cash cost of production per ounce of gold has dropped to a negligible level for the by-product calculation (let's ignore that theoretically they should have moved to co-product reporting) and much less, but still noticeably for the co-product calculation without management having to do anything. Management can take the glory for a substantial decline in the cost per ounce simply because of external factors. The above illustrates that comparing unit cost over time would be influenced by relative metal price movements.
Problem 3: Inconsistency Between Companies of the Definitions Adopted
The next problem is that the adoption of unit cost reporting is totally optional to the companies and also what definition they use. Each company is free to report their costs in the manner they wish. All they need to do is put some sort of vague statement in their reports and caution that their definition may differ from what others use.
The table below indicates what is evident from the Management Analysis and Discussion ("MDA") reports of a number of companies (represented by their stock code on the Toronto Stock Exchange) that are subject to the obligation to submit such reports to Sedar in Canada.
I cannot vouch that I have captured it all correctly, but any oversight is simply the result of the company not having been explicit enough about what it has included and what not.
The point of including the table is to show how inconsistent the definition is applied, which makes a mockery of comparing figures between companies, which is the reason for the cost reporting in the first place.
Problem 4: The Measure Does Not Really Capture Cash Generation by the Company
To demonstrate that AISC is meaningless as a gauge on how much a company generates in cash the performance of the same ten companies in Table 4 has been reviewed over seven quarters since March 2014. The reason for this starting period is that this is when most companies started to report AISC figures as per World Gold Council suggestion.
The approach to the comparison is to reproduce the cash flow statements of each company, sum the seven quarters' results and arrive at the change in cash over the period. This is then compared to the notional amount of generated cash based on the difference between revenue and cost per AISC, accepting that the AISC can include a relatively minor amount of non-cash items.
As some of the difference can be attributed to finance raised, which is captured in the cash flow statement, but not by the AISC, this has been deducted from the difference between the two cumulative numbers.
Table 5 and Table 6 give the results for the ten companies, which have been selected to encompass a wide variety in size, type of operations, and jurisdictions where they are operative.
As a bit of a digression, the table also shows that of the ten companies, only four companies paid dividends, of which two negligible amounts compared to the notional cash amount generated determined using AISC. Of the four, Endeavour Mining Corporation and Centamin (OTCPK:CELTF) could obviously afford to pay dividends as the first paid back loans without its cash balance dropping to the same extent and the latter needed no financing and increased its cash balance at the same time.
At face value, Semafo only managed to pay dividends from financing, but upon closer inspection, it is apparent that the dividend was paid in the September 2015 quarter with good cash flow allowing the company to pay back loans and increase cash.
Goldcorp (NYSE:GG) stands out in that it paid its dividend by partially funding this from finance raised and running down its cash balance. This information is simply not available from looking at AISC, as using this metric implies that the company had a margin of US$1.9 billion.
Of all the companies in Table 5, Claude Resources (OTCQB:CLGRF) is the only one that shows a very close match between actual cash generated and the notional margin derived from the difference between revenue and AISC. Semafo (OTCPK:SEMFF) differs by only 12% from the quoted change in cash in the Cash Flow statement before the effect from Financing Activities, but Argonaut (OTCPK:ARNGF) and First Majestic (NYSE:AG) show very large unexplained differences, which cumulatively amount to approximately 1.3 times that of the notional margin.
Table 6 includes in general larger companies than in Table 5, but, with the exception of Centamin for which only three quarters could be calculated, all show much larger variances, with an unexplained difference generally more than 1.5 times the notional generated cash calculated using AISC.
The above two tables illustrated that deductions using AISC are meaningless and serve no purpose in comparing between companies. What is the explanation for the large variance? Apart from income taxes not captured in the measure, the big item is of course acquisitions and investments in new projects with the companies most active in this respect having the largest variances. These initiatives may well be required to replace production from mines that are needed to be developed just to remain at current levels, yet they are treated as "growth" projects and expenditure is therefore excluded. It is this aspect that makes the use of AISC in its present format so flawed. The exclusion of outlays on "growth" projects also allows management to hide its incompetence. As is observed in an article in kitco.com "The Real Cost of Mining":
Companies capitalize significant expenditures year after year as Investments in Mining Property. Then every few years they take major write-offs to clear out the balance sheet. That effectively hides underperformance in bad years and then allows future years to ignore those costs.We submit that gold mining write-downs are more a result of marginal operations than expensive acquisitions. The earnings that get written-off would not have been earnings if costs were originally classified as expenses instead of capital items."
In other words, companies do not account properly for bad investment decisions on a quarterly basis and "adjust" at larger intervals, expecting shareholders to shrug their shoulders on these book entries as being historical and not important to the immediate and future cash flow of the company.
How To Resolve The Inadequacy of AISC Reporting
It is not that the shortcomings are not recognised by various parties and institutions. The Canadian Institute of Mining and Metallurgy ("CIM") has constituted a committee in 2013 "for the establishment of best practices for cost reporting."
Given that we are in 2016 and such recommendations have still to be published shows that the matter is not straightforward, probably because the CIM would want to come up with a perfect set of standards. We all know that perfection is the enemy of good.
So what do I suggest? Well, first, any system for reporting comprehensive cost needs to be as simple as practically possible. The suggestions by the World Gold Council in Table 1 are frankly too complex and open to interpretation.
Cash Operating Cost of Production
So why not use what is already available without exception? The Consolidated Statement of Cash Flow is part of every financial statement, is checked by auditors and reconciles to something that cannot be fiddled with: Cash and Cash Equivalents at the End of the Period. Apart from all mine expenses, including mine site G&A, realization cost for selling the product, it also includes Corporate G&A, income taxes paid and interest, items that are really cash outflows in the normal course of operation. It is the true cash cost of production.
A choice must be made about whether to use the amount before or after changes in working capital items. As these changes should cancel each other out in the long term, the choice is not terribly important. However, as escalation/inflation has an impact on having to invest in working capital items in the long run, I suggest that the number after working capital changes should be used as the true long-term metric.
As the cost is established from the consolidated statement, it does not capture the cost of each operation individually, but is that important to the investor anyway? Should management wish to account for the individual mine's cost structure for in-house control and reward purposes, it can allocate back the corporate, income taxes and finance cost to its operations.
Proposed Revision to All-in Sustaining Cost Definition
Here the suggestions of the World Gold Council are appropriate and all costs and investments required for maintaining production at an existing mine should be included such as delineation drilling/exploration for mine planning purposes, capitalized underground development and expenses for the replacement of equipment. It should differ however as follows:
- Corporate G&A has already been captured by the Cash Operating Cost of Production.
- All study costs up to start of feasibility studies should be included under AISC. Many studies on so-called growth projects come to naught and should be "expensed" in the AISC measure, irrespective on how it is dealt with in the financial statements for tax purposes.
The industry is already including the Sustaining Capex component and no major changes in policy are required.
Proposed Revision to All-in Cost Definition
The use of All-in Cost as proposed by the World Gold Council (i.e. AISC plus expenses not related to current operations) has two major shortcomings. Firstly, it gives the effect of immediate outlays and does not attempt to match benefit with cost (the matching principle in accounting) and, secondly, it also ignores the purpose of such investment in the first place: to earn a return on investment.
Some have suggested including a depreciation/amortization charge on project development costs, but that would require discretion on the method and period over which to apply such a charge. Furthermore, Amortization of pre-production expenditure would not be enough as a charge as it does not capture the hurdle rate for an acceptable return on investment.
I therefore suggest that a charge per unit of product is included, which I would call the Return on Investment ("ROI") charge. First inclusion would be after commercial production has been reached and can be based on the life of mine plan as per feasibility study and the actual investment until start of commercial production, including investments in working capital. The charge is calculated per unit of production - for gold mining companies ounce of Au - that will give the required rate of return that is the hurdle rate for this particular company.
- Upon start of commercial production, the amount of funding for the project is established, including the investments in net current assets until that date, assuming full equity financing at start of construction, irrespective of how the actual finance came about and should include the feasibility study costs incurred before construction starts. The reason for assuming full equity financing is that a project's return should be based on its own merits and not depending on financial engineering and not reward management for assuming financial risk. Improving the rate of return of a project is a decision separate from the merits of the project and should be kept out of it for the purposes of how well management has performed in giving the go-ahead for the project.
- The actual amount of funding that was required is divided by the ounces that are planned to be produced over the LOM. This gives the investment per ounce of gold to be produced.
- The required future value of each ounce of gold in each year is calculated by escalating this using the hurdle rate on a compounded basis. This gives the Required ROI Charge (US$/oz) for each year. With these measures it does not matter for the hurdle rate of return when a particular ounce of gold is produced, as long as it is produced at the Required ROI for that particular year.
- Based on the latest production plan, the required future value in each year can be calculated by multiplying planned production with required future value of ounces produced in that year.
- After the first year of commercial production, the actual value in excess of Sustaining Cost is established and divided by actual gold produced (Actual ROI Charge). I am afraid that here the principle of keeping it as simple as possible falls a bit apart, because should a company have elected for some sort of debt financing, the cost and repayment of such financing should be added to the excess over Sustaining Cost so as not to penalize the financing decision double.
- When Actual ROI/Required ROI >1 than the company is doing better than hurdle rate, when <1, worse.
- If the performance after the first year of commercial operation has differed from plan, or if there is a revision in LOM gold production, the Required ROI Charge need recalculation accounting for the remaining future value required and remaining ounces to be produced.
- This process may need repeating in every following year.
For companies with a number of operations, the reported figures will be a weighted average of the performance of the individual mines.
The above approach and possible requirement for annual revision may look to some complex and onerous, but is actually simpler than, for example, the calculation of the value of share options using the Black-Scholes equation, something that is routinely carried out by companies for options granted to management. When they can make the effort for themselves, surely they can make an effort for the owners of the company by providing information on how the owners are being rewarded (or not).
Based on the above, Table 7 summarizes the cost definitions in a similar format as Table 1 with World Gold Council guidance.
From the table, it is evident that the project construction outlays on growth projects are not accounted for as and when these outlays occur, but are only brought to book after start of commercial production. This means that in reality, the changes in cash as per Cash Flow statement can still be far apart from the notional cash margin established from these metrics. However, the advantage of the proposed All-in Cost is that it will impose a financial discipline on management in actually delivering on its promise of giving returns on historical investments as professed by them in their presentations.
Teranga Gold Corporation (OTC:TGCDF) ("Teranga") states in its MDA that they "evaluate all growth initiatives, including organic and inorganic opportunities, as well as new capital projects using an after-tax internal rate of return ("IRR") target to govern our capital allocation and investment decisions. For incremental mine site organic growth projects we set 20 percent as the minimum after tax IRR threshold." It sounds all fantastic, 20 percent after tax IRR, but, as can be seen from Table 6, its shareholders have not seen anything close to this. If management promises such lofty return it will directly impact the ROI charge in All-in Cost and shareholders can judge every quarter if management is meeting this commitment.
Inclusion of a ROI-charge will impose a degree of financial discipline on management that will deter them to embark on new business development adventures and waste precious cash flow.
What To Do Until a System of Reporting on Return on Historical Investment Is Formally Adopted By the Industry?
For an investor to make up its own mind about how the company is doing, I suggest to sum the following:
Cash from Operations After Changes in Non-Cash Working Capital/Ounces Produced in that period
Plus: the Difference Between AISC/oz and Cash Cost/oz as reported by company for that period
Plus: US$200/oz Au (for Ag: US$2.6/oz) as ROI Charge
The US$200/oz rule-of-thumb is based on a company that needs to invest 1.5 times annual revenue in pre-production capital expenditure and aims to have a compounded real return of 5%. For a company such as Teranga with its promise of 20% IRR after tax, you must add US$300/oz.
My experience with evaluating mining companies has been that an investment much higher than 1.5 times annual revenue makes it very difficult to arrive at a decent return as the cash operating profit needs to be then exceptionally high. That is why 1.5 x revenue multiple for pre-production capex is a good base to calculate required return per unit production on.
Whereas the above discussion has focused on gold/silver producers, it can be applied to iron ore, coal and base metal producers as well.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Editor's Note: This article covers one or more stocks trading at less than $1 per share and/or with less than a $100 million market cap. Please be aware of the risks associated with these stocks.