The Terrible Truth of Gold Mining Cost Reporting

Exploration Insights
by Brent Cook

Gold production costs as reported by mining companies have always consisted of a somewhat confusing, usually obscure, and often misleading set of metrics. It was my intention this week to make sense of cash costs, total cash costs, and the currently popular “all-in” costs, laying out what it all actually means.

To straighten out some inconsistencies I called Joe Hamilton, a good friend, fellow river rafter, and one of the best mining analysts and economic geologists I know. Joe is now a partner at Primary Capital, a high-quality boutique market dealer that provides M&A advice and finance to exploration and development companies.

After going through the complexities of how gold mining costs are reported a few times, it became obvious I was in over my head. Fortunately, Joe agreed to put his thoughts to paper in. . .

The Rant

The Terrible Truth of Gold Mining Cost Reporting
By Joe Hamilton

Mining companies generally point towards the “Cash Cost” of production to make profitability easy for investors to understand: The difference between cash-costs and commodity spot prices is the “margin” or profit that a company should post to its bank account. Investors are asked to use this as a gauge when comparing investment opportunities between two or more producing companies: lower cash costs obviously provide more cash for shareholders. Nothing could be easier and nothing could be further from the truth! But let’s dig a little deeper into the gold producers. . .

The Cost Numbers

In the last fifteen years we have seen companies report “Cash Costs”, “Total Cash Costs”, “Total Costs” and, now, a move by Goldcorp to report “All-in Cash Costs”. Confusing, to say the least since these ratios and numbers are not signed-off by auditors or accountants and are not even recognized by any Accounting Standards Board anywhere in the world. The companies can use these calculations in any way they wish and frequently the equations for the calculations vary by company. Confounding the entire issue is the concept of “Co-product” accounting or “By-product” accounting – the selection of which has the most fundamental impact on all of the above “Cost measures” but which many companies don’t even bother to specify in their financial statements.

First let’s look at some definitions:

Cash Cost:       the cost of production at the mine site, not including head office costs, interest expense, capitalized development or stripping, off-site costs (like smelting or refining costs), taxes or depreciation

Total Cash Costs:    Cash Costs plus off-site costs, head office costs, and sometimes interest

Total Costs:     Total Cash Costs plus depreciation, interest, and reported taxes (not necessarily paid)

All-in Cash Costs:    Cash Costs plus exploration expense, head office costs, and sustaining capital

Before understanding these ratios and numbers, we must look at where a mining company spends its cash and where it gets its cash. The first thing to understand is the difference between a capital item and an expense. For instance, fuel and explosive costs are an expense: the fuel is used this month.

But what about the cost of the storage tank?

It will be used for 10 years and therefore it is not a cash expense, but the cost is averaged over the life of the tank – this is a capital expense and shows up as depreciation over the life of the asset. Our assessment must include both capital and operating costs.

The largest component of cash expenses is usually the mine site costs directly related to production: salaries, fuel, water, grinding media, electricity, reagents, tires, grease and sometimes even sewage disposal. When we total up all of these expenses we get the “Cash Operating Cost”. The auditors look at this number with great scrutiny and it is almost impossible to misstate this number. But this is where the auditors leave our discussion. What about all those other things that consume the cash from a company balance sheet? What about cash consumed today so that I can mine tomorrow? What about stripping costs, underground development in advance of mining, shaft sinking, equipment replacement and new haul roads? Every company must use cash for these items; this is “Sustaining Capital”:  money that is spent to merely sustain a production level but is not included by the auditors as an expense. It’s possible to have two mines that produce at $700 per ounce cash cost, but one mine requires $300 per ounce in sustaining capital and the other requires $700 per ounce in sustaining capital. Unfortunately, the investor is only given the Cash Cost number and asked to differentiate between two “identical” gold mines. 

While we are at it, shouldn’t we also look at interest costs, debt repayments, head office costs, taxes, and royalties? These are items that are compiled and reported separately by auditors and don’t show up in the Cash Operating Costs. If this is overwhelming, don’t worry, its easily calculated because auditors are not dumb and they give us all the information we need to “follow the cash”. We just need to know where to look.

Before we go to the financial statements and apply some “Terrible Truth Equations” we need to understand the difference between “Co-product” and “By-Product” accounting.

These concepts were developed in base metal mines that always produced multiple metals: gold, copper, silver, zinc, lead, mercury, antimony to name a few. Auditors stopped using these terms decades ago. These concepts were never meant to be applied to gold mines and were known to be flawed, but they were resurrected by some marketing genius. Too bad for investors!

By-product Accounting

In any case, the mine gets paid different amounts for each product and must usually pay smelting or refining charges too. The difference between co-product and by-product accounting depends on the contribution to revenue for each metal produced. Generally, to use by-product accounting a single metal has to contribute at least 80% to revenue – the rest of the metals that contribute 20% to revenue are “by-products”. If a single metal cannot pass this test, then all the metals are thought to be “co-products”. This is important but frequently ignored by companies when reporting cash costs.

By-product accounting lets companies deduct the revenue that is received from the by-products from the costs associated with producing all metals at the mine. For instance, a mine may produce 1000 oz of gold and receive $1.5 million in revenue for gold. It may also produce 10,000 oz of silver and receive $250,000 in revenue. Because the silver revenue is less than 20% of total revenue the Company can use by-product accounting in reporting unit operating costs. Let’s say that the total mine-site operating costs were $1 million to produce the gold and silver. 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:  ($1 million - $250,000)/ 1000 would allow the company to report $750 per ounce cash costs of production. Hmmm – if I look at the cash costs I assume that the Company made $750 per ounce on 1000 oz of production and I could be safe in assuming that the company booked a profit of $750,000. But wait a minute – they received $1.5 million in revenue and had $1 million in costs which the auditors will tell you allowed them to book $500,000 in profit. Leprechauns must have buried another pot of gold! By-product accounting is false, but many companies still knowingly report these numbers – or their evil twin, “Cost per gold equivalent ounce”.

Co-product Accounting

This gets a little better with co-product accounting, which attempts to assign operating costs to each metal produced based on its relative contribution to revenue. Let’s look at a gold mine that again produces 1000 oz of gold but 40,000 oz of silver and has operating costs of $1 million. Revenue increases to $2.5 million ($1.5 million from gold and $1 million from silver) with a 60:40 split between gold and silver. If we were to use by-product accounting we would deduct the silver revenue from the operating costs and divide by the gold ounces: ($1 million - $1 million)/ 1000 – whoa – we are producing gold for nothing! Magic - we are probably seeing leprechauns by this time. Any company that reports “negative cash costs per ounce of production” is using by-product accounting. Under the co-product accounting, we need to look at the contribution to revenue for each product. In this case gold contributes $1.5 million to the $2.5 million in revenue with the remainder silver (60% gold: 40% silver). The operating costs are then split on the same basis and divided by the units of production: for gold ($1 million * 60%)/1000 = $600 per ounce AND for silver ($1 million * 40%)/40,000 = $10 per ounce. The Company should report that it produced 1000 oz of gold at cash costs of $600/oz and 40,000 of silver at cash costs of $10 per ounce. Now the back calculations work too, with the investor able to see the correct margins for each commodity. It’s a little better and a little more transparent. But it's a lot different than saying, "We produced 1000 oz at a cash cost of $0"! In addition, we still haven't considered all those uses of cash like head office costs and interest on our debt.

This brings us to the Financial Statements. 

The REAL Cost

Mines don't produce gold or copper or nickel - they produce cash! Really good mines produce piles of cash. Really old mines or bad mines produce cash but then have to reinvest it just to keep production steady, with no real cash going onto the balance sheet.

There are two truths in analyzing financial statements: the bank balance and outstanding debt have been checked by the auditors; and the cash flow statement never lies. 

Cash and debt are easy to check, but we can't forget "current debt": that portion that is due in 12 months. If it isn't obvious how a company will repay its current debt then I suggest that the savvy investor will wait for the upcoming equity issue to invest - and it will come! Long term debt is easier to understand - a company has at least two years to make the money from operations. But they need a good cash margin, which is not the one you think they have if you look only at the reported cash costs.  Fortunately we have a calculator, the always honest cash flow statement, and the Terrible Truth Equations. . .

The first thing to look at on the cash flow statement are the three subheadings: Cash from Operations, Cash from Financing Activities, and Cash from Investing Activities - there is no other place for cash to come or go except into one of these three bins. The auditors have signed off on these numbers and the company can't mess around here. Let’s look at Barrick’s Cash flow Statement from their Q1 2013 Financials (Fig. 1 below):

(Fig. 1: Barrick cash flow statement)

Cash from Operations

This is all the cash from all the mines. This number has been checked and re-checked by the auditors - it's real and it cannot be warped, misstated, recalculated or spun in any way. It's cash! It includes all expenses paid in cash in the period and includes taxes, interest, all office costs, and all disbursements that are not capital expenditures or debt repayments. The number next to "Net Cash provided by Operation Activities" is the true cash margin for the Company from selling its metals:  it’s the total of all cash brought in by the mines after all expenses (except capital costs and debt repayments as we shall see). If this number is negative then we have a real problem.

Deduct this number from the total revenue (found on the Consolidated Statements of Comprehensive Income = $3437 million for the quarter) and you have found the true, no-nonsense cash cost of production. In Barrick’s case, the true cash cost of production is $2352 million. This number can't be changed - it's real. What you do with it is up to you, but let's just look at the absolute number for now. You could calculate the co-product cost of production for each metal (copper, silver and gold) if you had 30 minutes to kill. For now, let’s just divide the cash cost of production by the number of ounces produced: (Total revenue - cash from operations)/ oz sold. This is Terrible Truth Equation #1: the true cash cost of production. Since the company sold 1.747 million ounces in Q1, its cash cost fully loaded to gold was $1,346 per ounce. This is a rather simple number and should really be adjusted for changes in working capital (payments to suppliers for materials that were not used immediately but are sitting at mine sites as inventory – like tires, engines, and reagents). It also includes all cash expenses charged by the company (mine cost, head office costs, interest cost, cash taxes paid, royalties, pension contributions, lunches, and toilet paper). It is something the companies don't want you to know and they don't do the math for you. Most gold producers style themselves as a single commodity Company. Let’s go with that and assume that they are mining only gold, and any other metals are incidental to producing gold - they get paid for the other metals, but there is no cost attributable and the gold production must bear all the operating cost. Seems fair to me if you call yourself a gold company instead of a gold-copper or zinc-copper-gold company.

Let’s look at some numbers

Agnico Eagle reported 2013 Q1 "Average total cash costs" of $740 per ounce of gold. But if we follow the cash and apply the Terrible Truth Equation we see that the company really produced an ounce of gold for $1207. For Barrick, the reported cash cost in Q1 of 2012 was $561 per oz, but the Terrible Truth is that it was really $1120 per ounce. Goldcorp gets kudos for reporting both by-product ($565 per oz gold) and co-product ($710/oz) costs for Q1 of this year. The Terrible Truth is that neither number is correct, and the true cash cost was about $1190/oz. Newmont does the best job of reporting by correctly using by-product accounting for copper production, but still shows a reported $1008 per oz for Q1 2013. The Terrible Truth is that the real number is $1308 per oz. And, these numbers are before any capital or debt repayment, but they do include exploration cost, those somewhat bloated head office costs, and Board expenses, interest, and taxes. This is the cost of doing business and it is a number which companies don't report-- but it is there in the financial statements.

The following table shows some basic numbers from Q1 financials. The last two columns show the cost of producing an ounce of gold if gold production had to pay all the costs for the company (100% gold burden) and the cost if we use co-product accounting (and assume that gold production has a pro-rata share of costs based on its contribution to revenue). 

Q1 2013

Total Revenue

Gold Revenue

Cashflow from Operations

Working Capital Adjustments

Gold Production

Realized Gold price

100% Gold burden Cost per ounce

Co-product Gold Cost per ounce














































We admit that this is equally confusing and does not really give companies credit for multi-metal mines that generate cash – and this is, after all, why we mine.

Cash from Investing Activities

This box in the cash flow statement records capital expenditures to build and sustain operations. It's still cash out the door, but it is sometimes hard to differentiate between new mine development costs and the capital required at existing operations. But this leads to Terrible Truth Equation #2: the Burn Rate calculation.

Let’s compare two numbers: cash from operations and cash from investing. The first number is the cash generated from operations after all expenses have been paid. The second number is the cash consumed in construction activities at both existing and new mines. In Agnico’s case, the company made $146.1 million in Q1 of 2013 from mines and spent $141.5 million in construction. Hmmm - what happened to the $740 per oz and the implied $871 profit? There was almost no cash generated! Goldcorp in Q1 2012 brought in $294 million from operations and spent $533 on construction projects. Fortunately, Goldcorp added $1.5 billion in debt to its balance sheet in Q1 and ended the quarter with $1.46 billion in cash, otherwise they would be broke. For Barrick, in Q1 of this year operations brought in $1.1 billion in cash, but they spent $1.4 billion on construction projects. That's for the first quarter! Barrick took on $2 billion in new debt in Q1 but repaid $1.2 billion, and paid $200 million in dividends. Give Barrick a nod for borrowing money to pay its dividend! Newmont made $433 million from operations in Q1 but spent $507 million in new capital and also paid $211 in dividends. The company borrowed $80 million in new debt to cover the capital expenditures and reduced its cash position to pay the dividend.

To sum this whole discussion up - "Don't believe everything you are told". The mining companies continue to spin investors by quoting cash costs but only including partial costs in the calculation. To give them credit, the tables in the back of 10Q and quarterly MDA filings have started to show more information, but the tables are becoming impenetrable in the process. The cash cost numbers seem to be chronically low and some would say possibly misleading. Thankfully the auditors give you the numbers, and by digging a little deeper the Terrible Truth can be laid bare.  Remember that Colonel William Boyce Thompson, the founder of Newmont, was asked "Where did you get the name for your Company?".  His reply: "Well son, I've got two mines, one in Montana and one in New York." (I made that up; I don't think Newmont ever had a mine in Montana! But it summarizes the story nicely.)