Katchum's  Instablog

Send Message
Albert Sung is the author of the Katchum Macro-Economic Blog, monitoring breaking economic news from a day to day basis. He started investing in 2008 because of the economic crisis and holds a masters degree in chemical engineering. Previously, he worked several years as a process engineer at... More
My blog:
Katchum's Macro-economic blog
  • The Marginal Cost Of Gold/Silver Production (2012) 17 comments
    Dec 23, 2012 11:46 AM | about stocks: GLD, SLV

    UBS recently put out a number for the cost of producing an extra ounce of gold. As you can see on chart 1, the cost has skyrocketed from 2008 onwards to today. Costs almost doubled in 2 years time.

    It shows us that if the gold price were to go to $1500/ounce, nobody would go out and search for gold as it would be unprofitable.

    Chart 1: All-in Cost of Gold Mining

    Chart 2 gives an operating cost of $700/ounce for gold, but it's important to notice that the large bulk of the costs go to construction, maintenance, exploration and taxes. Also note that the lowest gold went in 2008 is exactly at $712/ounce in October 2008, which was 10% below marginal cost of production at that time. That low in today's terms would be $1350/ounce.

    Chart 2: Replacement cost for an ounce of gold

    With all of this in mind, I believe gold will never go below $1350/ounce. This will be the ultimate floor. If it does, it will quickly rebound.

    And for people who are interested in the marginal cost of production in silver, it is around $30/ounce as suggested in this article. So I expect silver to rebound soon.

    Disclosure: I am long AGQ.

    Stocks: GLD, SLV
Back To Katchum's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (8)
Track new comments
  • The_Hammer
    , contributor
    Comments (5141) | Send Message
    Katchum absolutely wonderful work and NO comments! hello!
    although i do think it is difficult to target a marginal cost since mines have different characterisitics and localities. The main thing to understand is that cash mining cost are NOT total all in cost like acquisition price, development, exploration, construction. maintenance and taxes. It makes sense to calculate what your total cost are until sale of end product to see if you make a profit.
    26 Dec 2012, 02:14 PM Reply Like
  • Swayze8
    , contributor
    Comments (2) | Send Message
    I agree it is important to look at all-in costs when evaluating the individual merits of a company, but the marginal cost of production for the industry has no bearing on the direction of the gold price. Annual mine production only fulfills a fraction of annual gold demand. It's so small that global production could be doubled or cut in half and it wouldn't have a significant impact on price.
    This is one of the many reasons that currencies were once backed by gold - as it controls the rate at which governments could print money.
    You should view gold as more of a currency, and less as a commodity. This means looking at in terms of other currencies - not just the US dollar.
    6 Jan 2013, 02:49 PM Reply Like
  • Katchum
    , contributor
    Comments (615) | Send Message
    Author’s reply » That's right, I forgot about that...


    Mine production is only 1% of all gold in the world so it doesn't have a lot of effect. But the other way round it does have an effect. Gold companies depend on the gold price itself.
    6 Jan 2013, 05:05 PM Reply Like
  • The_Hammer
    , contributor
    Comments (5141) | Send Message
    marginal production costs sure do impact the gold price. gold supply is expanding at about 1.5% a year. Even with the elevated prices supply is not pouring onto the market.
    If the price of gold falls below cost of marg production "New" supply will dry up. Cos cannot mine at a loss for long. In a sense it will put in a long term support price eventhough it can undershoot to the downside or overshoot to upside.
    Commodities are in relative abundance but at a cost. The cost for gold, oil and other commodities getting more expensive to extract. The low lying stuff has been taken.
    7 Jan 2013, 12:48 PM Reply Like
  • Katchum
    , contributor
    Comments (615) | Send Message
    Author’s reply » Let's say all mines in the world shut down. Then we have gold that is going to be bought by the Chinese, the investors, Russia, Brazil, Iran, Turkey, Japan, etc...


    Who will supply this gold to them as nobody is going to sell their gold and just hold it? We now know that the Western Central Banks aren't selling their gold because the trend in gold holdings at Western Central Banks is flattening out.


    So the price needs to increase if mine supply drops, because otherwise the Eastern Kingdom can't buy any gold at any price.


    But this is all talk, let's get to the real quantitative numbers and see if Swayze8 is correct:


    Total gold supply is currently: 4300 tonnes/annum.


    Gold Mine supply is around 2800 tonnes/annum. http://bit.ly/Xf4L66


    So gold mines supply 65% of all gold demand. So how on earth does gold mine production not influence the gold price? If you shut down all gold mines, the gold supply would drop by more than 50%, while gold demand would stay the same. So the price of gold should skyrocket.


    So I do think that marginal cost has effect on gold price after all.
    7 Jan 2013, 01:47 PM Reply Like
  • Swayze8
    , contributor
    Comments (2) | Send Message
    Perhaps I didn't phrase that correctly.
    What i meant to say is mine supply adds a small amount each year to the total supply of gold. Most of the gold traded in the world is not the incremental new supply, it's the shifting among holders of the existing base supply.
    So annual gold production of roughly 2800 tonnes is of little significance to price because true supply consists of all the gold that has ever been produced (roughly 170,000 tonnes).
    Much like the increase in money supply dilutes purchasing power of existing money, an increase in the supply of gold shoud be understood as a dilution of the existing supply. So 2800 tonnes per year or a 1.6% increase in supply can be absorbed by the market via 1.6% price drop.
    If annual mine production were cut in half, this would translate in to an annual increase of ~.75% in the supply of gold. On the other hand, if a significant part of oil production were to be disrupted for an extended period of time, stocks would be exhausted after only a few weeks. This means it is much easier for gold to absorb any form of significant production expansions or shortages.
    Therefore, gold is precious because the annual production is so low relative to the total stock. This stability and safety is a crucial prerequisite for the creation of trust, and it is what clearly differentiates gold and silver as monetary metals rather than commodities.
    8 Jan 2013, 11:00 PM Reply Like
  • Katchum
    , contributor
    Comments (615) | Send Message
    Author’s reply » So you say: if we increase the supply by 2800 tonnes/annum, the price of gold would only drop 1.6%? I'm not sure, I need to investigate that too.
    9 Jan 2013, 11:33 AM Reply Like
  • Aricool
    , contributor
    Comments (6611) | Send Message
    Swayze, I like your thinking, but your thinking of Gold as if it were paper money just does not sit well with me. I don't think that tracks with how humans set an intrinsic value to anything. I mean, humans don't count how many things there is of something and prorate its value accordingly. The fundamental value assigned to anything is the cost (i.e., energy) to create it. Paper money has no cost of creation, thus it is just a counting exercise, and paper money is far from rare, so no premium value put on it. Gold is desired (and thus valuable) because it is very rare, which is another way of saying its cost of creation is very high. If the cost of creating gold, say, dropped to $10/oz and stayed there. Rest assured that the price of gold would rapidly drop to the cost of production soon enough. This dynamic does not exist with paper money, thus your equivalence of the two breaks down for at least that reason.
    30 Apr 2013, 12:25 AM Reply Like
  • rustic06
    , contributor
    Comments (53) | Send Message
    Swayze8, you are right. Bitcoin tried to recapitulate the gold market value model by creating a fixed amount of cyberwealth to which only a small amount of new wealth of known quantity could be incrementally added. Gold is not bitcoin (bitcoin could be hacked or otherwise contaminated) the gold model has been safe for over 3000 years which makes it precious. It is not a commodity, it is something of great value, not money in the strict definition, but closer to real money than fiat.


    You are right that gold should be valued as a rare painting or other highly valued art object rather than oil which is used up almost immediately after being refined. IMHO gold demand at this time is more important in the pricing of gold than production. As loss of gold production begins to be noted in the media and demand picks up, then the price will rise steeply. Gold is a relatively small market and its price is likely being manipulated at this time.


    I am buying mining stocks and CEF.
    29 Jun 2013, 11:44 AM Reply Like
  • udoran
    , contributor
    Comments (5) | Send Message
    Ketchum & Gentlemen,
    The definitive study of the business is as always from GFMS, and has some most interesting facts. Have a look at the graphs breaking out the major country costs vs world. Notice how the S. Africa costs are the top of the heap, as it is obvious that with the labor problems there coupled with declining grades TWO MILES or more down are a killer.
    Also to be considered are the industry grasping for ounces at almost any cost per the ABX debacle at Pascua Lama, and the mega costs at Pueblo Viejho. Some of the CIBC numbers above should be dissected for complex underground high costs operations, vs Gold Copper deposits with heavy smelter charges at 25 - 30% or more of gross off the top charges. Some of the Society of Economic Geologists and older GFMS excellent work on finding costs are worth having also. Here is the fresh study; http://bit.ly/14UjQxv
    6 Jul 2013, 01:41 PM Reply Like
  • aggie1961
    , contributor
    Comments (4) | Send Message
    What you say about gold may or may not be true because of the small amount of gold mined yearly compared with the total stock of gold. However, the argument is true for silver because so much silver is consumed and the amount of silver inventory available for sale is relatively small compared with annual mine production. Two factors mitigate against substantial immediate reduction of mined silver: (1) most mined silver is a by product of other mining, usually copper and gold and (2) miners who produce mainly silver have some flexibility to reduce costs so mining still cash flows so there may be no substantial immediate reduction of silver production. However, the cure for low prices remains low prices.
    8 Jul 2013, 12:45 AM Reply Like
  • hmorgan
    , contributor
    Comments (30) | Send Message
    Nicely written article. I think in general, the author makes a very common and potentially fatal mistake. Those who are predisposed towards precious metals tend to focus on the all in costs of production to opine that Gold should trade at $1500 (or more) and Silver at $22 (or more). Their analyses of costs is well thought out but they make a mistake in assuming that market prices gravitate or tend towards all in costs or average costs. This is flawed thinking. Market prices historically gravitate towards marginal production costs. This is the reason why pure commodity business (like airlines) are lousy businesses to own for the long term.
    Whats more, Gold operations have rising marginal production costs as less efficient mines are brought online with rising prices. (And vice versa). As Gold drops and less efficient mines are closed, the marginal costs will decline.
    The more important (from an investor standpoint) marginal production cost of Gold is likely $650-$700. With this in mind we believe Gold's tendency is towards those numbers. Finally, we think Gold would not provide attractive risk/reward until its price drops below (well below) marginal production costs.
    21 Apr 2014, 02:20 PM Reply Like
  • Aricool
    , contributor
    Comments (6611) | Send Message
    re "Market prices historically gravitate towards marginal production costs. " this is likely not true. Gravitation happens over time, and over time it is 'all in sustaining" costs which the banks will determine to loan you new cap-ex $ or not. cash-costs only matter in the very short term, esp. if the miner has most of their (1 and 2 year) forward volumes hedged near all-in costs. For recent examples, just look at bulk shipping day rates, and Nat Gas prices... Gold/Ag should be no different...
    24 Apr 2014, 12:06 AM Reply Like
  • The_Hammer
    , contributor
    Comments (5141) | Send Message
    24 Apr 2014, 07:31 AM Reply Like
  • Aricool
    , contributor
    Comments (6611) | Send Message
    hammer, nice article. thanks. that actually makes the argument to short the silver miners, not buy them. This is one of the reasons I dumped all my silver positions (except PAAS) a few months ago. If you plot nearly all of the top silver miners (even proxy like SLW) you'll see that they are way over valued compared to SLV at the same price point last year. So, if SLV stays down another month or two, all the miners have to come down allot. which is one reason to wait and see.


    BTW, all the cost probs listed in that article are being experienced for nearly all other commodities worldwide.


    the prob. w/ Ag, though, is that all the issue the article list will first put the hurt (kill?) on the pure Ag miners b/c most Ag is byproduct which has none of those cost issues, and there is not even enough Ag demand to consume that zero cost byproduct supply.


    So, what is the logical argument to buy Ag miners now? Seems like they'll get way cheaper soon enough...
    24 Apr 2014, 08:51 PM Reply Like
  • The_Hammer
    , contributor
    Comments (5141) | Send Message
    i think the argument is to buy physical silver (especially silver eagles - find a dealer that has least premium) when it swings below marginal production costs which continues to increase due to lower ore grades and other production costs.
    Miners have been proven to be run by cash incinerators diluting shareholders anytime and exposed to risky jurisdictions that change the rules after the huge up front cap ex investments. Miners are a leveraged trading vehicle.
    you said ag byproduct production not covered by ag demand to consume? what is the breakdown ?
    25 Apr 2014, 07:26 AM Reply Like
  • Aricool
    , contributor
    Comments (6611) | Send Message
    "you said ag byproduct production not covered by ag demand to consume? what is the breakdown ?"


    Top level, by-product production is 72% of total, or ~567Moz- see:


    Ag industrial demand (where all the hope for demand growth comes from) is ~466Moz (and has been flat YoY since 2007)- see:


    So, by-product production growth will continue to grow, but industrial demand will stay flat, to down.


    Total mining production is growing steady, and over 787 Moz.
    But total supply (scrap + mining) is over 1100 Moz.
    Moreover, Total silver fabrication demand is down to ~847 Moz.


    So, Ag is obviously way over supplied by at least 200Moz (ignoring fickle implied investment demand), which is equal to all the pure Ag mining.


    So, if all the pure Ag miners stopped production tomorrow, the market would be in balance, and then Ag prices could increase.


    Thus, what is the argument to buy Ag miners if they are the ones required to curtail production, and thus "fall on their sword" and die?


    Moreover, Ag prices will have to stay low for a long time to force them to curtail production, and they'd have to permanently curtail it even if prices rise or the problem restarts.


    The only big growing demand is in solar panels, but they cannot tolerate Ag over $25/oz before switching to alternatives. Thus, seems like a catch 22 for Ag, and dead money for a long time, unless maybe if the world stopped using digital cameras and when back to film.... LOL!


    So, I laid out why it seems Ag is in a structural bear market.


    Can anyone lay out an equally supply/demand reasoned bull case which is not related to manipulation, dooms day or boogie boogie stuff?
    25 Apr 2014, 09:38 PM Reply Like
Full index of posts »
Latest Followers


More »

Latest Comments

Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.