Friday, May 31, 2013

A COMEX Challenge

I have an idea that I will probably act upon: buying a 100 oz gold bar via COMEX!  At this point, I am starting from "zero" so this will take some thought and effort to pull off.

I will only do this if I get positive feedback, and a lot of it (that means you)!!!  Those of you who support the whole idea of demanding delivery of 100 oz of gold vs. cash settlement, please comment here or email me!  This will cost me money, so unless I get a lot of "moral support" I will not bother.  I do not ask anyone for any money, but without a lot of encouragement, I will assume there is little or no interest out there among you for me to do this.

***

Broadly, here is my rough outline as of the moment, this of course is subject to change as I learn more:

1)  For me to find a good commodity broker who will guarantee that I will get my bar in the month I have it for delivery, guarantee in writing, with not only the broker but his boss, the branch manager, all the way up as far as I can get something signed.  No cash accepted.  I want the gold.  I will put up the 100% "margin" when it becomes necessary (discussion on that just below).

2)  I ask you, kind readers, for your recommendations as to a good commodity broker who will arrange for delivery of my gold.  I can pay him "extra", above and beyond his normal commission, so if any of you have a friend who is one, well, do him a favor...

3)  As I educate myself on this process, I will soon pick a month for delivery.  It will NOT be June as I need time to study this matter in detail.

4)  I ask/beg anyone who has accepted physical delivery of COMEX gold to drop me a line with any observations that would ease my task.

5)  I also invite anyone with the means and interest to join me in a parallel (but separate and independent) operation.  As long as it is not to quickly, I can adjust my delivery dates to match yours.

6)  I will (for sure) have the capability to assay any gold I receive.

7)  If I go ahead with this, there will be a complete report or set of reports on this process here at my blog, as I suspect it would be of interest to my readers and to the "retail gold community" as a whole.  Yes, there will be a picture or two...

8)  I am going to guess that I may need a lawyer, someone schooled in commodity law and in particular who would know COMEX procedures and regulations.  With some effort, I could find such a lawyer, but a recommendation would be much better.

***

If I can get the proper paperwork in order, I would expect to receive the gold.

If the paperwork is in order (100% in order), and I do not receive the gold but a check instead, then that will mean war!

This could be lengthy and perhaps expensive.  I want to have everything more-or-less ready to go before I contract with a broker.  This means work for me.


Update -- 1 June 2013

Alert Zero Hedge member "TPTB_r_TBTF" provided me a link to the CME's (COMEX) "rulebook", it is apparently already in their rules that if they so decide that they can supply money (FRNs) instead of any commodity.

So, that ends this escapade.  The COMEX need have nothing to fear from me now.

Tuesday, May 28, 2013

My Glimpse Into The Hereafter

Today's article is inspired by two other bloggers: FOFOA and "TwoShortPlanks".  Here are the links:

fofoa.blogspot.com

twoshortplanksunplugged.blogspot.com

(the title of my post tonight is inspired by FOFOA's recent post)

Both bloggers predict very large quantum leaps in the price of gold (the "reset"), FOFOA predicts on the order of $55,000 - $100,000 per physical ounce and TwoShortPlanks ("TSP") calls for $134,000 (these numbers are of course approximate).  These numbers, as far as I can determine, are non-hyperinflated...  If we have hyperinflation, those prices would be that much higher.

Disclosure: I definitely have a dog in this hunt, gold rising to $55,000 or more per oz would certainly make me smile...

Of course, no one really knows, and it would depend on a large number of hard to quantify variables.  Jim Sinclair (jsminset.com) seems to be coming around to the Freegold idea as well, writing that $50,000 could be perfectly reasonable.  It does appear that the very high numbers predicted by the above observers are becoming more frequent, which is an interesting notion just by itself...

It is NOT crystal-clear to me how we would arrive at those very high numbers, I recently did explore some scenarios that would be plausible (these are convincing enough to me to be plausible, and this does not include close followers of gold who know more than I do...):

http://robertmixblog.blogspot.com/2013/05/gold-arithmetic.html

If these numbers of $55,000 are even close to being correct (and I will accept as "correct" any price of gold
over $10,000 per oz), then I believe there would be a number of changes that are as yet unanticipated...

***

Fallout

Macro Fallout...

My best guess, apparently paralleling FOFOA, is that gold would take its place as the best Store of Value after the "reset".  There are a number of reasons that FOFOA goes into, at length (!), but at such high numbers as over, say, $10,000 it is clear that spending gold might be just as clumsy as it would be today, perhaps more so.  Imagine taking a 1 oz Gold Eagle to the grocery store...

Perhaps the biggest question would be the nature of society after such a reset.  It is easy to imagine a "Mad Max" (TEOTWAWKI) set of scenarios, but that of course is not certain.  FOFOA himself (in a private communication with me) said that a disastrous scenario is not necessary for Freegold to come about, but he is on record as saying that he thinks we will have a hyperinflation of some sort.  As far I am concerned, a functioning society free of massive-scale unrest and criminality is really the most important result that I would want to see happen.

Easy to imagine in a post-reset would be the risk of theft/robbery of physical gold.  If the amount of gold in a man's wedding ring (say, 1/20 of an oz) were to be worth some $2500 in today's money, I would certainly wonder about if almost ANY gold would be worn as jewelry.  I would also imagine that jewelry store robberies as well as burglaries would rise, perhaps significantly.  Because gold is easy to make into other forms (low melting point) -- hence making it liquid for thieves, I think this is an under-appreciated risk of holding gold after the reset.  Security of guarding one's gold pile would become a much bigger issue IMO.  Diversification (long-time readers of my blog know this is perhaps my favorite word...) of hiding places becomes more of a factor than it already is...

OK, how about starting a business that would store gold?  One of FOFOA's followers asked that very question at his blog (comments section at http://fofoa.blogspot.com/2013/05/hold-on-to-those-gold-coins-and-bars.html).  FOFOA's answer (short version, my words): the banks or Brink's (companies that already exist) would likely offer that service.

At $50,000 per oz, what about fake gold?  Oh yes, that could be a big problem!  Reports are out there about Silver Eagles already being counterfeited (by China, who else?), here is a link to some information on precious metal fakery now: http://www.tfmetalsreport.com/comment/316667#comment-316667.  At that comment is a video of Mike Maloney and James Anderson (goldsilver.com) showing and testing a fake Silver Eagle alongside a real one as well as comments on fake gold as well.  Right now, it is not really economic for counterfeiters (even in China) to be making counterfeit Gold Eagles (particularly in small sizes, say 1/4 oz) out of tungsten (the metal closest to gold in density) because tungsten is very hard to work with, it is very hard and has a  very high melting point. There have been two notable cases of fake gold bars are a 500 gram bar and a 1 kg bar.  But, at $50,000 per ounce, those economics change a lot, we should expect to see a huge amount of counterfeit gold in the market place...

OK, how about testing equipment for gold?  Yes, testing and initial-stage reprocessing of gold (eg, melting scrap gold) becomes a more viable way to make a living (and probably big business as well).  An "ultrasound thickness gauge" is a device that can measure sound velocity through metals (the sound velocity is much slower through gold than through tungsten).  I am going to get one of these sometime...  Amazon.com sells cheap-looking ones for some $350 or so, one that *looks* better to me is the "GemOro AuRACLE Electronic Tester" ($459 at roseco.com, 800 527 4490).  Another angle on this would be the business (assayonwheels.com) I saw advertised in Barron's one weekend: http://robertmixblog.blogspot.com/2011_11_01_archive.html.  This might be a fantastic business to be in in a post-reset world...

Um, those 20,000 tonnes + in India are going to make that country RICH!  Well, yes.  Luck happens...  But, the gold is widely distributed among the people (gold is an "India Thing"), so not all that many new billionaires would arise from India.

There will be LOTS of undeserving people who get rich...  Well, again, yes.  Those with the foresight to have bought gold will do very well.

Silver Fallout...

Silver is a hotly debated metal, and here I would like to mention its possible role in a future gold price reset.  FOFOA believes that silver will not make the huge quantum leap in price that gold will.  I agree, though I am not certain (smile).

It may come down to what kind of world we find ourselves in re silver.  If things get very bad (TEOTWAWKI), then silver would become very useful for spending.  Bullets and cigarettes too...

My skepticism about silver mostly rests on this: if silver were so rare and industrial users of silver so vulnerable to a physical shortage of silver (however brought about) then why are industrial users (like Apple) not stockpiling it?

Taxation / Confiscation after a Reset...

FOFOA mostly assuages these concerns as far as I am concerned.  Much of his argument here that gold would NOT be confiscated or taxed very highly rests upon two ideas:

a) that gold "must flow", there are sound reasons he goes into that I will not here

b) gold would become the major source of capital for investment

He allows that the government might be irrational in the short-term (they may TRY to confiscate or tax it), but that would ultimately be shown to be unwise...  (I would argue that the government might be more irrational than he apparently does, and for longer, but I still agree in the end)

***

Who wins?  And who loses...?

I mentioned above that India and all other holders of physical gold would be winners.  Who are these winners?

In India, gold is widely held, and yes, there will be many winners who own, say, 1 - 3 ounces.  Their lives would change for the better.  And in many cases, a lot better.

Here in the United States (and perhaps Europe and Japan might be similar, at least similar to the USA in scale even if holding gold is somewhat more popular in Europe), it is thought that only about 1% - 3% of people own any non-jewelry gold, and that 1% number I have seen more often and better-argued as well.  TwoShortPlanks wrote the below in his latest piece (http://twoshortplanksunplugged.blogspot.com/2013/05/gold-going-goinggone-in-this-post-i.html):

6.       Less than 1 in 1,000-5,000 people have any meaningful amount of Gold.

I do not know where he gets those numbers, I know two people (besides myself) who have "meaningful amounts of gold", and I know less than 1000 people (but maybe my friends & family are smarter than I thought...).  Nor is "meaningful amount" defined.  Nonetheless his point is correct, that only a small number of people would benefit meaningfully from a reset.

I would like to take a crack at that idea of a "meaningful amount" of gold, and I will write with an American perspective in mind, please allow me to engage in another "Thought Experiment" as I am so wont to do...  Americans (Europeans, Australians, Japanese and so on count here) are perhaps "ten times" richer than the average person here on Earth, and so should perhaps own ten times as much gold as "average" (the average being just under one oz of gold per capita worldwide).  OK, then each American (man, woman and child) should own 10 ounces, a family of three should have thirty ounces...  If we say that a family of three actually DOES own 30 oz, well, that would without doubt count as "meaningful" to me...  Further, if "TSP" is right in what will happen to gold's price, then such a family would then have approximately $4,000,000 worth of gold (using his price of about $130,000), a game-changer for most families...

Who are the losers?  (And I do not discuss those who would lose in the almost inevitable decline in economic activity in such a transition -- which might be pretty rough and cause losses...)

If we wind up in a hyperinflation, then anyone without gold would lose!  Those holding other hard assets would do, probably, "OK", but would not see the huge increase in wealth that gold owners would.  In hyperinflation, hard assets will still be worth something, maybe even a little more than just "something".

If the reset does NOT happen in a hyperinflationary world (as FOFOA grants as possible, although he thinks hyperinflation is more likely), losers would include almost anyone who sells assets to those with gold...  I am thinking, perhaps, of my county that might need capital and have to raise it by selling assets (a toll bridge for example) or possibly buying the two gas stations in my town on the cheap...  Imagine!  If by trading a (relatively) small amount of gold by today's standards that you wound up with a toll bridge or the only two gas stations in town...  You would win, but someone else loses, at least relatively.

***

Finally I have not read anyone else's (even from the two bloggers) detailed explanation how we get to the seemingly astronomical numbers mentioned above...  When I bumped into FOFOA's blog in 2009, what hooked me (of course!) was that radical prediction of $55,000 per ounce!  What owner of physical gold could NOT be at least somewhat entranced by that?  Easy money!

It took some time for me to understand the basics of Freegold, but I essentially believe that something like that will occur (on the other hand, I am ready and diversified in case it Freegold does not come about).

In both cases, I would like to see more of the calculations of each blogger in reaching their very high numbers.  Hey!  I took my shot recently at deriving high gold prices, I would like to see theirs, or anyone else's calculations who is more sure than I am about a quantum reset.

Tuesday, May 21, 2013

Ameru Bearing Sale to Costa Rica

I am happy to announce that Ameru Trading del Peru S.A. has completed its first export sale!  Our company is the exclusive distributor of KBC bearings (Korea, owned by FAG/INA/Schaeffler of Germany) for Peru.  The KBC distributor in Costa Rica is a company called "MIGSA", run by a gentleman named Guillermo Aymerich, who my wife and I had the pleasure to meet several years ago in San Jose.  He treated us to dinner, so I owed him...

MIGSA needed certain special KBC bearings that are hard to get.  Hard to get meaning in this case that KBC only rarely makes these specialized items, he needed them and then contacted us.  We had some of the list he emailed us.  Ameru then worked with MIGSA and the Peruvian authorities to see the best way to ship the pieces to Costa Rica would be (lowest cost being the major factor).

The sale amount is $1377.65 (US dollars), the bearings are all from KBC (some are "FAG" / Korea brand, but from the same factories).  All pieces are made by very few manufacturers is the world.  The last piece (KBC DT-458448) is a wheel bearing for the Hyundai H-1 "Grand Starex" that I have mentioned before in other blog articles.  All pieces are for Korean vehicles.

Hyundai vehicles are very popular in Costa Rica as well as Peru -- in both countries Hyundai is even stronger than in the USA.

The Invoice does not come through very well, but here it is, *click* on the image for a better view (at the bottom of this article is a larger copy):


I would like to thank my sister-in-law Lily and her husband (and General Manager) César for their hard work in figuring out how to do an export order on short notice for the first time, muchas gracias!

Gracias as well to Mercy & Roberto, who helped advise us and complete the order as well!

And many thanks to my friend (and now customer of Ameru) Guillermo Aymerich, nos vemos en Costa Rica en enero!



Sunday, May 19, 2013

Gold Arithmetic

There is a lot of commentary about where the price of gold is going and even what the "correct" price of gold should be now.  I certainly do not have many answers to this whole subject, but I wanted to provide some perspective on gold's place in wealth holdings as well as its place in the financial system.

My own personal view is that FOFOA's numbers are likely to be seen when all of this (the distortions and contortions in our financial system) are addressed, whether deliberately (unlikely) or the events precipitating the gold reset (transition to Freegold) bring about much higher gold prices.  The numbers most kicked around at FOFOA's blog and comments are a range roughly between $55,000 to $100,000 per oz.

As I write this, however, we are currently in a bear market for the price of gold, now at about $1360 per oz off about 25% off its all time high.

I contend that, for the moment, the price of "paper gold" and physical gold are fairly close, with a premium of physical to paper of between 3% and 20% (depending on the form of gold and the size of a purchase).

The most dramatic effect of Freegold will likely be the divergence of paper gold to physical gold, FOFOA has explained that the paper price of gold will eventually be zero as investors flee paper and want the real thing.  This article explores some of the themes and issues associated with revaluation of gold.

Please note that all of this arithmetic is approximate, and is really only exploratory in nature.

***

One way of looking at a gold price reset, at least from a US-centric viewpoint, is to look at our national debt vs. the value of our gold reserves.  Most observers (myself included) believe that the USA indeed does have the 8130 (or so) tonnes of gold that the Treasury claims (most at Ft. Knox, lesser amounts in Denver, San Francisco and at the Federal Reserve Bank of NY).  Our arithmetic lesson starts here:

8130 tonnes = 8100 * (1000 kg / tonne) * (32.15 toz / kg) =  262,000,000 toz

262 million ounces * ($1360 / toz) = $356,000,000,000 ($356 billion)

[$356 billion dollars is not all that much money...  That would be a little over $1000 worth of gold for every man, woman and child in the USA (315 million) or less than ONE oz per capita.]

Compare to our current national debt of (approx.) $16.75 trillion.  That $356 billion worth of gold works out to being only about 2.1% of our national debt!

An idea floated by FOFOA is that the Treasury could just unilaterally revalue the price of our nation's gold to an arbitrary high price, he suggested that it could be in the $5000 - $10,000 range.  The Treasury "would put out the call to buy any and all gold for $10,000 / oz (say)".  The idea here would be to allow our gold to better offset our national debt by revaluing the gold (by essentially devaluing the dollar).  Well let's look at that a moment:

($10,000 / $1360) * $356 billion = $2.62 trillion, value of our nation's gold at $10,000.

OK, $2.62 trillion is a "better" number, but would we trade our nation's gold just to eliminate less than ONE FIFTH of our debt?  No way.  FOFOA then went on to write that this arbitrary revaluation would likely just be a stepping-stone to even higher numbers.  What if we wanted pay off our entire debt using our gold?  What would the price have to be?

$16.75 trillion / 262,000,000 oz = $64,000 per oz

And this (again) assumes that we would use all of our gold just to pay down the debt, which of course will not happen.  That $64,000 per oz falls squarely within FOFOA's target price range.

***

If we look at how other countries would fit into all of this, here is some more information.  Total gold held by central banks is estimated (2010) to be some 30,500 tonnes, but many consider that figure to be too low for today, the information is old, and it is thought that China and perhaps Russia & Saudi Arabia hold more than they claim.  So, "let's say" that the central banks of the major countries hold 45,000 metric tonnes (China claims only a little over 1000 tonnes held, but most informed observers think that number is really between 4000 and 8000 tonnes).

A quick scan over at wikipedia shows that most developed countries of size as well as a few developing countries (overall: "countries that matter") will have varying but mostly similar to the USA's debts in comparison to their GDPs (the figure for the USA is about 90% now, debt / GDP).

http://en.wikipedia.org/wiki/List_of_countries_by_public_debt

Let's "average this down" a little, as many (though smaller) countries have a significantly lower percentage of debt to GDP.  Call it 80%.  Total world GDP is around $83 trillion (wikipedia: http://en.wikipedia.org/wiki/Gross_world_product).  Arithmetic...:

80% (0.8) * $83 trillion = approximately $66 trillion of total national debts

45,000 tonnes gold = 1.45 billion toz * $1360 / oz = approximately $1.97 trillion worth of gold at today's price.

$1.97 trillion / $83 trillion of world national debts = approx. 2.3% value of gold vs debts.  There is not much difference between the world vs. the USA when it comes to gold holding as a percentage of national debts (part of this can be explained by the fact that the USA is almost 20% of world GDP and the fact that almost all of the major developed countries are all roughly in the same boat we are...).

So, by upwardly revaluing the world's gold to about $60,000 / oz would "cover" all national debts (gross) worldwide.

"FOFOA's numbers" still look about right...

***

A different way to look at "future price of physical gold" is to ponder the "quantity" of paper gold out there vs. the amount of physical gold.  An often-mentioned figure is that there are approximately "100 times" as much paper gold (futures, gold leases by/from bullion banks, etc.) as there is physical.  I am not in a position to know whether this is indeed true, but I have seen this mentioned many times.

Many derivatives (and that's what paper gold really is, a series of derivative products) have two parties, a "buyer" and a "seller".  Thus, it is not clear to me whether just multiplying the the amount of physical gold's price by 100 (and risk "double counting") or 50 is more correct.  Keep in mind my initial remarks, that these calculations are exploratory in nature, really a tool for thinking about this issue of "correct price in the future for physical gold".

Also, it is probably NOT correct to just multiply the price of gold times 50 or 100 to arrive at "correct prices" but it is likely that we would be approximately correct!

$1360 * 50 "paper claims" on each oz of gold = $68,000 per oz

$1360 * 100 = $136,000 per oz

My best guess, GUESS!, is that the $68,000 number is likely a better value.  If so, "FOFOA's numbers" again look good.

***

Still a different way to look at a future price of gold would be to think about wealth.  Gold is probably best thought of as a store of value, or as a form of wealth.  Wealth is possessed by whomever owns it.  Trying to calculate gold's portion of wealth is pretty tricky and full of arbitrary decisions, therefore I cannot make any kind of good judgements as to whether or not gold's present or future prices make sense.  Wiki again provides some insight into "wealth": http://en.wikipedia.org/wiki/Wealth (the wiki article shows as it FIRST picture several gold bars as a symbol of wealth).

Wealth is typically thought of as the net amount of land, personal property and money that someone (or a family) owns.  Personal property (cars, house furnishings, etc.) is clearly not as much as either land or money (money in its various forms, which include stocks & bonds, etc.).  Wealth, depending on the definition you accept, can include intangibles like education and health.

Wealth is unevenly distributed around the world.  The total measurable wealth per capita in the USA is about $130,000.Worldwide the average is about $23,000.

http://en.wikipedia.org/wiki/World_distribution_of_wealth

very rough worldwide total wealth would be about $150 trillion.  An even rougher estimate of the portion of total wealth represented by gold would be:

7 billion people * $1360 / oz = $9.5 trillion or 6.3% of world wealth is held in gold.  That number includes gold held by central banks and in jewelry form (the latter of course would count as wealth, but central bank holdings..., perhaps so, wealth of a nation).

Here in the USA, only about 1% of the population owns a significant amount of gold (one toz or more).


Gold's role in storage of wealth may vary through time as well.  Long ago (ancient Egypt), silver and gold each had about the same value per ounce!


So, due to the difficulty and arbitrary nature of defining wealth (note that we would have to subtract liabilities,  land for example -- borrowing), I really do not have a good feel what the future price of gold should be re gold's role as a Store of Value for the world's wealth.

***

This little exercise does hint that gold "should" be priced higher than it is, not even taking into account that gold has no counter-party risk and all of the other reasons why physical gold advocates suggest owning the metal.  But, the above do involve an unacceptable level of arbitrariness in order to accurately make a good prediction of future gold price.  Still, I would like to get more information on how others (especially FOFOA) arrives at HIS numbers..., would he tell us?

Monday, May 13, 2013

Review Of American Hard Assets -- Issue No. 3

I was pleased to receive the third issue of American Hard Assets on Friday.  The issue is loaded up with the usual articles than many of us are interested in...  The Cover highlights the security theme: "Secure Your Assets", as there are six articles related to the whole theme of security.

***

Before the Cover Story are there "World New Updates" columns.  Thieves stole $50,000,000 in diamonds from Brussells Airport, the diamonds were stolen right out of the cargo hold soon before take-off to Zurich.

Two metal detector enthusiasts on Jersey (UK) found a BIG HOARD of Iron Age coins.  These are thought to have been buried just before Roman General Julius Caesar showed up.  Estimated worth: perhaps $15 million!  Wow, guys, way to go!

In New York, two friends bought a Long Island home as an investment, they paid $2500 more to get the paintings of recently deceased Armenian-American painter Arthur Pinajian.  They got 3000 paintings!  Value could wind up being up to $30 million!  Another big score!

***

The first major story is by our own Department of Homeland Security (DHS)!  "Trade-Based Money Laundering" is about how various bad guys (mostly drug-smugglers (aka Transnational Criminal Organizations, TCOs) and terrorists) are using gold to launder their ill-gotten gains.  Trade-based money laundering is a set of schemes that use the international trade system to disguise money flows.  A classic example is under-invoicing by importers, in which an importer asks the exporter the state that their import shipment is worth LESS money (for example) so that they pay less import duty, etc.  Depending on what the dirty international trader wants to accomplish, under-invoicing and over-invoicing (the opposite, if the importer wants to send money OUT of his country for example) have been used...

The authors (David M. Marwell and Hector X. Colon) are part of DHS's Homeland Security Investigations (HSI), a group I had never heard of.  HSI has over 7000 special agents, and is the second largest criminal investigative agency in the entire federal government.  How about that!

Marwell and Colon describe various schemes that the bad guys do, some history of and what HSI does and their work with precious metals dealers to combat trade-based money laundering.

One interesting scheme is that money-launderers have been melting down gold and painting it to look like hand tools.  There are MANY other schemes used by these guys.

Editor comment: It would be MUCH easier to be sympathetic to HSI (and the IRS, etc.) if our government were not showing such totalitarian tendencies...  While I hate terrorists and drug-smugglers as much as the next guy, the militarization of our law enforcement (especially DHS, the agency also responsible for the odious TSA) has made millions of us hostile to the whole growth of government's law enforcement...

Just above I mentioned the DHS, TCOs, HSI, TSA and IRS.  Do these government guys like their acronyms or not?

***

Mark Brasfield (Nashville Safe House, a merchant of safes I imagine) writes "Ten Things to Know When Buying a Safe".

He discusses gun safes (typically large but not very strong), fire safes and burglar-resistant safes.

Also, if you buy a safe, get one a little bigger than you think you may need, think about the future!

***

AHA Editor Ryan Kasmiersky interviews Trident Response Group (TRG) President Clint Bruce.  TRG offers many of the services that many companies in that space do: training, intelligence gathering and the occasional foreign rescue.

Bruce details what they do as well as what individuals and families can do to protect themselves in dangerous places.

***

Wayne Simanovich (visit his website: www.simanovich.com) writes an article on guard dogs.  Guard dogs, of course, cover a part of home security beyond that of alarms and having a gun handy.

***

"Company Spotlight" features The Armored Group (TAG -- acronym city today, folks), a manufacturer of armored vehicles.

They just received "blast certification" for their Toyota Land Cruiser and Toyota HiLux vehicles.

TAG's manufacturing facilities are ISO 9001 certified.  Their customers include the US Army, the US Air Force, other US agencies, the UN and other customers around the world.

***

Fred Reed (who writes on numismatic matters, including for AHA) writes an interesting article on numismatic crime.  It turns out that theft of valuable coins is rather common.  So common, in fact, that a retired police chief and now coin show promoter Doug Davis has founded the Numismatic Crime Information Center (NCIC) to coordinate information flows regarding thefts of valuable collector coins.

Reed writes that he found over 21,000 Google hits on just the exact terms "coin theft".

Reed also goes on to describe a lot of such thefts, many are connected to TCOs...  There is occasional BIG MONEY, Reed describes a robbery of a coin worth $1,000,000.

Be careful, vendors, at coin shows!

***

"Online Buying Guide" (c'mon, guys, all articles should mention who the author is...) outlines buying gold and coins via the Internet.  It is suggested that doing your due diligence is important.

Described and mentioned favorably (selection, shipping, prices, etc.) are:

APMEX (apmex.com)
Gainesville Coins (gainesvillecoins.com)
Provident Precious Metals (providentmetals.com)

***

Doug Kale writes "Top 5 Free Gold Apps"!  I already use kitco's (kitco.com) for my iPhone.  Kitco's is mentioned and given five stars.  OPM Gold Alert and Merite Gold are also rated five stars, Gold Price Live abd US Gold Bureau are rated at four.  All of those are free.

Kale also rates Gold Tracker at five stars, but that app that costs $2.99.

***

Ed Estlow (who wrote on diamonds last issue and is a jewelry designer: www.edestlow.com) writes an article on other gemstones (that is, other than diamonds).

He suggests that if anyone wants to get into other gemstones, that you ought to just go ahead and get certified as a gemologist or seek out the advice of trustworthy experts.

***

Scott Wayne writes ("Let's Get Physical") that it is better to pay a little more and have your gold stored as "allocated" gold (the bar is actually yours, you get the bar info, etc.) rather than "unallocated" (pooled), where you would own a percentage of a gold pool...

Why not just find a place at home?  Corzine stole allocated gold.

***

Here we go, Mark O'Byrne (of GoldCore fame, probably familiar to readers of Zero Hedge) writes on rhodium!  South Africa and Russia produce 96% of the world's rhodium.

Rhodium is a platinum group metal that is used as a catalyst as well as for plating silver jewelry.  Rhodium has also had a history of majestic rises and falls in price (it now runs some $1300 for a one oz bar from Kitco), it reached a peak of $10,000 just before the financial crash in 2008.

O'Byrne writes that rhodium is the rarest of the precious metals.  But, low investor (and general worldwide) demand keep the price reasonable.

(the RAREST non-radioactive metal is osmium, also in the platinum metals group, but osmium has very few applications)

***

Nicholas Forrest writes about alternative ways to buy art.  The art market is opaque and really no very "investor-friendly".  Most art is illiquid, there is no standardization (part of the appeal of fine art, of course).  Still, there is investor interest, and where there are investors with money, a way will be found for them to invest.

And so this is happening in the art world.  There are many types of investment funds now set up to invest in art, Forrest writes.  The global art investment fund market increased an impressive 69% from 2011 to 2012 (with 2012 amount an impressive $1.62 billion).  There is even a fund dedicated to just violins: the Marquis Carbonelli Investment Company Limited!

The most well known is the Fine Art Fund out of London, they even have launched several other funds to seek out niches.

An industry association is there to provide information: artfundassociation.com.

Why the demand?  These funds hire experts, and investors can have relative safety in owning a piece of different artworks.  There are other variations too: the Liquid Rarity Exchange seeks to be able to securititze classes of artworks.  Hmm, uh, I dunno...  There are other efforts underway to have investors participate in renting out art to corporate accounts Art Equity of Australia buys paintings from $10,000 - $50,000 and rents them out (www.artequity.com.au).

***

Numistatists would likely enjoy John Dale Beety's article "Odd Denominations".  He examines various US coins with unusual denominations (half-cent, two-cent, three-cent and 20-cent coins).  Nice pictures.

***

Once again, collecting is profiled in an article in AHA.  To collect properly, it seems, you should:

-- have a long-term plan
-- learn about what you collect (yes, of course)
-- get information from experts and anywhere you can
-- have lots of money...

OK, I wrote that last one, but it's true.

***

The most interesting advertisement of the issue is form Beretta!  They have a new "technopolymer" frame automatic pistol (similar I would think to the Glock) in 9 mm Parabellum, it is called the "Nano".  Very impressive looking and made in the USA.

***

Matt Insley (dailyresourcehunter.com) wonders how much gold China really has.  China SAYS they have a (lowball for sure): 1054 metric tons ("tonnes"), according to him "we know" they have about 3900 tonnes, he goes on to calculate that they have 7000 + tonnes...  The USA has about 8100 tonnes if no one is lying.

And China buys more every month.  They are the world's largest producer (yes) yet export NONE of it.  They import a lot via Hong Kong, that is the source of a lot of people's guesswork in trying to figure how much gold really has.

***

AHA's "Mining News" columns describe recent events El Dorado Gold's Greek gold mine (lots of protest against it there).

New executives promising (really, they really are promising, really!) to cut costs at BHP Billiton, Newmont and Anglo American.

Hecla Mining appears to be on track to restart its Lucky Friday silver mine in Idaho (training..., maybe they had an accident?) as well as agreeing to buy Aurizon Mines (of Canada).

Turquoise Hill Resources (on the NYSE: ticker TRQ, although they are majority owned by Rio Tinto) is operating a BIG copper and gold mine (Oyu Tolgoi) in Mongolia.  They have suffered from numerous delays, but they may be back on track.  This is truly a big mother mine, I have been reading about on occasion for years.

***

"HindSight" is again written by John W. Garibald.  He discusses the case of Cyprus, and how that small country has affected so much.  This is turf that Zero Hedge has covered very well, so I will not repeat what is already well-known.  Garibald wonders about the future implications of Cyprus and the whole situation in Europe.

And well he should.

***

Verdict: Yes, if you can find this magazine and have interest in the above topics, buy it!

Thursday, May 9, 2013

Gold: Bits & Pieces

Over the past several days I have dipped into various topics of the Gold Universe, none of which seemed like they were big enough to devote a whole post to.  But, I want to at least mention each one, and discuss it to a degree or other.  I do not want to forget any of these!  And one or more of these may get further conversation going.

I titled this article in part using the words "bits & pieces", I would like to first explore tiny bits of gold and their utility.  I am starting to really like the very small 1/10 oz American Eagle coins, and even the 1/4 oz pieces as well.  The smaller coins are not as efficient a way of buy the most gold for the money, but they are a readily identified smaller amount of gold than the large 1 oz coins.  I now have the below "collection" of smaller gold bullion coins:

1/10 oz Gold Eagle
1/10 oz Maple Leaf
1/4 oz Gold Eagle
1/4 oz Maple Leaf
1/4 oz Krugerrand

I like these small pieces because they conceivably could be used in extreme TEOTWAWKI situations to make larger purchases than you probably could not easily do with silver, say a month's worth of groceries for the family or a used motorcycle.  And not have to worry as much about getting "change" back versus a 1 oz coin.  [They also make fine gifts, for raising consciousness of gold for example...]

***

And, yes, you can see a *slight* color difference between the Eagle and the K-rand.  The Eagle has about 3.0% silver as part of its alloy (the other 5.17% is copper) , while the K-rand has just 8.17% copper, so the latter is slightly more reddish in color.  Since I am putting up stuff in this post tonight, here is wiki's color-triangle on gold and its more common alloys (*click* on any image for a better view):



According to wiki "pure gold" (999 fine, at the top of the triangle) has a slight red-yellow color, although the 1/4 Maple (999 fine) looks pretty yellow to me...  Link to above: http://en.wikipedia.org/wiki/Gold

***

A couple of times now I have wanted to use FOFOA's very interesting graph from 2009, whether at Zero Hedge discussions or elsewhere.  Here is his graph, for my easy reference as well as for any of you!  Note, these are his 2009 estimates:

Note his annotation of "2009 non-hyperinflated dollars".  There seems to be misunderstanding of FOFOA's prediction the recent Zero Hedge thread about the paper-gold system coming to an end: http://www.zerohedge.com/news/2013-05-09/are-we-verge-witnessing-death-paper-gold-scam#new

It is my humble opinion (and I emphasize that it is humble, in that I have already written about my poor record as a predictor of things, especially in the future...), that even if the price of gold winds up at $25,000 (now considered fantastically high by most gold analysts -- "fantastic' in the sense of extreme and unlikely), well his whole set of arguments will be proven correct...

["Humility" is a subject that I am becoming more interested in...  I hope to write about that soon.]

And I would have a lot more money (however you define it),,,, what's not to like about that?!

***

FOFOA's "Freegold" is becoming a more visible meme now...  The destruction of "paper-gold" is discussed at the ZH link above, but this is not the only place outside of FOFOA-land where Freegold is getting noticed.  Why even wikipedia now has an article on Freegold!  There is, however, considerable controversy over how his ideas can be simplified to reach a wider audience.  And wiki's take is not all that persuasive, IMO, but at least it is there for the world to see, and this may jump-start additional conversation:

http://en.wikipedia.org/wiki/Freegold

Even "General Jim" Sinclair (jsmineset.com) is looking at and discussing Freegold.  It is not clear to me whether he has switched any of his thinking or not.  He has always been bullish (big time) on gold, but not to the very high prices as envisioned by FOFOA.

***

A book has apparently just come out that I want to get ASAP!  $10,000 Gold (by by noted gold analyst Nick Barisheff). I saw an ad for it just today, clicked on it and read the chapter that they (amazon.com?) allow you to read.  Nothing NEW in that chapter, but $10,000 in the book's title is enough to get lots of attention.

If I can get the book and read it soon, I will review it.  [But, there are other books in the queue as well...  The book I am reading now is about the Italian 'Ndrangheta -- Calabria's mafia.  More mobile and powerful than even the Sicilian mafia.  This book is very interesting: The Honored Society by journalist Petra Reski, 2013]

***

In my wanderings about the Internet I have run into Ohio Precious Metals' nice looking website (opmmetals.com).  They are a refinery in Jackson, Ohio, and are apparently the USA's largest refiner of gold bars that are COMEX certified.  All of their feedstock is recycled gold (source is mostly all the "We buy gold" places), they even have the three revolving arrow logo of certified recycled products.  Check out OPM's kilo bars (photo courtesy of Provident Metals (providentmetals.com):


Each kilo bar is about 117 mm * 51 * 9 (roughly 4 5/8" * 2 * 5/16), reference: http://www.goldbarsworldwide.com/PDF/NBA_76_Ohio_PreciousMetals.pdf.  Most other kilo bars have similar dimensions.

goldbarsworldwide.com is a wonderful website, hey, just sayin'...  Lots of pictures and useful reference data there.

Ohio Precious Metals is owned by privately held Global Metals Holdings, LLC (which also owns NTR Metals).

Provident, by the way, is now offering the above kilo bars for a very reasonable looking $5.95 per oz over spot (beating tulving.com by about $5.00 per oz, NTR suggested that I call Provident about OPM's kilo gold bars).  Three week delivery however...  And, do your own diligence!

***

Finally "a bit" (or a piece) for knowledgeable FOFOA follower "Michael dV", who has been kind enough to comment on my blog.  Michael takes a pretty hard line vs. silver, as many of FOFOA's followers do.  The main idea behind silver being dissed is that the choice of Store of Value of wealth has already been chosen: gold!  And so, silver will not ride gold's coattails in a price reset dramatically higher than precious metals prices now.  Just gold will go up 30 times...!  Well, I am not so sure (I have learned recently to be less sure of things, particularly in the future).

Michael commented at my recent "Preparations" piece, and is dubious about the value of silver in a SHTF, saying that fiat currency will work just fine.  I had no good response for him until now.  Yes, he is right about a need and use for currency, there will always be a need for a currency (even paper fiats).

But, if it were my gas station (liquor store, pharmacy, booth at the farmer's market, etc.), and the chips were really down, would I prefer to be paid with currency, or silver?  Hmm, for me that's an easy one!

Tuesday, May 7, 2013

Factor Analysis Of Ameru Sales Data

My blog has suffered a bit because one of my computers was down (now fixed!) and because I have been spending time with our sales data, exploring it to see what I can learn.

I have long been interested in a multivariate statistical procedure called "Factor Analysis", which takes a bunch of variables and "reduces" them into a fewer number of factors that hopefully will adequately model the data.  Each factor represents bearings that tend to sell together (if a customer buys some of one bearing, they are likely to buy some of one or more others).  Another way to think of a "factor" is as if they are "classes" of bearings, they more-or-less go along in the same basket...

In general, what we seek when using Factor Analysis are clean & descriptive factors that show a high correlation of bearings within that factor group as well as having each factor distinctive from the others.  Correlation is measured by the correlation coefficient ("C.C.", the second column in each table below), here all C.C.'s are positive (but they can and typically do in many datasets range from -1.00 to + 1.00).  In general, in the "social sciences" a C.C. of 0.3 is considered "significant", in that the two variable measured against each other would have a "moderate" relationship.  I am not looking for moderate relationships, however, I am looking for strong ones, and C.C.'s of about 0.7 or so are pretty strong, with 0.8 being even better and 0.9 almost a straight line relationship (charts below will illustrate these C.C.'s).

***

Before I go further into Factor Analysis, here is a brief overlook of our data as well as what correlation coefficients look like!

The below chart shows the sales of our top 248 bearings.  Our sales are "exponentially distributed", much like the famous "80/20 Rule" that you read about (how your top 20% of top pieces represent 80% of sales, etc.).  Actually, our data is more skewed than that, we are about 88/12 or so.  The first is all 248 bearings studied, note how important our top selling piece (KB-6007-RS) is!  The second chart is the same but with the outlier KB-6007-RS piece removed, the second chart looks much more like a classic exponential distribution.



Here are some examples of what different correlation coefficients look like (I made these numbers up and plotted them, but the correlation coefficients are accurate).  Note how close to a straight line 0.95 is, you can probably easily understand why I like to see high C.C.'s in my work!  A set of high C.C.'s will give me an idea of how many pieces to buy of each item in that group (factor), for example if we want to buy 500 pieces of our best-selling Daewoo Tico piece, I can be confident in buying, say, 200 of our next best-seller for Tico.  Imagine that each of the below graphs shows sales of each piece vs. its factor for a one month period:




***

Onward.  Basic info on the programs and dataset:

--  Program: S-Plus v. 6.1.2 (2002)
--  248 bearings (variables), representing some 79% of our sales, (total of +/- 1200 pieces)
--  400 customers (representing some 97% of our sales), total of 700 or so customers
--  Time period: 2011 - April 2013

Procedure:

All data comes from my MS Access analytical database.  That data comes up from Peru on MS Excel in a different format.  I took the Access data and made a "crosstab query", and found that Access limits crosstabs to 250 columns.  OK, then I looked at our top 248 top selling bearings.  I transferred this data to Excel, as S-Plus works with Excel.  After I had S-Plus' version of a worksheet, I was then able to execute the Factor Analysis routine.

Now that you have an idea of our overall sales data and what C.C.'s look like, we return to look at the results of the Factor Analysis I did on our dataset.

Explanation of below results:

-- I chose bearings that loaded highly (high C.C.) with each factor.
-- I chose 16 factors to be extracted.
-- "Bearing Ref" is our internal coding, the first two alphanumerics are the brand.
--  Bearing coding is quite irregular!  It is like having to learn languages...
-- "C.C." is the correlation coefficient of each piece to its Factor group.
-- "Factor" is the factor number, Number 1 is the most important.
-- "Eigenv." means eigenvalue, here equivalent to how important (loosely explained...).
-- "Comments" shows factor description or other info of note
-- "2011-13 Rank" is importance of each piece (sales, see highlighted bearing below).
-- "2013 Rank" is just for 2013 sales, there are changes, so I monitor them.

Factor 1 just below are (almost) all bearings for "Daewoo Tico", a small Korean & very common car in Peru, there are three bearings for the similar "Daewoo Matiz" that uses a few different pieces (here three).  Factor 1 has an eigenvalue of 14.9 (if I had run the routine to include all 248 possible factors, they would add up to 248, but I just chose to look at the 16 most important, so Factor 1 is (crudely) "worth" about 14.9 times an average bearing's sales in this dataset).

Some of these C.C.s are very high, I highlight in light blue background values over 0.95 (rarely seen in the social sciences).  I do not have any other data from other businesses, so I do not have any idea whether or not businesses see such strong relationships among their products...  Readers, care to share comments?

Some of our biggest sellers did not correlate well within just one factor (example: see last chart just above of our IL-212001).  My judgement is that these bearings sold to customers who bought lots of different kinds of bearings...  Many of our customers tend to specialists to one degree of another.  I mark three of these bearings in red in the tables below the term "highest ldg" means that these pieces do not fit well in that factor group, but it is the best fit...  You can note that the C.C.'s are rather low for those three pieces, again, I think it is because customers bought these three pieces with a variety of other bearings, not just for the same brand of bearing, vehicle model or type of vehicle (eg a bus).

This work is unfinished!  Note that I have made no comments because I am still trying to figure out how all of these pieces fit together!

High Factor Loadings -- Ameru Bearings (Data 1 May 2013)

2011-13 2013
Rank Rank
Bearing Ref. C.C. Factor Eigenv. Comments
C13RCT283 0.211 1 14.9 Daewoo Tico 15
C3RCT283SA 0.76 1 14.9 Daewoo Tico 44
KB30204 0.545 1 14.9 Daewoo Matiz 37 28
KB30205 0.518 1 14.9 Daewoo Matiz 21 21
KB32004X 0.824 1 14.9 Daewoo Tico 18 24
KB33005 0.754 1 14.9 Daewoo Tico 12 17
KB6000.2RS 0.964 1 14.9 Daewoo Tico 4 5
KB6002.2RS 0.929 1 14.9 Daewoo Tico 14 18
KB6007C3 0.813 1 14.9 Daewoo Tico
KB6007.2RSC` 0.811 1 14.9 Daewoo Tico 3 16
KB6007.RS 0.954 1 14.9 Daewoo Tico 1 1
KB6204A 0.866 1 14.9 Daewoo Tico
KB6204.2RS 0.981 1 14.9 Daewoo Tico 8 9
KB6205.RS 0.98 1 14.9 Daewoo Tico 7 8
KB6206.20 0.867 1 14.9 Daewoo Tico 30 34
KB6207.HLC 0.778 1 14.9 Daewoo Tico
KB6302.2RS 0.92 1 14.9 Daewoo Tico 11 11
KB68149.11 0.549 1 14.9 Daewoo Matiz 9 6

Below is Factor 2, note that I need to do more work to interpret this one better.  They are all wheel bearings, double-row type and hub & bearing assemblies (latter has a double-row bearing inside a steel housing).

DE255237 0.649 2 14.4 Suzuki
DE27BWK04 0.652 2 14.4
DE27BWK04.2A 0.717 2 14.4 26
DE27BWK06A1 0.616 2 14.4 Daihatsu/Toyota
DE30BWK11 0.584 2 14.4
DE387038 0.705 2 14.4 Daihatsu/Toyota
DE387139 0.568 2 14.4 Toyota Probox 19 19
DEE90A 0.474 2 14.4 Nissan -- highest ldg 2 10
DEHUB042.32.82 0.713 2 14.4 22
IL141008 0.515 2 14.4 Toyota Probox
MB38BWD22 0.671 2 14.4 Toyota Probox 5 4
MB40BWD15 0.653 2 14.4 Nissan
MBMUB0814 0.645 2 14.4 27 23

Factor 3, more work as well...

CH3874 0.925 3 14.2 Toyota
DE27BWK03ABS 0.612 3 14.2
DE2DUF050.7 0.819 3 14.2
DE51KWH01 0.735 3 14.2
DE54KWH02 0.921 3 14.2 Toyota Hi-Ace 25 10
DEHUB042.32 0.745 3 14.2
IL141005 0.941 3 14.2
JV949100333 0.808 3 14.2
JV949100348 0.911 3 14.2
JV949100382 0.885 3 14.2
IL212001 0.463 3 14.2 highest ldg 3 2
KB800736 0.595 3 14.2
KB800738 0.823 3 14.2
KBF569214 0.702 3 14.2

Factor 4 is almost all "tapered roller bearings" ("Timken type"), for wheels of commercial vehicles.  The first is an exception (it is for a truck, but it is a clutch bearing) while the last piece is an automotive wheel bearing that slipped in here...

C458TKA3703 0.5 4 13.8 +…
KB30207 0.758 4 13.8 trbs for Com Vehs +
KB30208 0.603 4 13.8 trbs for Com Vehs +
KB30210 0.722 4 13.8 trbs for Com Vehs +
KB32011X 0.689 4 13.8 trbs for Com Vehs +
KB32207 0.843 4 13.8 trbs for Com Vehs +
KB32209 0.737 4 13.8 trbs for Com Vehs +
KB33012 0.846 4 13.8 trbs for Com Vehs +
KB33013 0.508 4 13.8 trbs for Com Vehs +
KB3780F1.20 0.525 4 13.8 trbs for Com Vehs +
KBHC6307TH 0.803 4 13.8 trbs for Com Vehs +
KO02474.20 0.676 4 13.8 trbs for Com Vehs +
KO28584.21 0.764 4 13.8 trbs for Com Vehs +
KO30210 0.855 4 13.8 trbs for Com Vehs +
KO30211 0.737 4 13.8 trbs for Com Vehs +
KO30212 0.785 4 13.8 trbs for Com Vehs +
KO32207 0.608 4 13.8 trbs for Com Vehs +
KO3579.25 0.711 4 13.8 trbs for Com Vehs +
MBM4074CW 0.779 4 13.8 +…

Factor 5 is double-row bearings, hub & bearing assemblies and three unusual sizes of KBC bearings, the last piece is KBC as well, but is a double-row bearing.  Item IL-111001 is a double-row piece for Hyundai Accent, it is our 6th best seller both for this year 2013 as well as for the longer period 2011 - 2013.

DE28BWK06D1 0.679 5 13.6
DE28BWK15 0.708 5 13.6 40
DE28BWK19 0.863 5 13.6
DE42450.02070 0.555 5 13.6 Factor 6 too.
DE513297 0.519 5 13.6 62
DE515013 0.955 5 13.6
DE498448 0.719 5 13.6
DE559060 0.698 5 13.6
IL111001 0.391 5 13.6 highest ldg 6 6
IL111004 0.544 5 13.6
IL223003 0.688 5 13.6
IL223028 0.859 5 13.6
KB6308.32 0.574 5 13.6
KB806023 0.811 5 13.6
KB806093 0.825 5 13.6
KBDT255237 0.669 5 13.6

Factor 7, bearings for old cars, unusual pieces, etc.:

IL131006 0.592 7 11.6 Peugeot
IL131008 0.797 7 11.6 Peugeot
IL710001 0.683 7 11.6 housing for bearing…
KB576079 0.506 7 11.6 KB oddball
KB62.28.2RS 0.677 7 11.6 KB oddball
KB63.28.2RS 0.639 7 11.6 KB oddball
KO30207 0.684 7 11.6
MBM0895W 0.685 7 11.6 Chrysler
MBSet10 0.571 7 11.6 old car rear wheels
MBSetM80 0.554 7 11.6 old car rear wheels
MBSetM9 0.651 7 11.6 old car rear wheels

Factor 8 is "bread & butter" pieces for Korean cars (wheel bearings).  FYI, "carnecitas" is Peruvian slang for nice selling pieces ("little meaty ones").

IL111002 0.601 8 10.4 K. carnecitas for Korean 54
IL111003 0.876 8 10.4 K. carnecitas for Korean 15 29
IL111004 0.635 8 10.4 K. carnecitas for Korean 63
IL111006 0.829 8 10.4 K. carnecitas for Korean 120
IL111009 0.847 8 10.4 K. carnecitas for Korean 26
IL112001 0.588 8 10.4 K. carnecitas for Korean 94
IL113002 0.91 8 10.4 K. carnecitas for Korean 13
IL113006 0.549 8 10.4 K. carnecitas for Korean 133
IL113015 0.891 8 10.4 K. carnecitas for Korean 17
IL113017 0.917 8 10.4 K. carnecitas for Korean 81
ILH0400.1 0.596 8 10.4 Factor 10 too…
KB11749.10 0.752 8 10.4 K. carnecitas for Korean 82
KB38KW01 0.438 8 10.4 Hyundai, highest ldg 34
KB44649.10 0.512 8 10.4 K. carnecitas for Korean
KB45449/10 0.733 8 10.4 K. carnecitas for Korean 111
KB69349.10 0.647 8 10.4 K. carnecitas for Korean 60

Factor 9 is our cheap large tapered bearings from ZWZ (China), these are trailer bearings mostly.  Note high correlations within this group.  These bearings did not correlate at all with other brands...  This hints at just buying these bearings once in a while from this supplier, only few (4!) other pieces sell well for us.

ZW212049.11 0.924 9 8.7 ZWZ trailer bearings 67
ZW218248.10 0.925 9 8.7 ZWZ trailer bearings 50
ZW32219 0.829 9 8.7 ZWZ trailer bearings 56
ZW33213 0.924 9 8.7 ZWZ trailer bearings 165
ZW518445.10 0.864 9 8.7 ZWZ trailer bearings 38
ZW580.572 0.829 9 8.7 ZWZ truck bearing
ZW594A.592A 0.862 9 8.7 ZWZ truck bearing 140
ZW663.653 0.885 9 8.7 ZWZ trailer bearings 83
ZW683.672 0.779 9 8.7 ZWZ truck bearing 64
ZW749.742 0.568 9 8.7 ZWZ trailer bearings

Other than ZWZ bearings, we are fairly weak in big truck bearings, we have to skip down to Factor 12 to see more.  The items with "truck" in column 5 are either BIG bearings or moderate sized pieces for big trucks (pinion bearing or transmission).

KB598AS.592 0.568 12 6.5 truck bearings…
KBTR6513051 0.424 12 6.5 truck bearings…
KO30309D 0.525 12 6.5 truck bearings…
KO30311D 0.713 12 6.5 truck bearings…
KO32207 0.56 12 6.5
MB88509 0.545 12 6.5
MB8MTH7205R 0.684 12 6.5 truck bearings…
MB309L 0.791 12 6.5 truck bearings…
MBUY1307TM 0.545 12 6.5 truck bearings…
NSB17.99 0.651 12 6.5
ZW749.742 0.473 12 6.5 truck brg, Factor 9

KBC makes a few pieces that few others make, for Korean vehicles, Factor 16 shows this well, note the low eigenvalue of 3.6, this explains relatively little of our overall sales...

KB30306C 0.703 16 3.6 K. oddball trbs for K.
KB30307C 0.702 16 3.6 K. oddball trbs for K.
KB30308D 0.375 16 3.6 K. oddball trbs for K.
KBTR458020 0.628 16 3.6 K. oddball trbs for K.

...

S-Plus is kind enough to advise if the number of factors chosen is a "good model" or not, using a statistical notion of "p-value".  In this case, I need to get a p-value of over about 0.20 or so for the model to be pretty descriptive of the whole dataset (248 bearings).

With the above 16 factors extracted, my p-value was zero.  So, I tried extracting larger numbers of factors, the highest number I tried was 27 factors.  Factor 27 had an eigenvalue of 0.88 or so, less than one, so that looked to be about all of the useful relationships I was likely to find in each of the 248 bearings and their correlation coefficient with each factor.  Even extracting 27 factors, however, the p-value is still zero...  But, the three below factors (numbers 19, 22 and 23) did show some decent correlations among some pieces and their factor.  Note Factor 22 finally broke out the Daewoo Matiz pieces from Factor 1.  Factor 19 shows me that the wheel bearings for older Toyota Hi-Ace (a van used a lot in public transport in Lima) sold similarly in both KBC and Koyo brands.  Factor 23 are two Koyo ("KO" prefix) pieces that are a different type of bearing (belt-tensioner bearings, as in the bearing used in fanbelts for alternators, etc.).

                                    Factor    Eigenv.       Comments

(27 Factors…) 19 3.6 KB & KO Toyota Hi-Ace
(27 Factors…) 22 3.1 Daewoo Matiz
(27 Factors…) 23 2.4 KOPU1262 & PU2462