Particularly as demand for physical gold does not appear to have slowed down, as shown in Figure 1a of queuing customers inside a gold shop in the gold district of Yaowarat Road in Bangkok's Chinatown in Thailand on Sunday morning 23 June 2013.

For the past 3 years on 28 June, the US$ gold price has been ~$1600/oz (2012), ~$1500/oz (2011), and $1260/oz (2010). The last time gold was ~$1192/oz was May, July or August 2010.

With the fall in the gold price there have been a number articles trying to unravel its mystique, as there are a number of conflicting and confusing issues. Take a known factor like the official reported central bank gold reserves, the Gold Survey (formerly GFMS) had a total of ~32,245t at the end of 2002, which is close to the 32,200t of the World Gold Council (WGC) at that same point in time.

However, from the end of 2002 to the end of 2008, the Gold Survey's CB reserves fell by ~6,000t to 26,350t, whereas the WGC's supposedly same figure had only fallen by 2,400t to 29,800t. We don't have the updated figure for the Gold Survey, but the WGC to 30 Apr 2013 now has holdings ~31,800t for the central banks, including the IMF, BIS etc (or an ~2,000t increase in their figures since the end of 2008). The WGC's figures do include swaps and trading which may account for part of the difference.

The main focus for setting the gold price has gradually become "what happens in Comex trading". Historically the focus was on London and many (like us/ERA) still use the London pm gold fix, and there was an endless debate on the significance of Comex compared to London. Although, we have noticed increasingly a 3 point timing focus on the gold price, being the open and close in : London, New York & Hong Kong (in 24-hour trading).

It is generally acknowledged that the LBMA (London Bullion Metals Assoc) handles about 80% of the global gold trade compared to Comex at 20%. However, Comex publicly discloses a number of statistics about its trading whereas the LBMA (as far as we know) doesn't.

A recent (27 June 2013) report by Incrementum AG entitled "In Gold we Trust 2013" referred to some statistics from The Alchemist (Q1 2011), from which it can be determined that in 2011 the LBMA was trading at an average of ~6,000tpd of gold or ~30,000t of gold per week. Little seems to have changed, we can recall attending a Gold conference many years ago at which a central banker commented that the total world (CB) holdings of gold are turned over every 3 to 5 days - and they still appear to be.

Some surprising statistics on the gold market have been unveiled such as comments that the net commercial short position at 3.52moz is the lowest it has been in gold since August 2002. Or in other words the commercial traders have been mostly net short gold (their total short positions [expecting the gold price to fall in 1 to 3 months] have been more than their total longs [expecting the gold price to rise]). Or in other words they have been mostly expecting gold to fall since August 2002. Just as a point of information, the gold price in August 2002 was ~US$310/oz.

Apparently the commercials were last net long in 2001, when the gold price was ~US$270/oz. The bullion banks were briefly net long on 4 June 2013, and prior to that July 2008 with gold at ~$940/oz near the bottom of the GFC.

The long and short positions are summarised in the COT (Commitment of Traders reports on the US CFTC website) plus the percentage of the top 4 (usually referred to as "da boyz" and comprised of JPMorgan, Merrills, Goldmans & Canada's Bank of Nova Scotia") traders and top 8 traders or "commercials" (non-commercials being small and large [hedge fund] speculators). In total there are ~300 traders in gold on Comex.

Having a net short position does not clearly reveal the actual short position which was apparently about 10.2moz for the top 4 and 4.6moz for the 5 to 8 (or ~14.8moz for the Big 8) being ~46% of Comex trade. The total short position being ~15.8moz for the main commercial traders, or ~33moz in total short gold positions (excluding the spreads, total open-interest being 39moz).

This of course compares to Comex's actual gold holding of ~7.5moz, of which 1.35moz is classified as registered and seems to be steadily reducing as it is being withdrawn on an almost daily basis and then being held elsewhere than Comex. There have been comments that at some stage Comex will not be able to settle in gold and may then only be able to settle in cash.

To the above short position can be added the short positions on GLD or the ETF's which are about 2.9moz. And then there are the short positions on the LBMA that are undisclosed. At the time of the "raid" in April 2013, it was widely rumoured than one of "da boyz" were overexposed with short positions on the LBMA in both gold and silver, with the gold short covered in the subsequent 4 days of the gold price falling through $1550/oz.

While it is understandable that the silver price falls with the gold price due to trading ratios (such as gold:silver) and there are by-product relationships, one of the major puzzles has been that all the precious metals have been rising and falling together at almost exactly the same time, or platinum and palladium are moving along with gold and silver. Platinum and palladium have nothing ore deposit-wise to do with gold and silver, they are not significant by-products of gold or silver and vice versa (gold and silver are not significant by-products of platinum and palladium).

It is no wonder that the manipulation question has been raised.

There is a small metal sludge in the gold process that holds pge and base minerals that is sometimes called osmiridium (being a name derived from combining osmium and iridium - but it actually contains mostly platinum and palladium and unextractable [if any] osmium or iridium), and can be extracted by a specialist company.

All these short positions in gold are OK provided no one actually wants delivery, because then the shorts can get squeezed and the gold price possibly rise to a peak. Although the gold price run to $850/oz was mostly attributed to the oil price and to some degree rising inflation, there was probably a short squeeze too.

The perception of the gold price has changed significantly since 1977 when we first modelled companies at $110/oz with a lower level of $90/oz and an upper level of $130/oz (which at the time was regarded as overly optimistic), but within only 2.5 years, a gold price of US$850/oz was achieved on 20 January 1980, which shows that the gold price can increase quite significantly over a short period of time. Admittedly communications were much slower then, the increase to $850/oz was by observation of numbers on a Reuters "ticker" tape / print-out machine.

This potential volatility is all very well for the trader, but gold companies cannot mine their orebodies on the basis of such volatility. At the current gold price of ~$1200/oz, exploration has already ground to a halt, construction is being minimized, mines are closing or being deferred, and the knock-on effects into communities have to gradually occur.

Some governments are going to discover that the gold mining industry is not a perpetual gravy train that can be additionally taxed whenever you need some money. There have been ghost towns before, and they can occur again. When a mine closes, the community (unless it can survive on other local trade) has historically often moved on too.

Nick Holland of Gold Fields stated on 28 June that the South African gold industry cannot survive at $1200/oz, it needs to be ~$1500/oz, and that is before the coming annual wage talks from 11 July, in which the South African mining unions are demanding that the lowest basic wages be doubled, and workers subsequently laid off have to be re-instated or the companies may face "further action".

In our September 2009 Goode News column in Paydirt (titled "The last shoe to fall"), the IMF intended to sell ~400t of gold at US$1000/oz and brokers were forecasting gold to fall to $900/oz during the coming year, whereas we thought that it was more likely to stay above $1000/oz for some time.

In that September 2009 column we referred to a CBA presentation at an Excellence in Mining Conference in Sydney (21/22 September 2009), which showed that on an inflation adjusted price, the ~US$660/oz average price for January 1980, was equivalent to US$1850/oz in real August 2009 terms (or the

1980 peak of $850/oz was equivalent to a gold price of ~US$2380/oz in August 2009 terms).

There have been many elaborate forecasts of what the gold price should rise to due to the US printing US$. Any other country that has historically printed money has faced soaring inflation and a depreciating currency. The logic was there as in gold does have an inverse correlation to the strength of the US$, it used to be based on comparison with the DMK/US$, but now has become the US$ index. (Although there have been a number of times recently when the US$ index has fallen and gold has fallen too, or they have even occasionally both risen together. Another one of those manipulation questions, as in abnormal market behaviour).

The difference in hindsight is that the countries that have failed like Germany post WWII or Zambia or Zimbabwe (all of which have used wheelbarrows of paper cash to try and buy something), were not reserve currencies. As in not the US$. Argentina's president stated on 2 July 2013, that her greatest problem was not having a "little machine" that prints US$, Argentina can only print pesos and the now new "cedines", with Argentina banning its citizens from holding US$ after 30 September 2013 (it doesn't mention gold hoarding or buying).

India has taken a war on gold restricting imports, raising tariffs, preventing lending to buy gold, so as to improve its deficit, which has caused a partial shift to silver and reversion to smuggling (again). However, while 920t of silver imported into India in May 2013 sounds like "a lot" it is only ~15t in gold equivalent terms.

During the 1980's and into the 1990's gold at ~$400/oz was perceived to be about the norm with fluctuations above and below it, achieving at least a margin of $100/oz, but with rising inflation, the gold price and its margin to rising costs gradually became too low and mines closed.

In the 1990's the ideal mine according to Anglogold had to satisfy the triple 2 (in order to be considered for acquisition), namely 2moz in reserves/resources, production of 200,000ozpa and cash costs of ~$200/oz with total costs resulting in ~$300/oz (for an expected gold price average of $400/oz). $400/oz in August 2009 terms would be ~$1120/oz.

At the Paydirt gold conference in March 2010 (and covered in our Goode News column titled "Gold Mining industry facing rising costs"), Gold Fields expected its total costs to have risen from ~$900/oz to ~1100/oz in the then coming 3 years. Under that scenario, we expected the gold price would probably need to be in the vicinity of $1200/oz to $1300/oz, for viability.

Whereas for Gold Fields and a number of other companies, their total costs are now ~$1150/oz to $1200/oz, with a number of operations in excess of that. With inflation a $100/oz margin is not enough (especially pre-tax) and perhaps the margin should be closer to $200/oz or more. That would increase the gold price to a required base level of ~$1400/oz.

While it seems that the bears would like the gold price to fall to $1000/oz or lower, that does not appear to be realistically sustainable as the gold mining industry would virtually close down, and demand appears to be too great to satisfy with negligible annual gold production.

Although it is well-known that the gold price has an inverse correlation to the strength of the US$, real interest rates (being interest rates less inflation), and the Dow Jones Index, it is also a fact that gold vacillates between being a commodity and a currency.

The unprecedented physical demand that has been almost unabated started once gold dropped below ~$1450/oz and has accelerated where it has not been prevented (eg India - also note returning Indian citizens usually used to buy their gold in Dubai). However, it should also be recognised that the physical demand in parts of Asia and probably elsewhere is not necessarily to simply hoard gold for investment, for some it is a temporary measure.

Recently when we were in Thailand, someone commented that "people don't settle large transactions in baht, they settle in gold - that's why I have some gold in the safe - it is the standard form of currency in Thailand". "Lower gold prices means you can swap your paper baht for even more gold". So part of the queuing shown in Figure 1 is probably simply conversion of paper baht into gold for use as currency, and may apply to other countries too.

As to what else is going to cause the gold price to rise, well perhaps one of the factors is that it appears to becoming more difficult to engineer its fall. In April the margin on futures was 5%, so a $20bn sale (or ~400t of gold at $1550/oz) only cost $1bn (virtually petty cash for a hedge fund, especially if a group of them collude together). That margin has steadily increased to 15%, 25%, and now 2 July : 33%. To execute that $20bn sale would now cost ~$6.7bn, or a $1bn sale becomes $3bn worth of gold (or ~78t of gold at $1200/oz, or 60t @ $1550/oz).

The recent margin increase from 25% to 33% coincided with the increase in the gold price up from ~$1190/oz on Friday 28 June. Perhaps restriction attempts on futures trading are coming to protect investors. As ASIC stated at the Stockbrokers' Conference in May 2013, "if it looks and behaves like a market, it has to be regulated", and further restrictions on futures trading were expected after the G20 meetings.

China intends to make the RMB / Yuan a reserve currency, the strength of which is usually gold (eg the US which has supposedly never sold any of its gold, but may have lent or swapped some). So China's CB is probably buying, but it is not going to declare it, because if it did, the price would rise. China has realised/learnt that it has been the cause of paying higher commodity prices, so what it needs to do is to get the prices to weaken and then buy when they are weak. Once the transaction has been done, the hand can be played openly (even if it is ~5 years' later).

Either way, the sentiment towards gold, the gold price and gold price charts has been damaged and riven with uncertainty (and markets hate uncertainty). It has

been commented that gold could rise to $1400/oz by the end of 2013, which does appear to be a realistic target, unless the shorts are forced to scramble for cover in the August traditionally seasonal demand period.

Instead of answering the question as to what should the gold price actually be ?. it appears to be more a case of what does the gold price currently need to be (probably ~$1400/oz) for the gold mining industry to have a degree of viability. Any higher gold prices realised above that result in greater profits for producers, depending on what levels they can increase their production to, at lower costs.

Disclosure and Disclaimer : This article has been written by Keith Goode, the Managing Director of Eagle Research Advisory Pty Ltd, (an independent research company) who is a Financial Services Representative with Taylor Collison Ltd.

Figure 1. A Gold Jewellery Shop in Bangkok's Chinatown on Sunday 23 June 2013