Chinese Gold Demand Slips Further But In Line With 2014 - So Far

by: Lawrence Williams


Gold withdrawals from the Shanghai Gold Exchange have been falling month on month and are down 25% year on year.

We could see the annual total fall to below 2,000 tonnes for the first time since 2012.

Chinese gold imports plus domestic production have been far in excess of consumption estimates from the major specialist consultancies.

According to the latest figures out of China's Shanghai Gold Exchange (SGE), year-to-date gold withdrawals out of the Exchange have totaled 1,090.6 metric tons (tonnes) up until the end of July. This is 25% down on last year's total at the same time - 2015 was a record year - but is still very much on a par with the 2014 figure when the annual total reached a little over 2,100 tonnes. But what should perhaps be a little worrying for gold investors hoping for a lift in Chinese demand is that July is the fifth month in a row where the monthly withdrawals figure was below that of the month before (and that has not been the pattern of previous years) - and at 117.58 tonnes is actually far less than half the withdrawals figure for July 2015 - which came in at 285.5 tonnes (see table below).

Shanghai Gold Exchange Monthly Gold Withdrawals (tonnes)
















































Year to end July




Full Year



Source: SGE,

*February withdrawals figures tend to be anomalously low due to SGE closure during the Chinese New Year holiday

Now according to the Chinese central bank - the Peoples Bank of China (PBoC) - SGE withdrawals equate to Chinese gold demand, although the specialist precious metals consultancies which publish prominent reports on global supply and demand - notably Metals Focus (which supplies the data used by the World Gold Council) and GFMS in London and CPM Group in New York, would all disagree - quite vehemently. They all place Chinese demand at considerably lower levels suggesting there are elements of double counting in the SGE figures. But primarily their far lower figures for Chinese consumption relate to their adopting fairly rigid parameters for what actually constitutes demand. They all tend to ignore, for example, the substantial amount of gold apparently being absorbed by China's commercial banking system, supposedly for use in financial transactions and as collateral.

We prefer to look at known gold flows into mainland China, plus China's own new mined gold production as the best guide for the nation's true gold demand. Countries like Switzerland, Australia, the UK and the USA all publish breakdowns of their gold exports to the Chinese mainland, as does the Chinese Special Administrative Territory of Hong Kong. China itself does not publish any gold import or export figures, but it is believed that what comes in does not go out again, gold exports being prohibited. For example our own analysis of known Chinese imports from the countries noted above, plus an estimate for gold from other unpublished sources probably came to around 1,250 tonnes last year. It could actually be more as many of the world's gold mining nations do not publish gold export data. Add into that China's own domestic gold production - estimated at around 450 tonnes last year - and it looks as if China absorbed at least 1,700 tonnes of gold in 2015 - and that's not taking into account any recycled gold. This is hugely above the major consultancies' figures for Chinese gold consumption, but still well short of the SGE withdrawals figure for the year.

Indeed Koos Jansen, perhaps one of the foremost authorities on Chinese gold flows, writing on, came up with the following statement:

"My assessment for China's total net gold import for 2015 is 1,565 tonnes. If we add domestic mine output at 450 tonnes and a few hundred tonnes of scrap supply and disinvestment, total supply has been well over 2,000 tonnes for the year. Not surprisingly SGE withdrawals accounted for 2,596 tonnes in 2015."

So Jansen puts the figure even higher than my own estimate and while neither is as high as SGE withdrawals for the year, they are at least in the same ballpark if we add in, as Jansen has, a figure for recycled scrap supply.

Thus the discrepancies from the specialist consultancies' figures for demand are enormous. But as pointed out above a significant amount of the balance may be gold flowing into the commercial banks. We have seen estimates that these bank holdings are in excess of 3,000 tonnes and, given China's commercial banks are effectively state-controlled we have suggested that these amounts of gold are being held on behalf of the central bank which claims only to hold a little over 1,800 tonnes. China has been notably secretive in announcing its full gold reserve figures and in the past - when it has announced big increases in its reserve figures after years of understating them - it has said it was holding these additions in accounts not reportable to the IMF. Many observers believe China does indeed hold substantially more gold than it says it does, and some, or perhaps all, of this could be one of the reasons for the huge discrepancies between consultancy demand estimates and apparent gold flows. If a nation wishes to maintain secrecy over a substantial part of its gold holdings for whatever reason, what better way to do this than to hold it in the vaults of the commercial banks?

But back to SGE withdrawals - the lower levels seen to date are somewhat confirmed by lower exports direct to the Chinese mainland, and via Hong Kong, from the normal channels. The UK, for example, appears to have exported virtually no gold at all to China this year, while Swiss exports have trended lower. To an extent this may well be that surplus Western gold has been flowing into the gold ETFs, meaning there could be something of a supply squeeze in effect. Even so it appears that supplies to China are currently adequate to meet local demand as shown by price premiums being very low. There is thus little doubt that Chinese demand has indeed slowed sharply, and may slow further, perhaps because the market there is price sensitive and is wary about purchases at the higher prices seen so far this year. We would not be at all surprised if SGE withdrawals for the year continue weak and the year end total comes in at perhaps well below 2,000 tonnes - for the first time since 2012.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.