I just spent an hour or so looking up weekly GLD (Gold ETF) inventory (in tonnes) from Harvey Organ's archives. This post provides the results. I think it could be significant for gold traders / investors.
The above chart gives the price of GLD (and thus of Gold itself) relative to its maximum over the last two years and the size of the GLD inventory relative to its maximum over the same period. Each point represents one week.
Here's some pretty clear observations and a conclusion:
(1) The two are obviously pretty darn correlated.
(2) Both have been falling hard for about a year.
CONCLUSION: It would take something pretty extraordinary to have the price of gold rise significantly without a rise in this GLD inventory. This means that Western gold demand is still pretty significant as far as the price of gold goes.
Now let's look at some price turning points and see whether GLD inventories lead or trail those price changes (gray arrows numbered 1 to 4):
(1) Gold's last price rise prior to the 2013 collapse... The price rise led the inventory rise by about a week. You could even say that the inventory had a little panic sell-out just as the price was turning up.
(2) Gold's last price top prior to the 2013 collapse... Price again led the inventory this time by about 10 weeks.
(3) The summer 2013 lows... Price of gold led by about 4 weeks.
(4) The Syria War Drums Price Top... Price was about coincident with the undersized rise in GLD inventory. I conclude that this signifies that the futures traders pushed the price up, but the mainstream investors (who buy GLD) were not really believers.
CONCLUSION: GLD inventories are a good confirmation indicator of significant price trend changes. The current strong down-trend is pretty bearish for the price of gold.