As anyone working with statistics knows, before analysing data, it is crucial to understand what the numbers mean. And to do that you often need to look at the accompanying notes and definitions - a dull but vitally important part of the job.

If Mr Robertson had done that before writing his article, then I'm sure he would have avoided including numerous inaccuracies, such as:

• We report fabrication and consumption numbers separately in our Gold Demand Trends data so that there is no confusion between the two.
• Our surplus/deficit number is "the difference between total supply and gold demand. Partly a statistical residual, this number also captures demand in the OTC market and changes to inventories on commodity exchanges, with an additional contribution from changes to fabrication inventories".
• Recycling is not fresh supply - it is "gold sourced from fabricated products that have been sold or made ready for sale, which is refined back into bullion".

We strive to be as clear and transparent as possible when reporting developments in the gold market, and I strongly believe that had Mr Robertson read our notes and definitions before writing his article, he would have realised that the arguments he presented in his piece were not robust.

John Reade
Chief Market Strategist and Head of Research
World Gold Council