Issue 99
Digging into the LBMA Precious Metals Trade Data
After almost two years of being able to see the daily turnover figures for trading in the four precious metals, it seems like a good time to look at what we know now and what it means for the market. A major reason for the collection of this data has been to demonstrate to regulators that the gold market is highly liquid relative to other comparable assets and, to me, the data has done this conclusively.
THE TOTAL US DOLLAR VOLUME OF GOLD BEING TRADED EACH DAY IS LARGE RELATIVE TO OTHER COMPARABLE ASSETS AND THIS IS THE VALUE THAT REGULATORS ARE MOST INTERESTED IN
SIZE OF THE MARKETS
The average daily volume in the gold spot market since November 2018 has been around $32 billion, with a record of $72 billion traded in one day in August 2020, and the total across the four gold categories has averaged $51 billion. Silver spot trades a much larger weight on average per day than gold, but is only 15% as large in dollar terms, while platinum and palladium only trade about 2% of gold’s dollar value per day on average. In all the markets, spot transactions dominate, accounting for about 60% of the US dollar volume, as seen here in Figure 1.
The total US dollar volume of gold being traded each day is large relative to other comparable assets, and this is the value that regulators are most interested in with respect to classifying an asset as a High-Quality Liquid Asset (HQLA). To put this in perspective, the total volume traded daily of all UK Gilts, as reported by the UK Debt Management Office, averaged £27 billion ($34 billion approximately) over the same period that we have data for the gold market. And this total covers more than 100 different traded Gilts, with the £27 billion of liquidity divided between these. This means that even on a day with low volume, the spot gold market in London is still more liquid than individual UK Gilts. The total volume of gold traded daily across the four areas reported is roughly equal to the total trading in all government bonds of the EU28(1).
1. www.afme.eu/portals/0/globalassets/downloads/data/government-bonds/ afme-prd-govt-bond-data-report-q2-2018.pdf
LIQUIDITY AND COVID-19
The LBMA Annual Review of 2019 argued that the Loco London and Loco Zurich gold markets were more liquid than government bonds based on the European Bank Authority 2013 report data, especially using the Amihud Illiquidity ratio for spot gold. The Amihud Illiquidity ratio measures the price change caused by each dollar of trading volume – the smaller it is, the more liquid the market is seen as. And this ratio for gold is smaller than for any other market for which I have seen data.
THE BEST DEFINITION OF LIQUIDITY I’VE HEARD IS THAT IT’S SOMETHING YOU HAVE UNTIL YOU NEED IT MOST
This result is driven by the high concentration of gold trading in one contract – spot gold. The various bonds traded in the government bond markets fracture those markets’ liquidity, even though the total volume can be quite similar to spot gold for larger countries. Calculating the Amihud ratio for the other categories of gold trading would give a much larger number, meaning less liquidity, as the volume would be divided, for example, between forwards of different maturities.
The daily turnover in millions of US dollars for the four markets is shown here in Figure 2, broken down into the four categories that are reported. We can see a rising volume for gold and silver over the period, while the PGM volumes are more volatile.
Liquidity is normally defined as something like the impact on the price of making a trade, sometimes trying to include the non-linear component of this where large trades have a bigger than proportionate impact on price. The Amihud ratio used by the European Banking Authority assumes that liquidity is defined as the price change relative to the volume traded. But the best definition of liquidity I’ve heard is that it’s something you have until you need it most.So the Covid-19 induced stock market turmoil that began in late February 2020 gave an opportunity for the first time to see how gold market liquidity would hold up under a period of significant stress – liquidity is no good if it’s not there when you really need it.
And the gold spot market worked well through the whole period, as would be expected from an asset that has been shown to be a safe haven numerous times over the years.
Figure 3 shows the price and volume of the S&P500 index, with the same for spot gold from February to April.
As the stock market began to fall in late February, gold spot volumes rose and remained above average all through March. The average daily volume is about 20 million ounces per day and, in March, the spot market averaged 30 million per day, showing that the market was able to provide lots of liquidity when needed. The day of lowest liquidity, seen early in April, was the Thursday before Easter, which explains the lower amount transacted with the market closed the following day.
LOTS DONE, MORE TO DO
The increasing level of transparency in the gold market will allow
much more research to be undertaken in future on this market’s liquidity by anyone interested in doing so, which should further increase the confidence of investors and regulators in the gold market going forward.
For example, now that data is being published on the gold and silver vault holdings in London, as well as the volume traded, we can calculate the Turnover Ratio for London, another measure used by the EBA to assess the liquidity of assets.
IN SEPTEMBER 2020 BASED ON VAULTING DATA THIS RATIO WAS 0.07 THE SAME MEASURE AS WAS FOUND FOR EQUITIES UNDER THE EBA REVIEW
In September 2020, based on vaulting data, this ratio was 0.07, the same measure as was found for equities under the EBA review. This translates as 7% of the stock of gold in London being traded daily. Coupling these datasets with the Volume in Price Tranche (VIPT) data that is now also available will allow regulators and researchers to assess what the spread of prices was during the days of real market stress in the over-the-counter (OTC) gold market and what that means for gold.
Dr Fergal O’Connor is a lecturer in Finance at Cork University Business School and was previously a lecturer at the University of York. His research on Precious Metals began when he was awarded the 2011 LBMA PhD Bursary for the study of the London Gold Market and he has gone on to publish a range of research on the topic. Currently, his research focuses on building a clear picture of the operation of London Bullion Market from its inception in 1919, as well as a daily price series for gold and silver.