By Rhona O’Connelll, Head of Market Analysis, EMEA & Asia, StoneX Financial Ltd | June 10, 2021
Welcome to our monthly analysis of LBMA trading volumes for the major precious metals.
As usual, there are some interesting patterns to explore. Volumes in gold and silver have continued to dwindle, as did platinum in spot and forwards, but a few exceptionally high-volume days saw Options and LoanLeaseDeposit volumes boost those platinum total volume monthly averages. Palladium volumes increased again, and here, too, options and LoanLeaseDeposit showed striking gains.
As ever, heavy volumes were almost always followed by changes in trend, or at least a correction, as sentiment changes to some positions are opened while others are closed. Note also that the CME lowered margins on 6th May for gold, silver and platinum futures.
Daily average trading volumes in May, compared with the average for the first four months of the year.
Gold spot trade volumes averaged 16.2M ounces daily during May, a drop of 22% against the average for the first four months of the year. In the main, volumes revolved around a range of between 12M and 19M ounces, but in the first week of the month, there were two consecutive days (6th and 7th, see chart below) of much higher volume, at 19.3M and 20.3M ounces in the spot market, and with the first day also posting high Forward and Option volumes as gold gained over 1% in price. This came as gold came back up through the $1,800 level and crossed above the 20-day moving average to take it to ten-week highs (and starting a bull run that lasted through the month), so some of this would certainly have been a combination of sentiment, technical action and momentum trades.
In the background, the U.S. dollar and U.S. bond yields were coming under pressure. This ties in with the fact that at this point, in what has become a recurring and dominant theme, Federal Reserve officials were out in force in what looked like a concerted effort to dampen inflationary fears and to make the point that a small amount of inflation is good for an economy – and also to underline the expectations that these forces would be transitory. On May 6, at least four senior Fed officials spoke publicly about this, including former hawks such as Loretta Mester from Cleveland. Backing all this was the clear intention on the part of the Fed to keep rates lower for [much] longer, which of course focuses attention on the fact that real rates (almost worldwide, with the notable exception of China) will remain below zero for a good while to come.
Conditions settled down thereafter until the 24th and 25th of the month when swap/forward volumes surged to over 12M ounces on each day as spot gold approached $1,900, perhaps suggesting hedging activity.
Finally, spot volumes rocketed on the last trading day of the month (28th) to just shy of 31M ounces, against an average for the rest of May of 15.4M ounces.
The spot trading range was narrow (this is also axiomatic – wider ranges are frequently accompanied by thin liquidity and vice versa) between $1,882 and $1,905 as gold traded upwards through $1,900 for the second time in the week. The key driver here was the mixed bag of U.S. economic data, the most eye-catching of which was the 3.6% year-on-year gain in the Personal Consumption Expenditure, which is the inflation measure that the Fed watches when framing policy. The core index was up 3.1%, the fastest pace since 1992 – although, of course, the heavy dislocation of April 2020 makes the reading somewhat distorted. These numbers were accompanied by a drop in the University of Michigan Consumer Sentiment Index to 82.9 from 88.3 in April, with the expectations index falling below 80 as consumers started to fret about inflation. Professional sentiment, though, became more buoyant.
While the chart does not run into June, the high volumes of 6th/7th and 28th May again preceded a change in direction, although gold didn’t turn down again until June 1 as the market ran out of momentum above $1,900.
There was, as usual, some similarity between silver and gold trading patterns, but not much overall when it comes to looking at outliers. Both enjoyed the early spike in spot volumes, but while gold’s spot spike lasted two days, silver’s lasted only one (6th May), along with high forward volumes. Silver’s options activity expanded on 6th May, as did that of gold, but then soared on the 7th and was still well above the monthly average on the 8th.
The trigger for silver’s lurch higher on 6th May, during which it posted spot volumes of 318M ounces (monthly average 210M ounces) and traded a 5% range, was, of course, the move in gold. As usual, silver was the outperformer (it is almost invariable, but not every time that silver outperforms in a bull run and underperforms in a bear) and the move on the 6th took it well into overbought territory. This move will almost certainly have triggered some forward selling, most probably from across the mining sector, as base metal producers must have been tempted to lock in silver’s by-product value at that stage. We can see this also in elevated volumes in the LoanLeaseDeposit volumes, which actually peaked, in this early part of the month, on the 11th, as the price recovered in mid-correction. They then soared on the last day of May to post 15.8M ounces.
Silver’s overall price action differed from gold in that gold’s bull run was almost unchecked over the whole month, while silver lost momentum in mid-May. Sure enough, the second and third highest volumes of the month came on the 17th and 18th, as the price peaked at $28.75 intraday, the highest since the Reddit-related action at end-January/early February. While gold pushed on, silver traced out a more or less horizontal pattern over the rest of the month in a narrow intraday range of $27.20-28.60, after a heavy volume catfight on the 19th and 20th, which were a down-day and an -up-day respectively as the market to some extent cleared itself out.
Over this latter part of the month, silver’s correlation with gold was slipping (but bear in mind that correlation does not equate to direction), while that with copper, though lower, was steady. This would tend to suggest that the markets overall were not convinced of gold’s further bull potential because when the markets are in that frame of mind, silver often decouples and goes into base metal mode; during this period, the trends of silver and copper did show similarities as some of the heat came out of the copper market.
Some clear peaks and troughs in platinum trading during May, the majority of which again presaged a change in direction or a correction. The initial bias was downward, with two notable dips, on the 7th and 12th before volumes started to trend higher through towards month-end. As noted above, spot volumes were down, registering falls of 26% against the first four months of the year, while forwards were down by 23%. Options, however, were 36% higher and LoanLeaseDeposit was up by 21%.
That said, the overall trend in the spot market was a gradual increase in volumes, apart from a clear spike downward; this was on the 12th, with a volume of 296,932 ounces, just 54% of the monthly average as a whole; this was the third day of a downturn that has persisted through to the time of writing in early June; it was also the day on which both the 10-day and 20-day moving averages were severed, and the inference is that participants stood back in order to see whether there would be a rebound. Clearly, there wasn’t, and volumes picked up again thereafter, with the downturn consistently guided by the 10-day moving average, which capped any uptick. The increase in options activity may well reflect the comparatively low volatility in the metal’s price, allowing for some value to be picked up.
The end of the month saw platinum try to regain the $1,200 level, but this was only short-lived, and the failure may well have been responsible for an increase in volume in spot, forwards and LoanLeaseDeposit on the final day as market participants hedged against further likely price falls.
May saw London Platinum Week, during which (17th May) Johnson Matthey released its PGM market report and Metals Focus (18th May) released its Platinum & Palladium Focus. Both reports projected a surplus of platinum this year as the mine disruptions of 2020 were worked off; NorNickel, later in the month, reinforced this, looking at a surplus of platinum in both 2021 and 2022. After platinum’s year-long price recovery from a pretty vicious seven-year bear market, these authoritative studies almost certainly helped to take some heat out of the market., although both houses pointed to the prospect of substantial increases in platinum loadings, especially in China on heavy-duty diesel as emissions limits continue to tighten. Another key development is the onset of some partial substitution for palladium with platinum in the appropriate autocatalysts as a cost-containment exercise.
Palladium volumes continued to increase, with small gains in spot and forwards (4.6% and 4.1%) respectively against the first four months of the year; Options were ahead by 28.3%, and LoanLeaseDeposit was up by a sizable 44.7%, including four massive outliers, two at the start of the month and two at the end.
The first trading day, immediately after the Easter break, was the heaviest, at over 750,000 ounces, compared with a monthly average of just over 355,000 ounces.
This was the point at which the price peaked for the year-to-date, heralding the end of the bull run that had started in late January and starting a month of an overall downtrend that has persisted into early June. The trigger was the third failure at the $3,000 level, which prompted liquidation, while action in the forwards and LoanLeaseDeposit in the following few days would suggest that suppliers may have taken some forward cover. There was some bargain hunting on the decline towards $2,800 mid-month, which was reflected in a swelling in spot and forward volumes (and also, to a small degree, in the Exchange-traded products), but this time the technical resistance on the charts prevented an assault on $2,950.
This failure, on 19th May, saw another upsurge in volume as the price reversed course. Volumes accordingly dwindled thereafter, allowing prices to drift lower once more.
It does look as if palladium enjoyed substantial industrially related business during May. The stale bull liquidation and other activity at the start of the month were followed by increased buying in mid-month that then fizzled out. We had a very strong flourish at month-end with two days of spot in excess of 650,000 ounces each day, forward volumes rising by 50% over the levels of the previous week, and options and forwards surging to generate another run towards $2,900. It looks as if industrial users, notably the auto sector, which has been living hand-to-mouth, may have taken the view that the $2,800 level may be a floor and been replenishing inventory accordingly, despite the semiconductor shortage that is hampering auto production as the communications industry is further up the queue for chip supplies. The auto sector is estimated by independent consultants to account for only 11% of global semiconductor supplies.
Platinum week saw Johnson Matthey and Metals Focus both pointing to the recovery in the auto sector as a key component in a widening deficit this year, along with the flooding (now all repaired) at two Norilsk mines; NorNickel itself concurred at end-month, but for the time being the $3,000 level has proved too tough to penetrate.