Gold ETF Flows: March 2021

North American gold ETF funds drove 86% of Q1 net outflows


March highlights

Global gold ETFs lost 107.5t tonnes (t) (-US$5.9bn, -2.9% AUM) in March, marking outflows for the fourth month out of five. This was also the second month in a row in which net outflows ranked the top 10 worst outflows historically. Global assets under management (AUM) stand at 3,574t (US$194.5bn), back to levels last seen in June 2020.1 Since the peak asset levels in November 2020, gold ETF holdings have fallen nearly 9% in tonnage terms, on a par with the approximate loss in the price of gold over the same period.2


ETF flows chart

Data as of

Sources: Bloomberg, Company Filings, ICE Benchmark Administration, World Gold Council; Disclaimer

March regional overview

At a regional level, outflows were once again driven by North American funds of 68.5t (-US$3.8bn, -3.5%), mainly from the largest funds but also from some low-cost gold ETFs which had previously been spared heavier outflows in recent months.3 European funds saw outflows of 45.3t (-US$2.5bn, -2.9%), from funds across multiple countries. Funds in ‘Other’ regions had minor outflows of 0.9t (-US$64mn, -1.8%).4 Asian-listed funds, led by China, had another strong month of inflows of 7.2t (US$427.2mn, +5.8%).

Q1 2021 overview

North American funds represented 86% of global net outflows, falling by 145.4t (-US$8.1bn, -6.7%) in the first quarter of 2021. SPDR® Gold Shares lost US$7.5bn (-10.5%), followed by iShares Gold Trust, which lost US$1.1bn (-3.6%). There were a few bright spots in the region: SPDR® Gold MiniShares saw inflows of US$380mn (9.5%), the largest of any fund globally, followed by newly launched CI Gold Bullion Fund, which added US$369mn.5 European funds lost 51.7t (-US$2.5bn, -2.6%) during the quarter, with UK-listed funds accounting for most of the outflows (39t).

While funds in ‘Other’ regions had small inflows of 2.5%, the major positive trend during the quarter came from Asian-listed funds, which collectively added 17.8t (US$1bn, 14.6%), driven by China, India, and Japan. The strength of investment demand in these markets has been partially linked to opportunistic “dip buying” at a time of gold price weakness, although stock market volatility was the primary driver in China in particular. This was reflected by the local gold-price discount in India and China turning positive during the quarter.6

Price performance and trading volumes

Gold finished the month 3% lower at US$1,691.1/oz.7 It held above US$1,700/oz for most of the month, before falling back below that level in the final few days. By the end of March, gold was down over 10% year-to-date. Despite the downward move in price, gold’s one-month implied volatility – reflecting market expectations of future fluctuations in the price – finished the month close to 14%, well below its historical realised average of 16% and the average level of 20% seen in 2020. Our short-term price performance model suggests that the primary driver of gold’s decline during March, and throughout Q1, were higher interest rates, impacting the opportunity cost of holding gold (see Gold Market Commentary, March 2021).

Daily average trading volumes for gold fell 5% in March to US$162bn from US$168bn in February. While year-to-date volumes are 6% below the 2020 average of US$183bn, they are still robust relative to the 2019 average of US$146bn. The decrease in March was largely a function of lower over-the-counter volumes reported through the LBMA Trade Data. Net long positioning, via the recent Commitment of Traders (COT) report for gold COMEX futures, fell to 490t,8 below the 2020 average net long level of 871t and the lowest level since May 2019.9

Regional flows10

Western funds drove global outflows

  • North American funds had outflows of 68.5t (-US$3.8bn, -3.5% AUM)
  • Holdings in European funds fell by 45.3t (-US$2.5bn, -2.9%)
  • Funds listed in Asia had net inflows of 7.2t (US$427.2mn, +5.8%)
  • Other regions had outflows of 0.9t (-US$63.6mn, -1.8%).

Individual flows

SPDR® Gold Shares (US) represented the lion’s share of global outflows during the first quarter

  • In North America, SPDR® Gold Shares lost 56.1t (-US$3.1bn, -5.0%), followed by iShares Gold Trust, which lost 15.7t (-US$855.2mn, -2.9%). SPDR® Gold MiniShares had holdings rise by 1.0t (US$55.7mn, 1.4%), while Graniteshares Gold Trust lost 0.8t (-US$45.4mn, -4.2%). Recently launched CI Gold Bullion, a low-cost fund in Canada, had inflows of 5.8t (US$318mn) during the month
  • In Europe, four UK-listed funds led outflows. Invesco Physical Gold lost 11.9t (-US$656.8mn, -5.1%), iShares Physical Gold lost 7.6t (-US$446.6mn, -3.4%), Gold Bullion Securities lost 6.7t (-US$373.9mn, -9.2%), and WisdomTree Physical Gold lost 5.7t (-US$313.7mn, -4.7%)
  • In Asia, Bosera Gold added 2.5t (US$ 140.4mn, 11.7%), while E Fund Gold and Huaan Yifu Gold each added 1.5t and 1.3t respectively.

Long-term trends

Global AUM has fallen over 20% since the August 2020 high

  • The downward trend in overall global flows continues to be dominated by US and UK funds
  • Gold ETF flows have been largely correlated with gold prices and interest rates in recent months
  • Asian gold ETF holdings continue to grow assets despite other regions faltering.


  1. We regularly review the global gold-backed ETF universe and adjust the list of funds and holdings based on newly available data and information.

    • Based on the LBMA Gold Price PM as of 31 March 2021.

    • Low-cost US-based gold-backed ETFs are defined as exchange-traded open-ended funds listed in the US, backed by physical gold, with annual management fees of 20bps or less. At present, these include Aberdeen Physical Swiss Gold Shares, SPDR® Gold MiniShares, Graniteshares Gold Trust, and Goldman Sachs Physical Gold ETF.

    • ‘Other’ region includes Australia, South Africa, Turkey, Saudi Arabia, and the United Arab Emirates.

    • The CI Gold Bullion Fund was launched in January 2021.


    • Based on the LBMA Gold Price PM as of 31 March 2021.

    • As of 30 March 2021.

    • Net longs represent Money Manager and Other Net long positioning in the COMEX futures market.

    • We calculate gold-backed ETF flows both in ounces/tonnes of gold and in US dollars because these two metrics are relevant in understanding funds’ performance. The change in tonnes gives a direct measure of how holdings evolve, while the dollar value of flows is a finance-industry standard that gives a perspective on how much investment reaches the funds. There are some months where the reported flows measured in tonnes of gold and their dollar-value equivalent seem inconsistent across regions. Both figures are correct. The disparity is due to the interaction between the performance of the gold price intra-month, the direction and movement of the US dollar, and the timing of the flows. For example, hypothetically, if European funds were to experience outflows early in the month, when the price of gold was low, but gained assets later in the month when the price of gold increased, and/or if the euro/dollar currency rate moved meaningfully when there were flows, there might be a discrepancy between tonnage change and flows.