Americans Buying Gold

Adam Hamilton, CPA

American stock investors have suddenly started buying gold again.  After largely ignoring most of gold’s monster upleg, they just started flocking back fueling big builds in major-gold-ETF holdings.  If this stunning shift proves a persistent trend, these capital inflows will drive gold much higher.  American stock investors remain radically underinvested in gold, commanding vast pools of capital they can deploy to chase it.

Normally I don’t write about major-gold-ETF capital flows very often, maybe a few times a year.  But just two weeks ago, I published an essay “Americans to Chase Gold”.  It analyzed the incredible anomaly of gold soaring in an extraordinary monster upleg with virtually no differential gold-ETF-share buying!  I led off with “American stock investors’ overdue return is the single-most-bullish argument for gold today.”

It concluded “American stock investors will almost certainly yet chase gold’s extraordinary monster upleg.”  While that seemed inevitable, the timing of market normalizations is unpredictable.  So I didn’t expect to expand this research thread again for a few months, and had an entirely-different essay planned.  Then out of the blue in this past week or so, gold-ETF capital inflows exploded to their highest levels in years!

Some context helps illuminate this major development.  The world’s dominant gold ETFs are GLD SPDR Gold Shares and IAU iShares Gold Trust.  They pioneered these effective conduits enabling stock-market capital to easily migrate into and out of physical gold bullion, launching way back in November 2004 and January 2005.  The World Gold Council tracks all the world’s physically-backed gold ETFs every quarter.

Exiting Q4, GLD and IAU together represented fully 39.3% of all the gold bullion held by all the world’s physically-backed gold ETFs!  A UK one was a distant-third at just 6.2%.  During plenty of quarters, the holdings swings in GLD and IAU alone account for a large majority of overall total world gold-demand moves.  All that has made GLD+IAU capital flows one of gold’s primary drivers for about a decade-and-a-half.

Gold uplegs reach monster status once they achieve 40%+ gains with no interrupting 10% corrections.  And gold uplegs simply didn’t grow that large in this modern gold-ETF era without being fueled by big GLD+IAU-holdings builds.  Without American stock-market capital flowing in, there just wasn’t sufficient persistent buying to fuel huge uplegs.  Understanding the simple mechanics behind gold ETFs explains why.

Physically-backed ones essentially act like conduits for the vast pools of stock-market capital to slosh into and out of gold.  GLD and IAU shares have their own supplies and demands independent from gold’s own.  So if they are being bought or sold at rates different than gold’s, their share prices will decouple from its and fail their tracking mission.  Gold ETFs require constant active management to equalize any differentials.

When American stock investors buy GLD+IAU shares faster than gold, their share prices soon threaten to disconnect to the upside.  To prevent this, managers issue sufficient new shares to offset that excess demand.  The resulting cash proceeds are then immediately plowed into buying more physical gold bullion.  So rising GLD+IAU holdings reveal American stock-market capital flowing into gold, bidding up its global price.

Monster gold uplegs are quite rare, and before today’s the last two both crested in 2020.  The first peaked in early March at 42.7% gains over 18.8 months.  That was mostly fueled by GLD+IAU holdings soaring 30.4% or 314.2 metric tons on massive differential gold-ETF-share buying!  Then the pandemic-lockdown stock panic erupted, briefly sucking in gold.  That epic fear forced a 12.1% correction in just eight trading days.

Corrections reset uplegs, so gold’s next one soon started soaring in that panic’s wake.  Over the next 4.6 months into early August, gold rocketed up 40.0%.  American stock investors overwhelmingly drove that, with GLD+IAU holdings blasting up 35.3% or 460.5t!  Average these last two monsters, and gold soared 41.4% on 32.9% or 387.4t GLD+IAU-holdings builds.  Gold uplegs could no longer grow huge without them.

Love or hate GLD and IAU, they are frictionless gold vehicles for stock investors.  Though they do have small annual percent-of-assets management fees of 0.40% and 0.25%, their shares are easy, quick, and cheap to buy.  That gives instant portfolio exposure to gold price moves.  Before gold ETFs, investors had to sell stocks, transfer cash, buy gold coins in stores or online for 5%+ commissions, then securely store them.

Later selling was similarly-involved, with dealers buying coins back around 2% discounts.  So a two-way gold “trade” cost investors a 7%-to-10% spread in addition to being cumbersome!  It is also a hassle to find time to visit a coin shop or post office to pay for insured shipping.  So with gold-ETF shares effortless and almost free to trade, it’s easy to understand why GLD and IAU grew so popular with American stock investors.

Yet just a couple weeks ago, one of gold’s mightiest uplegs ever was blasting higher with no meaningful differential GLD+IAU-share buying.  As of February 13th, gold had powered an incredible 60.9% higher in 16.3 months without any 10%+ corrections.  Yet shockingly GLD+IAU holdings had slumped 1.6% or 20.0t in this span!  That’s an extreme disconnect, a shocking anomaly unprecedented in this gold-ETF era.

This updated chart shows the crazy lack of differential GLD+IAU-share buying as gold soared.  My last essay on this a couple weeks ago explained why.  Enthralled by the AI stock bubble, American stock investors felt little need to diversify their mega-cap-tech-dominated portfolios.  So central banks, Chinese investors, and Indian jewelry buyers took the gold-buying baton to fuel gold’s latest monster upleg.

Even as gold started breaking out to new nominal record closes a year ago, American stock investors didn’t care.  They were too busy piling into NVIDIA on hopes of hundreds of billions of dollars of spending on AI infrastructure.  On March 8th, 2024 as gold achieved a six-trading-day record streak, unbelievably GLD+IAU holdings slumped to a crazy 4.5-year secular low of 1,203.3t!  Then they ground along there for months.

Capital inflows into gold-ETF shares finally started to resume as gold surged to more records last autumn leading into the elections.  But since GLD+IAU holdings were digging out of such a deep hole, they were still flat through gold’s newly-monster upleg.  In late October they crested at 1,273.5t, still a bit under their 1,277.9t in early October 2023 when this gold upleg was born.  By late January 2025, they fell back to 1,248.9t.

While gold’s bull run was awesome largely thanks to very-strong central-bank demand, it sure felt like American stock investors would never return.  But they had to sooner or later, probably after that AI stock bubble inevitably burst.  They love chasing winners, piling in to ride established upside momentum.  And it was only a matter of time until either record-achieving gold prices or slumping mega-cap techs brought them back.

GLD is much larger than IAU, usually commanding over 2/3rds of their aggregate gold-bullion holdings.  Older and better-known, GLD tends to be preferred by retail investors.  But since IAU’s annual fees are 15 basis points lower than GLD’s, institutional investors tend to gravitate to it.  Just last week as gold forged deeper into nominal-record territory, big differential gold-ETF-share buying erupted in both ETFs.

Last Tuesday the 18th, GLD’s holdings surged 0.8% in a sizable daily build.  That was just the biggest in one month, but more significant than that due to the timing.  Gold stocks have seriously underperformed gold in this monster upleg and recent years, leaving them anomalously-undervalued.  Partially because of that and exacerbating it, plenty of traders have decided to instead trade GLD options for leveraged gold exposure.

They’ve exploded in popularity and volume in recent years.  Universally across stock markets, all monthly options expire on each month’s third Friday.  So for some years now, GLD has periodically seen big daily builds on third Fridays.  Those almost certainly involved options dealers squaring hedging positions, as those third-Friday GLD builds were isolated.  They were single-day events with little or no follow-up buying.

Excluding third Fridays, last Tuesday’s 0.8% GLD-holdings build was its biggest since late June 2024.  I saw that in real-time as I watch these holdings every trading day, but didn’t think anything of it.  But then the next day Wednesday the 19th, GLD’s holdings surged another 0.9%.  GLD hadn’t seen similar back-to-back big-build days ex-third-Fridays since early March 2022, which sure piqued my interest more.

Then on Thursday the 20th, GLD’s holdings surged another 0.7% while IAU’s climbed 0.8%!  The latter was IAU’s biggest daily build since mid-November as gold’s post-election pullback was bottoming.  And GLD’s holdings had soared 2.4% in three trading days.  Outside of occasional three-day spans including third Fridays, nothing like this had happened since mid-April 2020 early in gold’s last monster upleg!

So for the first time in gold’s current monster upleg, American stock investors were sustaining differential GLD-share buying.  Were they finally starting to return to begin chasing gold?  Remember just before that GLD+IAU holdings were down 20.0t during this monster upleg, compared to those enormous 387.4t average builds in gold’s last two monster uplegs.  A similar 400t+ swing soon would propel gold much higher.

That brings us to last Friday the 21st, which was February’s options-expiration day.  GLD’s holdings skyrocketed 2.3% or 20.7t that day alone!  We probably should dismiss that as being options-squaring-related, as there’s plenty of precedent of that during this gold upleg.  GLD’s holdings surged 1.8% on October 2023’s third Friday, 1.5% on November 2023’s, 1.8% on March 2024’s, and 1.9% on December 2024’s.

Those isolated big third-Friday builds continued this year, with GLD’s holdings surging 1.2% on January’s.  Had this past Friday’s huge 2.3% GLD-holdings build been the only one, I would’ve assumed it was just another isolated third-Friday thing.  But with that big three-day build leading into it and its enormous size, it might be more significant.  Last Friday’s GLD-holdings build ranks as one of the largest ever witnessed!

In percentage terms we’re talking top-0.8% in GLD’s entire history since November 2004, 5,102 trading days.  But big daily percentage builds were naturally much more common in GLD’s early years when its holdings were way smaller.  Excluding pre-2007 big percentage builds, last Friday’s ranks in the top-0.4%!  And in raw-tonnage terms, it is even higher at top-0.4% absolutely and top-0.3% excluding pre-2007 ones.

So while options squaring was likely a large part of Friday’s massive build, it was still exceptional even by third-Friday standards.  GLD hadn’t seen a bigger single-day build on any day including third Fridays since late January 2022, a third Friday.  Across those four trading days last week, GLD’s huge 4.8% holdings surge was the biggest since late March 2020 just emerging from that pandemic-lockdown stock panic!

American stock investors were just resuming chasing gold then, piling in to ride its upside momentum.  Again that culminated in an epic 35.3% or 460.5t GLD+IAU-holdings build in 4.6 months, catapulting gold 40.0% higher in that short span!  I wouldn’t bet on another 40% in coming months with gold’s monster upleg already so remarkable, but American stock investors returning would drive gold considerably higher.

Making this GLD-build streak stand out more, it actually continued this Monday the 24th with another 0.3%.  That wasn’t much, but most big third-Friday GLD builds have no follow-up buying at all.  So seeing even modest levels after such a gigantic GLD-holdings surge feels different.  Scroll back up to that chart and look at the giant jump in GLD+IAU holdings over this past week or so, it sticks out like a sore thumb!

Only time will tell if this stunning differential gold-ETF-share buying soon fizzles or proves the start of a sustained trend.  But we haven’t seen anything like it in gold’s entire current monster upleg!  The last time American stock investors flooded into GLD+IAU shares so hard, it was early in gold’s previous monster upleg and thus a super-bullish omen.  Speculators and investors interested in gold need to closely watch this.

American investors not only ignored the vast majority of gold’s current upleg, but their deployment in gold may as well be zero.  A quick proxy for that divides the total value of GLD+IAU’s physical-gold-bullion holdings by the collective market capitalization of all the elite S&P 500 stocks.  Midweek with gold just under nominal-record highs, GLD+IAU holdings were worth $122.8b.  That’s definitely a nice chunk of change.

Yet this known gold held by American stock investors is radically dwarfed by their stock holdings, worth $53,611.4b in S&P 500 stocks alone.  That implies a trivial 0.2% gold portfolio allocation!  That’s way too low, and will rise as this AI stock bubble inevitably bursts.  Merely to return to 1.0% which is still nothing, American stock investors would have to shift hundreds of billions of dollars into gold likely via these ETFs.

Midweek GLD+IAU held 1,308.8t of physical gold bullion in trust for their shareholders.  The all-time high was 1,800.5t in mid-October 2020 soon after gold’s last monster upleg.  Even that left gold allocations under 0.4% of the S&P 500’s aggregate market cap.  Mean reverting back up there would require a huge GLD+IAU build of 491.7t.  That really ought to happen given the vast pools of stock-market capital today.

So if you’re not watching GLD+IAU holdings, you need to.  They’re analyzed in all our popular weekly and monthly newsletters.  We also actively trade smaller fundamentally-superior mid-tier and junior gold stocks in them, which do still amplify gold.  Our 84 gold-stock trades last year averaged +43.1% annualized realized gains!  That far-outperformed the leading GDX gold-stock ETF’s dreadful +9.4% in 2024.

Our newsletter trading books are currently full of recently-added smaller gold-stock trades with fantastic upside potential as gold stocks revalue higher to reflect these awesome prevailing gold prices.  The gold miners are now reporting their best quarterly results ever with record earnings, and GDX is on the verge of a dozen-year secular breakout!  Both should greatly increase interest in and capital inflows into this sector.

The bottom line is American stock investors look to be finally returning to gold.  The combined physical-gold-bullion holdings of the dominant gold ETFs just soared their most in years, reflecting major capital inflows.  While monthly options-expiration position squaring played a role, the holdings builds went way beyond that.  For the first time in gold’s current monster upleg, Americans might be starting to chase it.

And they have vast room to buy, with implied portfolio allocations to gold of effectively-zero.  The single-most-bullish argument for gold after such an extraordinary run is Americans haven’t yet returned en masse.  But when they almost certainly take back the lead-buying baton from central banks, gold will power much higher yet on their vast capital inflows.  Deeply-undervalued gold stocks ought to soar on that.

Adam Hamilton, CPA

February 28, 2025
Copyright 2000 – 2025 Zeal LLC (www.ZealLLC.com)