What Is the Best Gold ETF to Buy in 2026?
The best gold ETF depends on the role the position plays in a portfolio. For pure cost minimization on a buy-and-hold position, GLDM at 0.10% is the lowest-fee option among major physically-backed funds. For deep liquidity, tight spreads, and the most active options market, GLD remains the benchmark despite its 0.40% expense ratio. For a middle-ground balance between cost and liquidity, IAU at 0.25% is the standard choice.
Investors allocating a small tactical position, say under $5,000, can use any of the six funds on this list without noticing a material difference. Investors committing $50,000 or more to a long-term gold allocation should not hold GLD. The fee differential between GLD and GLDM on a $50,000 position is $150 per year, compounding to roughly $2,500 over a decade assuming 5% annualized returns on gold.
The six ETFs ranked below all hold allocated physical gold in audited vaults. None use derivatives or swaps. The structural differences between them are real but modest. Cost is the primary decision variable.
How Do the Best Gold ETFs Compare?
| Rank | Ticker | Fund | Expense Ratio | AUM (approx.) | Structure | Custodian | Vault |
|---|---|---|---|---|---|---|---|
| 1 | GLDM | SPDR Gold MiniShares | 0.10% | $10B+ | Grantor trust | HSBC | London |
| 2 | SGOL | abrdn Physical Gold | 0.17% | $3.5B | Grantor trust | JPMorgan | Zurich, London |
| 3 | BAR | GraniteShares Gold Trust | 0.17% | $1B+ | Grantor trust | ICBC Standard | London |
| 4 | AAAU | Goldman Sachs Physical Gold | 0.18% | $700M | Grantor trust | JPMorgan | London |
| 5 | IAU | iShares Gold Trust | 0.25% | $32B+ | Grantor trust | JPMorgan | London, NY, Toronto |
| 6 | GLD | SPDR Gold Shares | 0.40% | $60B+ | Grantor trust | HSBC | London |
All six funds are organized as grantor trusts. This structure has two consequences. First, the investor is treated as owning a proportional share of the underlying gold for tax purposes. Second, because gold is a collectible under IRC Section 408(m), long-term capital gains in taxable accounts are taxed at a maximum federal rate of 28%, not the 20% rate applied to equities. This applies to every fund on this list.
1. GLDM: Lowest-Cost Core Holding
SPDR Gold MiniShares (GLDM) launched in June 2018 with a single purpose: offer the same physical gold exposure as GLD at a fraction of the cost. The strategy worked. GLDM’s AUM has grown past $10 billion, making it the fourth-largest gold ETF and a credible alternative to IAU.
Each GLDM share represents approximately 1/100th of an ounce of gold, producing a share price in the $40-50 range at current spot levels. This makes GLDM accessible for small incremental purchases and works well with fractional share functionality at most brokers.
GLDM uses the same sponsor (World Gold Council) and same custodian (HSBC London) as GLD. The gold bars are held in the same vault under the same audit procedures. There is no structural, custodial, or operational difference between GLDM and GLD. The 30 basis point fee difference is pure margin compression.
The one reason not to hold GLDM: active trading. GLDM’s bid-ask spread is slightly wider than GLD, and its options market is thin. For a retail investor buying and holding, this is irrelevant. For an institution running a collar strategy or rolling weekly options, GLD is still the right choice.
Best for: Long-term holders, IRA accounts, cost-sensitive allocations above $1,000.
2. SGOL: Swiss Vaulting at 0.17%
abrdn Physical Gold Shares ETF (SGOL) stores a portion of its gold in Zurich vaults, with additional holdings in London. The Swiss vaulting feature appeals to investors who view Swiss jurisdiction as a hedge against political or regulatory risk in the UK or US.
The 0.17% expense ratio undercuts IAU by 8 basis points and AAAU by 1 basis point. On pure cost, only GLDM beats SGOL. AUM around $3.5 billion is sufficient for tight NAV tracking, though trading volume is lower than GLD or IAU.
The Swiss storage element is often overstated. A London-vaulted fund and a Zurich-vaulted fund face substantially similar custody risks in any realistic scenario. The distinction matters if an investor specifically wants diversification across vault jurisdictions rather than concentrating all gold exposure in one location.
Best for: Investors who want Swiss vault exposure, those diversifying custody across multiple ETFs.
3. BAR: Independent Custodian
GraniteShares Gold Trust (BAR) uses ICBC Standard Bank as custodian, differentiating it from the HSBC/JPMorgan duopoly that dominates the rest of the list. For investors concerned about concentration risk in bullion bank custody, BAR offers a structurally distinct alternative.
At 0.17%, BAR matches SGOL on cost. AUM is smaller at roughly $1 billion, which translates to wider bid-ask spreads and lower daily volume. Tracking error versus the LBMA Gold Price PM fix has been minimal since inception in 2017.
BAR is a niche choice. The fund is operationally sound, priced competitively, and structurally simple, but it does not offer a meaningful advantage over SGOL or GLDM for most investors. Its relevance is specifically for those who want to avoid HSBC and JPMorgan exposure.
Best for: Investors avoiding the major bullion bank custodians, custody diversification strategies.
4. AAAU: Goldman Sachs Sponsored
Goldman Sachs Physical Gold ETF (AAAU) is a small fund at roughly $700 million in AUM, sponsored by Goldman Sachs Asset Management with JPMorgan as custodian. The 0.18% expense ratio is competitive, though AAAU offers no cost or structural advantage over SGOL.
AAAU was originally launched by Perth Mint in 2018 with a government guarantee, then sold to Goldman Sachs in 2022, which removed the Perth Mint features. The fund today is a conventional grantor trust holding London-vaulted gold, similar to most of the category.
For most investors, AAAU is a redundant option. Its main appeal is for Goldman Sachs-affiliated investors who prefer to stay within the GSAM fund family.
Best for: Investors with existing Goldman Sachs brokerage relationships or advisory mandates.
5. IAU: Mid-Range Balance
iShares Gold Trust (IAU) has been the standard gold ETF for cost-conscious investors since 2005. At 0.25%, it sits in the middle of the cost spectrum, above GLDM and SGOL but well below GLD. AUM exceeding $32 billion provides exceptional liquidity with bid-ask spreads typically one cent.
IAU’s share price is approximately 1/100th of an ounce, matching GLDM. This produces a similar share-level accessibility. The fund uses JPMorgan as custodian, with gold held across London, New York, and Toronto vaults.
The argument for IAU over GLDM is marginal: slightly more liquidity, more established trading history, larger fund size. The argument against: 15 basis points of annual fee drag. For a $50,000 position, that is $75 per year forever. Compounded, this becomes meaningful over a 15 to 20 year horizon.
IAU remains a reasonable choice, particularly for investors who want more liquidity than GLDM offers. Newer investors building a position from scratch in 2026 should compare the fee differential against their expected holding period before defaulting to IAU out of habit.
Best for: Investors prioritizing liquidity over cost minimization, mid-sized positions under $25,000 with shorter horizons.
6. GLD: Liquidity Premium
SPDR Gold Shares (GLD) was the first US-listed gold ETF, launching in November 2004. It created the gold ETF category. Two decades later, GLD’s AUM exceeds $60 billion, and it remains the most traded gold ETF by a wide margin.
The 0.40% expense ratio is uncompetitive for buy-and-hold positions. It costs four times what GLDM charges for effectively identical gold exposure. The premium is justified only by liquidity: GLD trades tens of millions of shares daily, its bid-ask spread is consistently the tightest in the category, and its options market is by far the deepest.
For a retail investor holding gold for years or decades, none of these features justify the fee. For an institutional desk trading billions in gold exposure, managing currency hedges, or running structured product mandates, GLD’s liquidity edge can be worth the extra basis points.
Active options traders in particular have no substitute for GLD. Weekly options, deep out-of-the-money strikes, and complex spread strategies require the volume and open interest that only GLD provides among gold ETFs.
Best for: Institutional investors, active traders, options strategies requiring deep liquidity.
When GLDM Wins, When IAU Wins, When GLD Wins
The practical question for most investors is not which fund is theoretically best, but which fund best fits a specific use case.
GLDM wins when the holding period exceeds two years and the position is primarily for wealth preservation or portfolio diversification. Every basis point saved compounds. GLDM’s liquidity is more than adequate for any retail-sized trade and for most institutional trades outside of fast-moving market conditions.
IAU wins when the investor wants modern liquidity without paying GLD’s premium. Mid-sized active accounts, trust accounts with moderate turnover, and investors who value the established brand and 20-year track record all fit the IAU profile. The incremental cost over GLDM is real but often worth the additional trading flexibility.
GLD wins when options trading is part of the strategy, when positions are large enough that one cent of bid-ask spread matters per trade, or when intraday execution quality is paramount. These conditions generally apply to institutional investors, dedicated traders, and sophisticated options users.
Our gold ETF deep dive covers these tradeoffs in more detail, including the mechanics of creation and redemption that underpin why all six funds track gold so closely.
Tracking Error Across the Category
Tracking error measures how closely an ETF’s NAV tracks its benchmark (typically the LBMA Gold Price). For the six funds listed here, annual tracking error above the expense ratio is consistently under 0.10%.
GLD and IAU have the tightest tracking thanks to their size and liquidity. Authorized participants can create and redeem baskets efficiently, keeping NAV aligned with spot gold. GLDM, SGOL, BAR, and AAAU track slightly less precisely due to smaller size, but the deviation is within a few basis points.
Over a 10-year hold, tracking error is typically dominated by the expense ratio. A GLDM position will lose approximately 1% cumulative to fees over a decade. A GLD position will lose approximately 4%. Tracking error adds perhaps 0.5-1% cumulative across a decade. The fund with the lowest expense ratio almost always produces the highest 10-year return in this category.
What Is the Tax Treatment for Gold ETFs?
All six funds are grantor trusts holding physical gold. Long-term capital gains (held over one year) are taxed at a maximum federal rate of 28%, the collectibles rate. Short-term gains are taxed as ordinary income up to 37%.
This 28% ceiling applies only at the top marginal tax bracket. Investors in lower brackets pay less. A taxpayer in the 22% bracket pays 22% on long-term gold ETF gains. The 28% cap is a maximum, not a flat rate.
In a traditional IRA, gains are deferred until withdrawal and taxed at ordinary income rates. In a Roth IRA, gains are fully tax-free after age 59.5 and five years of account existence. Both account types eliminate the collectibles penalty entirely. For investors building long-term gold ETF positions, the Roth IRA is the tax-optimal location when available.
Physical Gold vs the Best ETFs
The cost comparison between a best-in-class gold ETF (GLDM at 0.10%) and physical gold depends heavily on the premium paid at purchase and the cost of storage.
At a 3% premium over spot with free home storage (existing safe, no insurance cost), physical gold becomes cheaper than GLDM after 30 years. At a 5% premium with 0.5% annual depository storage, physical gold never becomes cheaper than GLDM. At a 2% premium with free home storage, the break-even is roughly 20 years.
These numbers reflect pure cost. The case for physical gold rests on attributes beyond cost: no counterparty risk, direct possession, utility during banking disruptions, estate planning flexibility. None of these appear in an expense ratio comparison. The physical vs paper guide addresses the full framework.
For most investors, the rational answer is a blend: ETFs for tactical, tax-advantaged, and accessible exposure. Physical for core long-term positions above a threshold that justifies storage infrastructure, typically $25,000 or more.
Frequently Asked Questions
What is the cheapest gold ETF?
GLDM (SPDR Gold MiniShares) charges 0.10%, the lowest expense ratio among major physically-backed gold ETFs. SGOL and BAR tie for second at 0.17%. GLD remains the most expensive of the major funds at 0.40%, though it trades with the deepest liquidity.
Is GLD or IAU better?
IAU is better for almost any buy-and-hold investor. Its 0.25% expense ratio undercuts GLD’s 0.40% without sacrificing meaningful liquidity for retail-sized trades. GLD retains advantages only in institutional contexts where deep options liquidity and tight spreads on very large trades justify the extra cost.
Do the best gold ETFs all hold physical gold?
All six funds listed here (GLD, IAU, GLDM, SGOL, AAAU, BAR) hold allocated physical gold in audited vaults. None use derivatives, futures, or swaps. Each publishes regular bar lists that investors can verify. This structural uniformity is why the comparison reduces mostly to cost.
Should I hold gold ETFs in a Roth IRA?
Yes, when possible. Gold ETFs held in a Roth IRA avoid the 28% collectibles tax entirely, and all gains become tax-free after the qualifying period. For long-term investors with access to Roth IRA space, gold ETFs are among the most tax-advantaged holdings because the collectibles penalty is so steep in taxable accounts.
Are there any gold ETFs cheaper than GLDM?
Not in the US-listed physically-backed category. GLDM’s 0.10% is the current floor. Some leveraged gold ETFs and futures-based products have different fee structures but are not comparable. Gold mining ETFs also have different expense ratios but provide mining equity exposure, not direct gold exposure.