How to Own Commodities: Futures Roll Costs, K-1 vs 1099 Funds, and Gold ETF Taxes
Commodity futures roll costs with a worked example, K-1 versus 1099 funds (DBC, PDBC, USO), gold ETF fees, and the 28% collectibles rate, as of September 2026.

A commodity ETF's return has three parts: the change in commodity prices, the gain or loss from rolling futures contracts, and the interest earned on the cash that backs the futures. The first gets the attention, and the second can dominate the result, as the example below shows. From 2008 through 2025, the Invesco DB Commodity Index Tracking Fund (DBC) returned -0.86% a year, while consumer prices rose 2.44% a year and the S&P 500 returned about 11% a year (data below). The 1959 to 2004 record was very different, and this guide looks at the mechanics that may explain the gap.
This guide covers how to own commodities: futures funds and their roll costs, why K-1 and 1099 funds differ, what the April 2020 oil collapse showed about single-commodity funds, and how gold ETFs are priced and taxed. For how commodities behave in high-inflation years, see the inflation hedging guide and the stagflation guide.
The record: what commodity funds have earned
Gary Gorton and Geert Rouwenhorst built an equally weighted index of commodity futures from July 1959 to March 2004 and found that fully collateralized futures earned about the same return and Sharpe ratio as equities, were negatively correlated with both stock and bond returns, and were positively correlated with inflation (NBER Working Paper 10595). That study is the basis of most arguments for commodities in a portfolio.
Recent evidence comes from DBC, a large broad commodity fund that reports annual returns in its SEC filings. We compiled its yearly returns at net asset value from its 10-K filings (2010, 2013, 2016, 2019, 2022, 2025) and compared it with the S&P 500 and gold in Aswath Damodaran's return dataset and with December-to-December CPI from FRED.
| Period | DBC, per year | S&P 500, per year | Gold, per year | CPI, per year |
|---|---|---|---|---|
| 2008 to 2025 (18 years) | -0.86% | +10.96% | +9.60% | 2.44% |
| 2011 to 2019 (9 years) | -5.60% | +13.29% | +0.89% | 1.78% |
| 2020 to 2025 (6 years) | +8.31% | +14.93% | +19.07% | 3.94% |
| 2021 and 2022 | +30.07% | +2.61% | -1.62% | 6.75% |
Some highlights from DBC's annual returns: -30.83% in 2008, +41.34% in 2021, +19.69% in 2022, -6.18% in 2023, +2.00% in 2024, and +8.41% in 2025. Over the 18 years, the fund's annual returns had a standard deviation of 18.6 percentage points, similar to the S&P 500's 18.0, and a 0.42 correlation with the S&P 500. It beat inflation in 9 of the 18 years. In the two big inflation years, 2021 and 2022, it returned $10,000 into $16,917. In 2023 with inflation at 3.4%, it lost 6.18%.
Why does a fund built on the same idea fall short of the 1959 to 2004 record? Roll costs, fees, and the timing of the study are all candidates, and this comparison cannot separate them. It does show that a broad commodity fund is a volatile hedge that paid off in some inflation shocks and lost money in others. The next section explains one of the main costs.
How roll yield works, with numbers
A futures fund cannot hold physical crude oil, so it holds contracts that expire, sells them before delivery, and buys later ones. The price difference between the two contracts is the roll yield. DBC's 10-K describes its index as selecting, for each commodity, the contract with the most favorable implied roll yield, and it states that rolling in contango tends to drag on returns (DBC 10-K).
Illustration: crude oil sits at $80.00 for a full year. Each month the next contract trades 1% above the expiring one, so the fund sells at $80.00 and buys at $80.80. A month later the contract it bought has converged to the $80.00 spot price, and the fund has lost $0.80, or 0.99% of what it paid. The fund's cash earns 3.90% a year (SOFR, FRED, used as a proxy).
| Scenario for one year | Roll yield | Fund total return |
|---|---|---|
| Spot flat, contango of 1% a month | -11.26% | -7.73% |
| Spot flat, backwardation of 1% a month | +12.82% | +17.30% |
| Spot up 10%, contango of 1% a month | -11.26% | +1.50% |
| Spot down 10%, backwardation of 1% a month | +12.82% | +5.57% |
| Spot up 10%, no roll effect | 0% | +14.37% |
In the first row $10,000 shrinks to $9,227, even though the price of oil never changed. A fund in a persistent contango needs spot prices to rise more than 11% a year just to break even before interest. The most useful check on a commodity fund is therefore its roll rule. Ask how it chooses which contracts to hold, whether it can move down the curve to reduce contango, and what it did in the last two years the curve was steep.
What April 2020 showed about oil funds
The CFTC's staff report on the WTI crude oil contract says the May 2020 contract fell from $17.73 a barrel to settle at -$37.63 on April 20, the day before it expired, the first negative price in the contract's 37 years on the exchange (CFTC). That was a drop of $55.36 in one session.
The United States Oil Fund (USO) holds the near-month WTI contract and rolls it to the second month before expiry, according to its 2025 annual report. It charges a 0.45% management fee on assets, plus brokerage and other costs. A single-commodity fund like this depends on one market and on the front of its futures curve, and its roll cost comes back every month. It suits a short trade better than a long-term holding.
K-1 or 1099: what the fund structure means for taxes
| DBC | PDBC | USO | GLD | |
|---|---|---|---|---|
| Exposure | Diversified commodity futures, contracts chosen for roll yield | Diversified commodity futures, held through a Cayman subsidiary | Near-month WTI crude futures | Physical gold bars |
| Legal form | Partnership | Regulated investment company | Partnership | Grantor trust |
| Fee | 0.85% management fee | 0.59% unitary fee | 0.45% management fee plus other expenses | 0.40% sponsor fee |
| Tax form | Schedule K-1 | Form 1099 | Schedule K-1 | Form 1099 |
| Gain treatment | Futures usually 60% long term, 40% short term | Distributions reported on Form 1099; character not confirmed | Futures usually 60% long term, 40% short term | Collectibles, 28% maximum |
DBC's 10-K says the fund is classified as a partnership for federal income tax purposes, and holders are liable for tax on their allocable share of income. Regulated futures contracts are generally Section 1256 contracts, and IRS Publication 550 describes their treatment: 60% of a gain or loss is long term and 40% is short term, whatever the holding period, and open positions are marked to market at year end. You pay tax on gains the fund realizes each year, even if you never sell shares.
Illustration: a $10,000 futures gain at a 24% ordinary bracket and a 15% long-term rate is taxed at (60% x 15%) + (40% x 24%) = 18.6%. At a 37% ordinary bracket and a 20% long-term rate it is 26.8%. Both are lower than the 28% ceiling for gold funds, and neither includes the 3.8% net investment income tax (IRS).
PDBC works differently. Its semiannual report to April 30, 2026 says the fund must qualify as a regulated investment company by getting at least 90% of its income from qualifying sources, and that its Cayman subsidiary is a controlled foreign corporation whose income the fund adds to its own taxable income. The report gives the unitary advisory fee as 0.59% of average daily net assets and shows $6.34 billion of net assets and a net asset value of $18.46 at April 30, 2026. Invesco also waived about $3.9 million of a $15.4 million fee bill for the six months. You avoid the K-1, and like any fund of this type it distributes income to shareholders. We could not confirm from the filing how those distributions are characterized, so check the fund's most recent tax notes before assuming they get the 60/40 treatment.
The main practical difference: a K-1 fund arrives late in tax season, can require state returns, and taxes unrealized gains at year end through the mark-to-market rule. A 1099 fund is simpler but can distribute gains you must report. If you want to owe tax only when you sell, neither structure gives you that.
Gold: ETFs, fees, and the 28% rate
Physical gold trusts hold bars in a vault, and the price of a share follows the price of gold minus the annual fee. As of September 29, 2026, the fees listed on the issuers' pages were:
| Fund | Annual fee | Source |
|---|---|---|
| SPDR Gold Shares (GLD) | 0.40% | State Street |
| iShares Gold Trust (IAU) | 0.25% | iShares |
| SPDR Gold MiniShares (GLDM) | 0.10% | State Street |
Illustration: $50,000 invested for 10 years while the gold price rises 5% a year. With no fee it would be worth $81,445. After fees: GLD $78,245, IAU $79,431, GLDM $80,634. The gap between GLD and GLDM is $2,389. That gap is the price of the same metal in a more expensive wrapper.
The tax rule is the larger issue. The GLD annual report states that gains individuals realize from selling collectibles, including gold bullion, held more than one year are taxed at a maximum 28%, and that a gain on selling shares in a trust that holds collectibles is treated the same way. The IRS says the same for collectibles generally. The 28% is a cap: if your ordinary rate is lower, that rate applies. Shares sold within a year are taxed as ordinary income.
Illustration: a $20,000 long-term gain costs $5,600 at 28%, against $3,000 at the 15% stock rate or $4,000 at 20%. With the 3.8% net investment income tax the gold figure is $6,360, against $3,760 or $4,760. Held inside an IRA or 401(k), gold ETF shares are taxed under the account's withdrawal rules and not at the collectibles rate. Gold mining stocks are ordinary stocks for tax purposes but behave like stocks, so they add equity risk and do not track the metal closely.
Decision rules
- Choose the purpose first. A broad commodity fund is a volatile inflation hedge with a 0.42 correlation to stocks in the DBC data. If you want long-term return, the 2008 to 2025 record argues for a stock fund.
- Read the roll rule before the expense ratio. Contango cost 11.3% a year in our example, more than any fee.
- Compare the tax form with your situation. K-1 funds can lower the tax rate on gains at high brackets because of the 60/40 rule, and cost paperwork and mark-to-market tax. A 1099 fund is simpler.
- For gold, use the lowest-fee physical fund and hold it longer than one year. Expect 28% at high incomes; the tax-efficient investing guide covers account placement, and the tax-loss harvesting guide covers losses.
- Keep the position small enough to hold through a decade like 2011 to 2019, when DBC lost 40%. A sleeve of 5% of a portfolio that loses 40% costs 2%. The rebalancing guide explains how to add to it on a schedule.
- Avoid single-commodity funds for long holds. Their roll cost repeats monthly and their risk is concentrated.
- Funds that combine commodities with bonds under one wrapper are covered in the risk parity guide, and the wider menu is in the alternative investments guide.


This guide is for informational purposes only and does not constitute investment, tax, or legal advice. Fund fees and prices are as of September 2026 from issuer pages and SEC filings and can change. Historical returns come from the sources linked, and we computed the comparisons; past performance does not predict future results. Commodity funds can lose money, and the tax treatment of futures funds depends on your circumstances. Consult a qualified tax professional and financial advisor before investing.



