ETFs everywhere
In Is anyone NOT yet trading oil? I wrote that I'm long oil via USO call spreads. I kept asking myself: why USO instead of oil futures directly?
That question led to a dozen more. Here's what I learned:
What is USO?
USO (United States Oil Fund) is an exchange-traded product that holds WTI crude oil futures contracts. It doesn't own physical oil. It owns paper claims on future oil delivery, rolled forward every month before they expire. When you buy USO shares, you're buying a fund whose NAV is determined by the value of those futures positions.
What is the relation between USO price and WTI price?
Directionally the same, when WTI rises, USO rises. But the relationship is not 1:1 and the gap compounds over time in a predictable direction.
On any given day, USO's return tracks front-month WTI futures closely, typically 0.85-0.95 correlation. The divergence comes from three sources.
Daily basis risk is small but persistent. Post-2020, USO holds a basket of contracts across multiple months rather than purely front-month. If front-month WTI moves 2% but second and third month contracts move 1.5%, USO moves somewhere between those - a weighted average of its basket. On most days this difference is a few basis points. Over weeks it accumulates.
The roll creates a periodic discrete cost or gain. On the days USO rolls, selling the expiring month and buying the next, it transacts at market prices on both sides. If the curve is in contango (next month more expensive than current), USO sells cheap and buys expensive: negative roll yield, a drag on performance. If the curve is in backwardation as it currently is during the Hormuz crisis, USO sells expensive and buys cheap: positive roll yield, a tailwind. This is the most important variable for anyone holding USO over weeks or months.
Management fees and transaction costs create a small permanent drag - 0.45% annually plus roll transaction costs.
The practical mental model: USO daily return ≈ weighted average return of its futures basket, minus daily accrual of roll cost, minus daily accrual of fees. In the current backwardated oil market, the roll yield is actually positive - USO is a better long oil instrument right now than it would be in a normal contango environment.
How do you compress futures that trade 24/5 into something that trades like US equity?
The ETF wrapper handles this through the creation/redemption mechanism. The fund holds futures continuously, those positions exist and are marked around the clock. But USO shares only trade during US equity market hours.
The gap is managed by authorized participants (APs) - large institutions who can create and redeem USO shares in exchange for cash at NAV. When USO trades at a premium to NAV, APs create new shares by depositing cash, pushing the price back toward NAV. When USO trades at a discount, APs redeem shares for cash. This arbitrage keeps USO's equity price approximately in line with its underlying futures value even though the futures market never closes.
Why Do ETF Wrappers Exist?
Three structural reasons why ETF wrappers for futures markets exist and attract capital.
Accessibility - retail investors and many institutional investors cannot or do not want to access futures markets directly. Futures require approved margin accounts, understanding of delivery mechanics, active roll management, and in some jurisdictions specific regulatory permissions. An ETF holding the same exposure requires none of this - it trades in any standard brokerage account the same way as buying Apple stock.
Smaller notional - a single WTI futures contract represents 1,000 barrels of oil, currently worth over $100,000. A single USO share represents a fraction of that. The ETF wrapper democratizes access to asset classes where the minimum futures contract size excludes smaller investors entirely.
Portfolio integration - institutional investors managing multi-asset portfolios often have mandates, risk systems, and operational infrastructure built around equity and bond instruments. Adding a commodity futures position requires separate custody, different margin treatment, and often separate regulatory reporting. Holding a commodity ETF instead slots into existing portfolio infrastructure without any of those complications. A pension fund whose mandate says "equities and bonds only" can hold GLD as an equity instrument; it cannot hold GC futures.
Continuous trading without delivery risk - futures have expiry dates and, in theory, physical delivery obligations. An investor who forgets to roll a WTI futures position could end up obligated to take delivery of 1,000 barrels of crude oil at Cushing, Oklahoma. The ETF wrapper eliminates this entirely - the fund manager handles rolls and delivery mechanics, and the investor holds a share that never expires.
Are there other examples?
The ETF wrapper pattern has been applied to almost every futures market and financial instrument requiring daily exposure management.
Commodity ETFs: GLD (gold), SLV (silver), UNG (natural gas), PDBC (diversified commodities). Each compresses a futures market into equity-tradeable shares with the same roll mechanics and tracking error dynamics as USO.
Leveraged ETFs: TQQQ (3x QQQ), UPRO (3x SPX). These hold the underlying plus swap agreements to achieve leverage and must rebalance daily to maintain their stated ratio - a mechanical flow with significant market impact on volatile days.
Inverse ETFs: SH (-1x SPX), SDS (-2x SPX). Hold short swap positions with the same daily rebalancing requirement.
Volatility ETFs: VXX, UVXY. Hold VIX futures and roll them daily. The compression of a volatility futures curve into a single equity-tradeable product creates persistent roll drag that suppresses near-term VIX futures prices - a structural feature rather than a bug.
The common thread: any time you compress a futures curve or leveraged position into a single equity ticker, you create rebalancing or roll mechanics that behave differently than the underlying.
Is USO equity or commodity?
Both, depending on the lens. Legally it is an equity, a Delaware statutory trust whose shares trade on NYSE Arca under securities regulations. Economically it is commodity exposure, its returns are driven entirely by oil prices. Regulatorily it sits between the two - the SEC regulates the ETF structure while the CFTC regulates the underlying futures USO holds.
In Book Review: The Complete Guide to Capital Markets for Quantitative Professionals I asked questions about ETFs versus futures that I couldn't answer at the time. Now I can:
Why are some indices available as both ETFs and futures while others are only ETFs?
Futures require standardization and sufficient institutional demand to justify an exchange listing a contract. SPX futures exist because the S&P 500 is the global benchmark for institutional equity exposure - pension funds, hedge funds, and banks need futures for hedging at scale. FTSE All-World futures don't exist in liquid form because the index is too broad and too fragmented across jurisdictions to standardize into a single deliverable contract. Most institutions manage international exposure through individual country or regional futures (Eurostoxx, Nikkei, FTSE 100) rather than a single all-world product.
ETFs fill the gap for indices where futures don't exist - the creation/redemption mechanism substitutes for the futures delivery mechanism. The ETF wrapper is essentially infinitely flexible in what it can hold, as long as the underlying assets have sufficient liquidity and the SEC approves the structure.
From a legal perspective, is it easier to create a new ETF than a new futures contract?
Yes, significantly. A new futures contract requires exchange approval, CFTC designation, standardized contract specifications the entire market agrees on, market maker commitment to provide liquidity, and demonstrated economic purpose. The exchange bears the economic risk of a contract that doesn't attract volume - most new futures contracts fail and are delisted within years.
A new ETF requires SEC registration, an investment advisor with appropriate registrations, a custodian, an authorized participant agreement, and an exchange listing application. The process is well-understood, template-driven, and typically takes 3-9 months. Hundreds of law firms specialize in ETF launches.
The result: over 3,000 ETFs exist in the US versus a handful of actively traded futures complexes. When a fund manager wants exposure to an obscure commodity basket, they create an ETF holding existing futures contracts rather than convincing an exchange to list new ones.
What's the difference between investing with futures versus ETFs?
Five meaningful differences.
Leverage - futures require only margin (typically 5-10% of notional). ETFs require full capital upfront.
Roll control - with futures you decide when and how to roll, bearing the cost consciously. With ETFs the roll happens mechanically on a fixed schedule regardless of curve shape.
Expiry - futures expire and require active management. ETFs don't.
Tax treatment - futures in the US receive 60/40 treatment (60% long-term, 40% short-term regardless of holding period). ETF gains are taxed on actual holding period.
Ecosystem - ETFs have options chains, retail analytics, and equity infrastructure. Futures have deeper liquidity, tighter spreads in the underlying, and professional-grade data.
Should I implement my strategy using ETFs or futures?
Depends on three conditions: capital base, data infrastructure, and holding period.
ETFs are the right answer when your capital is smaller than 5-10x the futures contract notional, when you're already operating inside equity options infrastructure, and when the tracking error and roll costs are smaller than the friction cost of building new futures infrastructure. For a directional trade held weeks to months, USO options slot directly into existing tools without rebuilding anything.
Futures become the right answer when your capital is large enough that contract size is appropriate, when you've built futures-specific analytics, and when roll timing decisions add enough value to justify the complexity. At that point the cleaner exposure, tighter spreads, and roll control of CL outweigh USO's accessibility advantages.
The transition from ETF to futures is a natural progression as capital and infrastructure grow - not a binary choice made once.
For now, I'm sticking with USO. My capital is small enough that futures contract size is awkward. My infrastructure is equity-focused. And the current backwardation makes USO's roll yield work in my favor rather than against it.