Is VGT or FTEC the Better Tech ETF? Here's How They Compare on Risk, Returns, and Fees

Fidelity's FTEC and Vanguard's VGT are near-identical passive U.S. information-technology sector ETFs, with FTEC charging a 0.08% expense ratio versus VGT's 0.09% and offering 288 vs 322 holdings. Key metrics are comparable: 1‑year returns ~21.66%, beta ~1.33, max drawdowns ≈ -35%, and dividend yields ~0.40%–0.41%; AUM and liquidity diverge materially at $16.7bn for FTEC versus $130bn for VGT. VGT is slightly more concentrated in Nvidia (18.19% vs 16.61%) and its top three holdings account for 45.41% of assets versus 44.34% for FTEC, making cost and liquidity the primary differentiators for allocators.

Analysis

Market structure: The practical winners are mega-cap tech stocks (NVDA, AAPL, MSFT) and primary-market makers — VGT’s $130B AUM vs FTEC’s $16.7B concentrates buying power into the same top names (NVDA 18.19% vs 16.61%), mechanically increasing demand for those shares and tightening spreads. Losers are small-cap techs and niche active managers who lose flow; for institutional-sized orders (> $1–5M) VGT’s liquidity and narrower spreads materially reduce trading cost vs FTEC. Cross-asset: concentrated ETF flows into NVDA raise equity option gamma and can push short-dated implied vol higher; risk-on into tech tends to steepen U.S. front-end yields and pressure USD carry in short windows around AI/data catalysts.

Risk assessment: Key tail risks are (1) regulatory actions on AI/chips or export controls within 3–12 months, (2) an operational shock at NVDA (foundry/yield or guidance miss) and (3) ETF redemptions that force outsized selling of top caps. Immediate (days) risk = liquidity/market impact for large trades; short-term (weeks–months) = earnings and AI cycle news driving +/-20–30% moves in NVDA; long-term (years) = concentration risk where 3 names could represent ~45% of returns. Hidden dependencies include creation/redemption mechanics and options market hedging that can amplify moves (delta-hedging into selloffs).

Trade implications: For core exposure, prefer VGT for institutional-sized positions (> $500k) because liquidity saves slippage; for retail, FTEC is marginally cheaper. Tactical: establish a capped NVDA exposure via options — buy a 6–12 month call spread sized to 0.5–1.5% of portfolio (e.g., 30–60% OTM) to express AI upside while limiting drawdown. Hedging: buy puts on VGT or pay a 1–3% cost to protect a tech sleeve if it exceeds 20–25% of total equity exposure; pair trade — overweight VGT vs SPY (long VGT, short SPY) sized 2–4% to express tech outperformance.

Contrarian angles: The market underestimates operational liquidity risk if NVDA gaps (a 20% NVDA shock could cause >5% intraday tracking error in VGT), so the “liquidity premium” of VGT is overstated for stressed scenarios. Expense-driven switching to FTEC is likely underdone for retail but irrelevant for large blocks; historical parallel: 2018 FAANG concentration where index crowding amplified drawdowns. Unintended consequence: crowding into two ETFs raises systemic gamma exposure — a catalyst like an earnings miss could cascade through options and ETF creation/redemption channels.

Key Decisions for Investors

• For institutional-sized core tech exposure (> $500k): allocate 3–6% of portfolio to VGT (ticker VGT) rather than FTEC to minimize slippage; cap incremental NVDA exposure so NVDA constitutes no more than 3% of total portfolio (rebalance if VGT NVDA weight >18.5% or NVDA position >3%).
• Tactical asymmetric upside on NVDA: buy a 6–12 month call spread sized to 0.5–1.5% of portfolio (e.g., long 40–60% OTM call, short 100–150% OTM) to express AI tail upside while limiting capital at risk; close if NVDA rallies >50% or falls >25% from entry.
• Protect concentrated tech sleeves: purchase 3–6 month puts on VGT equal to 2–3% of portfolio value (or a 5–10% downside floor) if tech exposure exceeds 20–25% of equities, or execute a collar on existing VGT holdings to reduce cost of protection.
• Relative/value pair: implement a 2–4% long VGT / short SPY position to capture expected tech premium over the next 3–9 months around AI adoption; trim if VGT outperformance exceeds 10% absolute or if implied vol for NVDA spikes >80%.