VFV vs XUS vs ZSP: The MER Is a Tie, So Tracking and Trading Decide

Quick answer

VFV is the narrow winner in this specific cost audit, not because its management expense ratio is lower—it is not—but because it combines the tightest disclosed historical average spread, the largest asset base, and the highest five-year ending value. The more important answer is that VFV, XUS, and ZSP are economically close enough that trading costs, brokerage features, and taxes can outweigh the observed difference.

All three issuers report a 0.09% MER.^[2][5][8] That costs about C$9 annually per C$10,000 before compounding. A fee-only comparison therefore produces a three-way tie. The decision starts to become useful only after adding realized returns, tracking, spreads, and fund structure.

Five-year all-in outcome

The following table uses each issuer's NAV or portfolio total return for the five years ended June 30, 2026. It assumes C$10,000 was invested at NAV, distributions were reinvested, and no commission or personal tax applied.

XUS publishes a five-year cumulative return of 111.22%, which converts to roughly 16.13% annualized.^[4] VFV reported 16.15% and ZSP 16.14% annualized.^[2][8] The result is almost comically tight: C$17.57 separated the highest and lowest ending balances after five years. That is 0.18% of the original investment in total, not per year.

This table does not say VFV will lead next. It says an investor should be skeptical of claims that one of these wrappers is dramatically cheaper. A C$10 commission on both entry and exit would exceed the full historical gap. Selling an existing holding to chase a few basis points can also create tax or market-timing costs.

Tracking is useful—and easy to misuse

VFV's five-year NAV return was 16.15% against 16.25% for its displayed benchmark, a lag of about 0.10 percentage point per year.^[2] ZSP reported 16.14% against 16.53%, a larger 0.39-point annual gap.^[8] XUS reported 111.22% cumulative versus 110.30% for its benchmark, an apparent 0.92-point cumulative lead.^[4]

That makes XUS look like the tracking champion, but it is not safe to subtract these rows as though all issuers used an identical benchmark calculation. Index tax treatment, valuation timing, currency translation, and linked series can vary. The funds also have different structures. The honest conclusion is that XUS tracked its own displayed benchmark very well, VFV's lag was close to its fee, and ZSP's reported gap deserves monitoring. It is not proof that XUS will outperform the other two.

The spread and liquidity test

VFV's July ETF Facts reported a 0.018% average bid-ask spread and 793,279 average daily units for the 12 months ended May 31.^[3] XUS reported 0.03% and 736,152 units for the year ended April 30.^[6] ZSP reported 0.02% and 1,146,487 units during 2025.^[9]

On C$10,000, those average one-way spreads represent roughly C$1.80, C$3.00, and C$2.00 if applied mechanically. They are historical averages across different periods, not executable quotes. A volatile open, a U.S. market holiday, or trading while the underlying U.S. stocks are closed can widen a Canadian S&P 500 ETF's spread. A limit order during overlapping Canadian and U.S. market hours is more important than a tiny ranking difference in an old average.

ZSP's higher unit volume does not automatically make it the most liquid. ETF liquidity also comes from the underlying S&P 500 securities and market makers' creation and redemption activity. Still, all three disclosures indicate active, tight markets rather than a material liquidity warning.

Structure: VOO, IVV, or direct holdings

VFV obtains most of its exposure through Vanguard's U.S.-listed VOO, while XUS primarily uses BlackRock's IVV.^[2][6][7] ZSP directly holds the index constituents.^[8] A fund-of-fund structure can make portfolio operations efficient, but it adds an underlying fund layer. The issuers cap or report fees so the published Canadian-wrapper MER remains the relevant headline cost.

Direct replication does not guarantee better tracking. It changes where trading, tax, cash drag, and securities-lending effects occur. XUS's 2025 report said the Canadian wrapper did not lend securities.^[7] Vanguard's Canadian disclosure similarly did not provide a comparable lending-income offset for VFV, while activity may occur in the underlying U.S. fund. The reviewed ZSP materials did not expose a comparable current lending contribution. With no consistent three-fund dataset, assigning imaginary basis-point benefits would reduce rather than improve the analysis.

Distribution timing differs but is not a return advantage by itself. VFV and ZSP pay quarterly; XUS pays semi-annually.^[2][5][8] An investor who needs frequent cash may prefer quarterly payments. An investor who reinvests automatically should focus on total return and any brokerage reinvestment constraints.

What would change the result

VFV's edge rests on measurements that can move. XUS could retain better benchmark-relative tracking. ZSP's direct replication could close its reported gap. Live spreads can reverse the historical order on any given trade. Fee waivers can also change, and foreign withholding or taxable distributions depend on account and fund structure.

The strongest reason not to switch is the tiny observed gap. In a taxable account, realizing a capital gain can overwhelm years of basis-point differences. In any account, being out of the market during a transfer can matter more than C$18 over five years. This comparison is most useful for a new contribution or for monitoring an existing fund—not as a sell signal.

Bottom line

VFV earns the highest score because the available evidence is consistently good: a 0.09% MER, the narrowest disclosed average spread, very large scale, and the highest ending value in the common-date five-year test.^[1][2][3] XUS is nearly tied and has the best issuer-reported tracking result against its own benchmark. ZSP offers direct replication and strong disclosed trading volume.

Choose based on the live quote, brokerage costs, distribution preference, and account consequences. If one of the three is already held efficiently, the data does not support paying tax or commissions merely to chase this ranking.