More asset managers are now running the same strategy through two wrappers: a legacy separately managed account or composite, and a newer ETF built to track the same mandate. The logic is sound. ETFs bring in a different investor base, offer better liquidity, and lower the entry threshold for allocators who cannot hold SMAs directly. But the moment both vehicles exist side by side, someone will pull up the monthly returns, notice a gap, and ask why.
The honest answer is that a gap is expected. Two vehicles running "the same" strategy are rarely running the identical portfolio at the identical moment, and the differences compound in ways that are easy to explain individually but hard to summarize in a single sentence. The real work is not eliminating the gap. It is being able to decompose it on demand, in a format an allocator or consultant can follow without a finance degree in ETF mechanics.
An ETF and its underlying model portfolio diverge for structural reasons that have nothing to do with skill or intent. The ETF is constrained by creation and redemption mechanics, authorized participant activity, and index-replication rules that force it to hold instruments in specific lot sizes. The composite, by contrast, is usually managed with more discretion: a portfolio manager can enter or exit a position gradually, hold slightly different weights across accounts, or delay a rebalance by a day if liquidity conditions warrant it. Neither approach is wrong. They are simply different operating constraints applied to the same investment thesis.
When a client or allocator asks "why did the ETF return 40 basis points less than the composite this quarter," a vague answer erodes trust even if the underlying management was sound. A structured answer, broken into four buckets, tends to hold up under scrutiny.
Portfolio construction constraints. ETFs often carry cash drag from creation unit mechanics, minimum trade lots that round position sizes, and index-tracking rules that limit deviation from a reference basket. Composites are freer to hold precise target weights. This bucket usually explains a meaningful chunk of the gap in periods of high turnover.
Factor and exposure drift. Over months, a rules-based ETF and a discretionarily managed composite can develop different sector, size, or factor tilts even when they started from the same model. Attribution here means comparing exposures at each rebalance point, not just at inception, since drift accumulates gradually and is easy to miss in a single snapshot.
Execution timing. The ETF trades through authorized participants at prices set by the creation/redemption process; the composite trades directly, at times chosen by the manager or trading desk. A one- or two-day lag in when a position is initiated or trimmed, multiplied across a volatile month, can account for a surprising share of the divergence.
Fee structure. Expense ratios and management fees are rarely identical between an ETF and its SMA counterpart, and the difference is mechanical rather than a judgment call. It is also the easiest bucket to explain, which is why it should not be the only one offered when a client asks the question.
The managers who handle this well do not wait for the due diligence call. They build the attribution into a recurring report, monthly or quarterly depending on the audience, and walk through the same four buckets every cycle regardless of whether the gap is large or small that period. This does two things. It normalizes the existence of a gap, so a wider-than-usual month does not read as a red flag. And it gives the client service team a consistent vocabulary to use with allocators, rather than reconstructing an explanation from scratch each time someone asks.
None of this works without daily transaction-level data for both vehicles, consistent inception dates so period comparisons are apples to apples, and benchmarks that are genuinely aligned rather than approximately similar. It also requires the ability to compare vehicles in something close to real time, since a client asking about last month's divergence during a due diligence call is not well served by a report that takes two weeks to produce.
A well-prepared answer in a due diligence meeting sounds specific: "the ETF underperformed the composite by 35 basis points this quarter, of which roughly 15 came from cash drag during two large creation events, 10 from a lag in the technology sector rebalance, 5 from fee differential, and the remainder from execution timing on the two largest trades." That level of specificity, delivered without hesitation, tells an allocator more about a manager's operational maturity than the return figure itself.
Kiski's reporting is built to support exactly this kind of cross-vehicle attribution, pulling daily transaction data across ETFs and composites into a single comparable framework so the breakdown above can be produced on a normal reporting cadence rather than assembled under deadline pressure.
