Why the related-company rule matters more than it looks

The problem hiding in plain sight

A common assumption when constructing a sector index is that each listed company in the relevant sector represents a distinct economic exposure. For most names on most markets, that assumption holds. But it breaks down the moment a parent company and one or more of its listed subsidiaries both qualify under the index’s eligibility rules.

In that case, a naive market-cap-weighted index will count the subsidiary’s economic value twice: once through the parent’s consolidated market cap, and again through the subsidiary’s standalone listing. The result is a hidden concentration risk that doesn’t show up in the published weights.

A concrete example

Take Artgen Biotech (ABIO), a Russian biotechnology issuer with two separately listed subsidiaries — Gemabank (GEMA) and Genetico (GECO). In RHIX v1.0, all three names independently passed the sector eligibility screen and were included in the index.

The hidden exposure: the value of GEMA and GECO is already reflected in ABIO’s parent-level market cap. By including all three, the index gave roughly double weight to that single economic story — and a hit to GEMA or GECO’s price would propagate through the index twice: once through the standalone constituent, and once through ABIO’s consolidated valuation.

The v2.0 fix

RHIX v2.0 introduces an explicit related-company rule:

Where a parent company and one or more of its subsidiaries are separately listed on MOEX and both qualify under the sector criteria, only the parent company (ultimate controlling entity) shall be included.

This is a hard rule, not a soft cap. It is also exempt from the buffer mechanism that ordinarily delays removals by one review cycle — a double-counted exposure should not persist for six additional months just because the rest of eligibility is satisfied.

The trade-off is loss of two index members and slightly more concentration in the remaining five. The benefit is that the index now represents what it claims to represent: five distinct economic exposures, not five constituents containing three overlapping ones.

When this matters in practice

The rule is most material in indices covering markets with significant holding-company structures — Russian, Korean, Italian, Indian — where listed conglomerates frequently carry listed subsidiaries. For users benchmarking active portfolios against the index, the difference between v1.0 and v2.0 weights is large enough to matter for tracking error.

For ETF replication, the rule also reduces operational risk: fewer overlapping cash flows on dividend dates, fewer simultaneous corporate actions to process.

It’s an unglamorous rule that no one will notice on a quiet quarter. It will matter on a noisy one.