Operational diligence on a new counterparty asks the right questions about the counterparty itself. Credit diligence asks the right questions about the counterparty's balance sheet. Counterparty concentration diligence asks a different question: what does the desk's exposure to this counterparty look like in the moments when several of its counterparties are simultaneously stressed?
Counterparties are not independent. They share LPs, they share clearing relationships, they share funding facilities, they share data-centre footprints, they share regulatory exposure. The desk that diversifies across three nominally-distinct execution counterparties may, in stress, find that all three are running through the same prime broker or the same emergency facility, and that the diversification it bought was nominal.
This piece is a framework for diligence on counterparty concentration as a structural risk — distinct from credit risk on any single name. It applies to institutional desks selecting their execution-counterparty mix in 2026.
The five concentration dimensions to map
Counterparty concentration shows up in five distinct dimensions, each of which the desk should map across its counterparty mix.
The first dimension is LP overlap. If a desk's three execution counterparties all source ~40% of their top-of-book from the same three bank LPs, the desk's effective LP diversification is not three counterparties; it is essentially one panel of three banks accessed through three intermediaries. In a stress window where those three banks pull back simultaneously, the desk has no resilient pricing across any of its three nominal counterparties.
The second dimension is prime-broker overlap. Many institutional execution counterparties hold their inventory and post their margin through one of a small number of prime brokers. If two of a desk's execution counterparties run through the same PB, a PB-side event affects both simultaneously — the desk's exposure to that PB is double-counted.
The third dimension is clearing and settlement-corridor overlap. Counterparties that settle through the same correspondent bank in the same currency are subject to that correspondent's cut-off, liquidity and operational events. A correspondent-bank holiday or operational outage hits multiple counterparties in the same window.
The fourth dimension is data-centre and connectivity overlap. Counterparties whose matching engines run in the same colocation building can be affected by a single building-wide event — power, cooling, network transit, regulatory inspection. The desk's diversification across two counterparties in LD4 is operationally less robust than diversification across LD4 and FR2.
The fifth dimension is regulatory exposure. Counterparties licensed in the same jurisdiction are subject to the same regulator's decisions — capital rules, leverage caps, restricted-jurisdiction lists, periodic audits. Diversifying across jurisdictions reduces single-regulator risk but increases operational complexity.
The diligence questions, per dimension
For each of the five dimensions, the desk should ask the counterparty a specific question and require a specific level of disclosure.
On LP overlap: ask the counterparty to disclose its top-10 LPs by routed volume, broken out by asset class. Most institutional counterparties will not disclose individual LP names but will disclose enough about the panel composition (number of bank LPs, number of non-bank LPs, share of total volume routed to the top-3 LPs) that the desk can model overlap across counterparties at the structural level.
On prime-broker overlap: ask the counterparty to disclose its prime broker(s) for inventory and margin. PB identity is generally not confidential and is a hard requirement for diligence.
On settlement-corridor overlap: ask the counterparty for the correspondent banks it uses in each major currency. Settlement-corridor risk is operationally specific and frequently overlooked.
On data-centre overlap: ask the counterparty for the specific colocation buildings hosting its primary and disaster-recovery matching engines. "Equinix" is not a building; "Equinix LD4" is. The granularity matters.
On regulatory exposure: confirm the counterparty's primary and secondary regulator and any pending regulatory matters disclosed in its most recent compliance review.
What good diversification looks like
A desk that has executed the diligence above and arrived at a counterparty mix with low concentration on every dimension should expect to be using three or four counterparties, each with a distinct PB, materially non-overlapping LP panels, settlement through different correspondent groups, primary engines in different colocation buildings, and regulators in different jurisdictions.
This is harder than it sounds. The number of institutional execution counterparties in 2026 that satisfy all five dimensions, even within a single asset class, is small. Most institutional desks end up with one or two "primary" counterparties whose concentration profile is acceptable, plus one or two "backup" counterparties whose concentration overlaps materially with the primaries but whose value is operational redundancy in case of a primary-counterparty event.
Backup counterparties are still useful for that operational redundancy. They are not useful as structural diversification, and the desk should not treat them as such in its risk reporting.
Where Drovix sits in the framework
Drovix's institutional positioning is as one of the counterparties in a desk's diversified mix, not as the sole counterparty. The structural choice supports this: Drovix maintains a curated LP panel that is materially distinct from the panels operated by the largest US and European institutional brokers; its prime-broker and settlement-corridor relationships are documented and available to counterparties on diligence; its primary engines run in Equinix LD4 with disaster recovery in FR2 and active-active footprint into NY4; its regulator is the FSC of Mauritius, distinct from the regulators of the dominant US and European counterparties.
The structural distinctness is the product feature. A desk that uses Drovix as one of three counterparties — with the other two structured around different LP panels, different PBs, different colocations and different regulators — has materially lower concentration risk on every dimension than a desk that uses three counterparties that all run through the same Tier-1 US bank PB and the same handful of Tier-1 bank LPs.
Drovix publishes the diligence pack required for the framework above on request. The point is to make the structural concentration question answerable in a single meeting, with documented answers, rather than left to the desk's risk officer to inferred from public sources.
The relationship between margin-model pro-cyclicality and concentration in a stress window is in margin pro-cyclicality; the LP-panel side of concentration is in the capacity curve of aggregated liquidity.
Analyst Desk
Drovix Research Desk
Institutional Research
Drovix Research Desk publishes institutional-grade analysis covering macro events, cross-asset correlations, and execution insights for professional market participants.