3.3 Risk Data Aggregation & Reporting

Why Risk Data Is a Risk Management Topic

Models, limits, and capital formulas are only as good as the data feeding them. After the 2007–2009 crisis, supervisors found that many large banks could not quickly aggregate exposures to a single counterparty, geography, or risk factor across legal entities. Risk data aggregation and risk reporting (RDARR)—often discussed in the spirit of the Basel Committee’s BCBS 239 principles for effective risk data aggregation and risk reporting—is therefore part of Foundations of Risk Management, not merely an IT footnote.

FRM candidates are not expected to recite every supervisory paragraph. You are expected to explain why strong aggregation matters, what goes wrong when data are weak, and which governance, architecture, and reporting traits characterize a capable firm.

---

Benefits of Strong RDARR

Strong RDARR shortens the lag between market moves and management action. In a fast crisis—sovereign stress, cyber event, or sudden credit migration—hours matter. Firms that already reconcile front-office positions to risk engines and finance books can re-cut the portfolio by counterparty, product, and risk factor on demand.

Illustrative Decision Cycle

1. Market shock hits a sector.
2. Aggregation engine pulls positions, collateral, and hedges tagged to that sector across all subsidiaries.
3. Risk reports update concentration, P&L sensitivity, and limit utilization.
4. ALCO / risk committee adjusts hedges, limits, or funding within the same day.

Without steps 2–3 working reliably, governance meetings debate stale or conflicting numbers.

---

Impacts of Poor Risk Data

Weak RDARR is not a cosmetic problem; it is an operational and strategic vulnerability.

Worked Qualitative Scenario

Suppose Bank North’s corporate book shows $2.0 billion exposure to Conglomerate X in the credit system, while the markets subsidiary’s securities financing desk holds another $800 million of X collateralized by lower-quality securities that are booked under a different client code. If aggregation keys do not map the two codes to one economic group, firm-wide reports show $2.0 billion. A 30% stress loss assumption applied to the incomplete total understates potential loss by 0.3 × $0.8bn = $240 million—before wrong-way or gap-risk overlays. That is how data taxonomy failures become capital and reputation failures.

---

Governance Principles

BCBS 239–style thinking emphasizes that data capability is a board and senior management responsibility, not an outsourced IT curiosity.

Core Governance Expectations

1. Ownership and accountability — Named executives own risk data quality and reporting integrity; boards receive assurance on capabilities and gaps.
2. Risk data strategy aligned to risk appetite — Aggregation priorities follow material risks (counterparties, products, geographies) rather than only what legacy systems already produce.
3. Policies and standards — Firm-wide definitions, taxonomies, retention rules, and change-control for critical risk data elements.
4. Independent validation and internal audit — Second- and third-line review of aggregation processes, reconciliations, and report production.
5. Resources and expertise — Sufficient skilled staff and budget; critical processes not dependent on a single “hero” analyst.
6. Group-wide scope — Banking, trading, and material subsidiaries included; material outsourcers contractually obligated to meet standards.

Governance fails when each desk invents its own exposure definition, when finance “adjusts” risk numbers without lineage, or when remediation of known data gaps is perpetually deferred as a low-priority IT ticket.

---

Data Architecture and IT Infrastructure

Strong aggregation rests on architecture choices made in calm times.

Architecture Characteristics

• Integrated data taxonomies — Common identifiers for legal entities, products, books, and risk factors.
• Authoritative sources — Clear systems of record for positions, static data, and market data; minimize conflicting copies.
• Automated reconciliation — Front-to-risk, risk-to-finance, and entity-to-group breaks aged and escalated.
• Controls over manual interventions — Dual control, logging, and expiry for overrides.
• Flexibility — Ability to add new products, scenarios, and cut dimensions without multi-year rebuilds.
• Historical depth — Sufficient time series for stress calibration, backtesting, and model development.

IT Capability Traits

A practical FRM framing: technology should let the firm answer “What is our exposure to X, now, on a consistent definition?” without launching a project.

---

Reporting Characteristics

Aggregation feeds risk reporting—the management information that boards, risk committees, and supervisors actually use.

Report Design Tips Aligned with RDARR Spirit

• Lead with exceptions versus appetite and limits, not only raw inventories.
• Show concentrations (top counterparties, sectors, countries) on consistent economic-group definitions.
• Distinguish measurement uncertainty (model and data quality flags) from market volatility.
• Keep crisis playbooks that reuse the same aggregation pipes as BAU reports—do not invent a parallel unaudited process under stress.

Mini Case: Improving a Weak Report Pack

A bank’s weekly board pack lists “trading VaR” from the market-risk engine and “credit exposure” from a loan system, but FX banking-book positions sit only in finance. After a currency crisis, directors ask for total FX sensitivity. The risk function spends five days merging three extracts with inconsistent currency codes. Remediation under RDARR principles would: (1) extend the taxonomy so banking-book FX feeds the same risk warehouse; (2) automate daily reconciliation of FX deltas; (3) add a standing board report cutting FX by entity and product; (4) test the cut in a fire-drill so the five-day scramble never repeats.

---

Connecting RDARR to Other Foundations Topics

• Governance of risk management needs reliable MI; without RDARR, appetite frameworks are decorative.
• ERM (next section) cannot deliver an enterprise view if data remain siloed.
• Stress testing and capital planning inherit every taxonomy and latency weakness in the aggregation layer.

For the exam, emphasize the causal chain: governance → architecture/IT → accurate timely reports → better decisions under stress. Memorize the qualitative benefits and failure modes; be ready to diagnose a vignette where inconsistent identifiers or manual overlays create concentration blind spots.