Why did Basel move from the AMA to the SMA for operational risk capital, and how does the SMA work?

The transition from AMA to SMA represents one of the most significant shifts in Basel III.1 (finalized 2017, implementing through 2028). The AMA allowed banks to use sophisticated internal models for operational risk capital; the SMA replaces this with a standardized formula.

Why the AMA Was Abandoned:

1. Excessive Variability — Two banks with similar risk profiles could produce wildly different capital numbers under AMA because their internal models, assumptions, and data differed dramatically. A BCBS study found AMA capital varied by a factor of 3-5x across comparable banks.

1. Model Risk — AMA models relied on combining internal loss data, external data, scenario analysis, and business environment factors. The complexity created opaque outputs that supervisors couldn't easily validate.

1. Insufficient Data — Operational risk losses are fat-tailed with rare catastrophic events. Most banks had only 10-15 years of internal data — insufficient to reliably model once-in-50-year events.

1. Gaming Potential — Some banks used favorable scenario assumptions or selective external data to lower their AMA capital estimates.

The Standardized Measurement Approach (SMA):

The SMA uses two inputs:

1. Business Indicator Component (BIC) — A proxy for operational risk exposure based on the bank's income statement. It combines:

• Interest, leases, and dividends component (ILDC)
• Services component (SC)
• Financial component (FC)

The BIC captures the idea that larger, more complex banks face more operational risk.

1. Internal Loss Multiplier (ILM) — Adjusts the BIC based on the bank's actual loss history. Banks with higher historical losses (relative to BIC) receive a multiplier above 1.0, increasing their capital requirement.

SMA Capital = BIC × ILM

The BIC Calculation:

The Business Indicator (BI) is calculated from financial statement data, then mapped to capital through a marginal coefficient structure:

Larger banks face progressively higher coefficients, reflecting the non-linear relationship between bank size and operational risk.

Key Difference from AMA:

The SMA removes model discretion. Every bank uses the same formula with the same coefficients. The only bank-specific input is historical loss data (for the ILM), and even this has prescribed calculation rules.

Exam Tip: FRM II tests the rationale for the transition (variability, comparability) and the mechanics of BIC and ILM. Know why larger banks face higher marginal coefficients.

Study SMA mechanics in our FRM Part II community.