Actuarial Notes

Operational Risk Margin

Study guide: Exam 6C

The Operational Risk Margin is the MCT component covering losses from failed internal processes, people and systems, or from external events — fraud, cyber incidents, model error, legal and compliance failures, business disruption. It is not derived from an insurer's own operational loss history; it is a formula applied to volume measures.

\text{Operational risk} \approx f(\text{premiums},\; \text{liabilities},\; \text{growth})

  • Formula-based by design. Operational losses are infrequent, idiosyncratic and poorly captured by any one insurer's data, so the MCT uses a proxy: charges on written premiums and on liabilities, with an additional charge on premium growth above a threshold, subject to an overall cap relative to the rest of Capital Required.
  • The growth charge is the interesting feature. Rapid growth is itself an operational risk indicator — systems, staffing, underwriting discipline and claims handling all strain when volume rises quickly — and it is a recurring precursor in PACICC's analysis of insurer failures. The MCT charges for it explicitly.
  • The margin is added after the Diversification Credit, which applies between insurance and market risk. Operational risk is not diversified against them, on the reasoning that it can strike at any time and often strikes hardest when other things are going wrong.
  • The formula cannot see quality. Two insurers with identical premium and liabilities attract the same charge regardless of the strength of their controls. That gap is what ORSA and Corporate Governance expectations are there to fill: the insurer must form its own view of operational risk and reflect it in the internal target.
  • Model Risk is an operational risk, and OSFI's model risk expectations sit alongside the capital charge — a mis-specified pricing or reserving model is an operational failure with insurance-risk consequences.
  • Cyber risk has grown into the dominant operational exposure for most insurers, and it is the clearest example of a risk the volume-based formula does not measure.

The Growth Charge in Action An insurer's operational risk margin is computed as 2.5\% of written premium plus 0.5\% of liabilities, plus an additional 2.5\% of the amount by which premium growth exceeds 20\%. Written premium rises from \$400 million to \$540 million; liabilities are \$700 million.

Compute the margin and comment.

Answer Growth rate:

\frac{\$540\text{M} - \$400\text{M}}{\$400\text{M}} = 35\%

Premium above the 20\% threshold:

\$540\text{M} - 1.20(\$400\text{M}) = \$540\text{M} - \$480\text{M} = \$60\text{M}

Margin:

\begin{align*}
&= 0.025(\$540\text{M}) + 0.005(\$700\text{M}) + 0.025(\$60\text{M}) \\
&= \$13.5\text{M} + \$3.5\text{M} + \$1.5\text{M} \\
&= \$18.5\text{M}
\end{align*}

The growth surcharge is \$1.5 million — modest in itself, and that is the point worth making: the operational charge is not where rapid growth mainly costs capital. The larger effect is in the Insurance Risk Margin, which rises with the liabilities the new business creates.

What the surcharge signals rather than funds. Its real function is to make growth visible in the capital calculation and to prompt the supervisory question. An insurer growing 35\% should expect OSFI to ask what changed — new distribution, a new line, a competitor's withdrawal, or a price that is below the market — and to look for the PACICC failure pattern of growth plus underpricing plus reserve deficiency.

What the formula cannot tell anyone: whether this insurer's claims systems, underwriting controls and staffing can actually absorb 35\% more business. That assessment belongs in ORSA, and if the answer is no, the internal target should rise by considerably more than \$1.5 million.