Models & quantification

Risk quantification

Risk quantification is the expression of risk in numbers, such as an expected loss, a loss distribution, a probability or a capital requirement, so that risks can be compared with each other and with the risk appetite. It contrasts with qualitative assessment using categories such as low, medium and high; semi-quantitative assessment with scores and scales sits in between.

Source: NIST SP 800-30 Rev. 1 (2012), section 2.3.2; IEC 31010:2019 (Annex B.5 and B.7); ECB Guide to the ICAAP (July 2026), principle 6; DNB Open Book on Supervision, confidence interval in the ICAAP (2016); Directive 2009/138/EC (Solvency II) Article 101; FSB, Principles for an Effective Risk Appetite Framework (2013)

NIST SP 800-30 Rev. 1 (2012) distinguishes three approaches. A quantitative assessment uses numbers whose meaning is maintained outside the context of the assessment; it best supports cost-benefit analysis of alternative responses, but its results need explanation of the assumptions, and its rigour is significantly reduced when subjective judgements are buried in the numbers. A qualitative assessment with categories from very low to very high is easy to communicate, but makes relative ranking difficult. A semi-quantitative assessment with scores or bins, such as a scale of 1 to 10, combines both, with a visible role for expert judgement.

In the financial sector, quantification is often a supervisory requirement. One of the seven principles of the ECB Guide to the ICAAP (revised in July 2026) is that risk quantification methodologies are adequate, consistent and independently validated. The Dutch Central Bank (De Nederlandsche Bank, DNB), which carries out the SREP for less significant banks, assesses capital adequacy in the ICAAP, as established practice, at a confidence interval of 99.9 per cent over one year (DNB, 2016). Insurers calibrate the Solvency Capital Requirement to all quantifiable risks (Article 101 of Directive 2009/138/EC, Solvency II).

For the board and supervisory board members, the question is not whether a figure is exact but whether it improves decisions: which assumptions drive the result, how sensitive is it to them, and have the models been validated? Quantification is not the same as risk assessment: assessment covers identifying, analysing and evaluating risk, and quantification is one way of carrying out the analysis. Not every risk can be captured well in money; the Financial Stability Board explicitly names reputation, conduct and money laundering risks as more difficult to quantify (2013). Through Risk Insight, ONE Risk Advisory uses Enterprise, Process and IT Risk Assessments to map the risks and opportunities that most affect an organisation's objectives, prioritised by impact and likelihood.

Last updated: · Editor: ONE Risk Advisory

Related service Risk Insight

Frequently asked questions

Is a quantitative risk assessment always better than a qualitative one?

No. NIST SP 800-30 (2012) points out that quantification mainly helps with cost-benefit analysis, but loses rigour when subjective judgements are buried in the numbers or when uncertainty is high. A qualitative assessment is easier to communicate. The choice depends on the data available, the decision to be taken and the extent to which the risk can be captured in numbers.

Which techniques are used to quantify risk?

IEC 31010:2019 describes, among others, Monte Carlo simulation, Bayesian analysis and event tree and fault tree analysis for determining consequences and likelihood, and Value at Risk and expected shortfall as measures of risk. In the financial sector these are joined by internal models for credit, market and insurance risk, which require the supervisor's permission and are independently validated.

← All terms