Fair value hierarchy

Accurate classification of assets and liabilities under IFRS 13's fair value hierarchy is under growing scrutiny, with rising demand for more precise and transparent valuation practices. As techniques evolve and markets shift, companies face pressure from regulators and stakeholders to ensure defensible fair value measurements. Errors in FVH levelling can lead to material misstatements, regulatory challenges, and reputational risk.

To address this, many organisations are turning to trusted advisers like KPMG for expert guidance to ensure alignment with industry best practices. Francisco Jimenez and Ni Zhong, together with other members of the KPMG Financial Instruments team, have delved deeper into the topic and best practices below.

Fair value hierarchy & IFRS 13: A regulatory perspective

IFRS 13 states that "an entity shall use valuation techniques that are appropriate in the circumstances and for which sufficient data are available to measure fair value, maximising the use of relevant observable inputs and minimising the use of unobservable inputs". The inputs used to determine fair value are categorised into three distinct levels:

US Treasury Bills and market-leading equities are prime examples of level 1 securities under IFRS 13. These instruments are traded extremely frequently, with transparent pricing and high liquidity, meeting the definition of having "quoted prices (unadjusted) in active markets" for level 1 inputs.

An example of a level 2 security would be a corporate bond or an interest rate swap. A corporate bond that is not actively traded may be valued using observable market data such as benchmark yield curves and issuer-specific credit spreads.

Similarly, interest rate swaps are typically priced using observable forward rates and discount factors derived from market data. Both instruments qualify as level 2 because they rely on "inputs other than quoted prices included within level 1 that are observable for the asset or liability, either directly or indirectly," as defined by IFRS 13.

Private investments are a common example of level 3 securities, as they are not traded on active public markets and lack observable pricing data for identical or similar instruments. Per IFRS 13, in such cases fair value must be estimated using "unobservable inputs for the asset or liability"—those not derived from market activity but instead based on internal assumptions.

Examples of these inputs might include liquidity discounts and volatility estimates. Similarly, collateralised loan obligations (CLOs) are valued using models that incorporate unobservable assumptions such as conditional prepayment rates (CPR), constant default rates (CDR), and loss given default (LGD).

These assumptions are designed to produce valuation results consistent with the exit price that a market participant would reasonably expect to transact at on the measurement date.

Key challenges and our insights on levelling

What constitutes an "active" market?

Interpreting fair value hierarchy levels under IFRS 13 often involves significant professional judgement, especially in defining what qualifies as an "active" market, which varies by asset type and market transparency. Key indicators include trading volume, trading frequency, exchange listings, and bid/ask spreads. Establishing thresholds for classifying inputs is subjective and requires robust documentation to remain defensible. A key pitfall is the misclassification of level 2 inputs as level 1, or vice versa, often due to misunderstandings around market activity or pricing transparency.

To clarify what qualifies as an "active" market under IFRS 13, firms should adopt a structured framework combining quantitative thresholds, such as trading volume, frequency, bid/ask spreads, and broker quotes, with qualitative factors like pricing reliability and valuation oversight. Consistent application across asset classes and periods, supported by clear documentation and regularly updated thresholds, ensures the framework remains relevant and minimises misclassification risk.

Maximise use of level 1

Firms sometimes overlook or underutilise level 1 inputs, even when they are readily available, which can lead to non-compliance with IFRS 13’s requirement to prioritise level 1 inputs when available. A common example is the consideration of evaluated prices for US Treasury Bills, resulting in a level 2 classification. However, these securities are among the most actively traded globally, with quoted prices in highly active markets. Failure to recognise this can result in misclassification and weakened financial reporting integrity.

Firms may exercise professional judgement to classify a security as level 1 when evaluated pricing has been used, provided there is a reasonable and well-supported justification—such as observing quoted prices in active markets. To ensure consistency and compliance with IFRS 13, these decisions should be well-documented, including rationale and relevant assumptions, and supported by a clear review process.

Regulatory scrutiny

In recent years, regulatory bodies like the IASB and FASB have intensified their focus on fair value measurement and disclosure. Standards such as ASC 820 and IFRS 13 now require more detailed and transparent reporting on the inputs and methodologies used to value financial instruments. Alongside this, there has been a growing emphasis on FVH compliance, with regulators expecting levelling decisions to be well-documented and defensible. These developments reflect a broader trend toward enhanced consistency and transparency in financial reporting.

To address this challenge, firms should ensure clear documentation, apply consistent FVH classification policies, and maintain robust internal review processes. Regular training and management oversight can help ensure practices are aligned with IFRS 13 and ASC 820, supporting transparency and defensibility in fair value reporting.

Assessing the significance and application of valuation inputs

Valuation inputs under IFRS 13 often involve significant complexity and subjectivity, particularly when dealing with unobservable or adjusted inputs. Determining whether an unobservable input is "significant" depends on its impact on the overall fair value conclusion, not merely by its presence. When multiple inputs from different levels are used, the valuation must be classified at the "lowest level input that is significant".

However, judgments around significance can vary across entities, leading to inconsistent classifications. Further complications arise when complex or adjusted inputs, such as internally derived discount rates, bespoke liquidity adjustments, or extrapolated market data—are used without a full understanding of their implications.

These inputs can obscure transparency and increase the risk of misclassification between level 2 and level 3, especially when they have been poorly documented.

To address these challenges, firms should implement a structured framework for evaluating input significance and observability. This includes setting clear thresholds, performing sensitivity analyses, and documenting how input adjustments affect valuation outcomes. Complex or bespoke inputs should be supported by robust documentation and a clear rationale for their use.

Additionally, targeted training for valuation teams can help ensure inputs are well understood and appropriately applied. A standardised process backed by transparent documentation supports the credibility of fair value hierarchy classifications and helps to ensure compliance with IFRS 13.

Complex valuation models

Complex valuation models are sometimes applied without being fully understood, increasing the risk of misclassifying inputs between level 2 and level 3 in the fair value hierarchy—especially when model assumptions lack transparency or documentation.

To mitigate this risk, firms should ensure robust documentation, conduct independent model validations, and provide training to valuation teams. For bespoke models, the rationale and methodology must be clearly supported to ensure credible and defensible fair value classifications.

Robust internal policy

A common challenge in fair value hierarchy classification is the lack of clear, robust internal policies to guide levelling decisions. In many cases, firms either do not have formal documentation or rely on high-level guidance that may lack the depth and clarity necessary for comprehensive understanding.

This can lead to inconsistent application of levelling criteria, especially in borderline cases or when dealing with complex or illiquid instruments. Without well-defined policies, it becomes difficult to ensure transparency and alignment with IFRS 13 and industry best practices.

Firms should aim to standardise their FVH classification policies using tools like decision trees and flowcharts which guide consistent application of processes. These processes must align with regulatory standards and should be independently reviewed to ensure objectivity, compliance, and ongoing improvement.

How KPMG can help

KPMG can assist you through the provision of independent assessments of fair value hierarchy classifications, comparing entity-assigned levels to our own and evaluating the reasonableness of inputs used.
We conduct detailed reviews of firms’ valuation methodologies, identifying any gaps or inconsistencies against IFRS 13 and industry best practices.
Our team monitors changes in asset characteristics, market conditions, and financial reporting standards to ensure fair value conclusions remain current and defensible.
We leverage a broad range of proprietary and third-party data sources to support accurate pricing and enhance the reliability of fair value determinations.
KPMG can provide actionable recommendations to enhance the entities’ fair value hierarchy process and ensure alignment with dynamic regulatory and industry standards.