Fair Value Hierarchy Explained

The illusion of precision in modern valuation

Finance, at its best, is the art of measuring the immeasurable. In theory, the idea of fair value should be simple. It is the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. But in practice, especially in private markets, there is often no observable transaction, no real-time quote, and no “market participant” willing to buy the derivative or complex financial instrument embedded in a startup’s balance sheet.

The fair value hierarchy under ASC 820 was designed to bring structure to that ambiguity. It classifies valuation inputs into three levels of reliability — Level 1, Level 2, and Level 3 — each reflecting how observable and market-based the inputs are. While this hierarchy has brought more consistency to financial reporting, it has also introduced a new kind of risk: the false sense of certainty that can accompany complex instruments valued on fragile assumptions.

Over the years, I have seen this firsthand. From embedded derivatives buried in convertible notes to stock warrants linked to performance milestones, the valuation work often began with spreadsheets but ended in negotiation. Valuation, in such cases, is not merely a measurement exercise. It is a dialogue among management, auditors, and investors about what the asset or liability truly means.

The scaffolding of ASC 820

ASC 820 defines fair value as an exit price — what the company would receive or pay to transfer the asset or liability in question. It does not depend on how the asset was acquired or how the liability was originated. Rather, it assumes a hypothetical transaction in a hypothetical market, among hypothetical participants.

The standard introduces a three-level hierarchy to prioritize the inputs used in fair value measurement. Level 1 inputs are the most reliable. They are quoted prices in active markets for identical assets or liabilities. Think of publicly traded stocks, Treasury bills, or exchange-listed derivatives. If you can find a Bloomberg terminal or a trading app to show you the price, you are looking at Level 1.

Level 2 inputs are still observable, but not quite as direct. They may include quoted prices for similar assets, observable yield curves, or implied volatilities from option markets. Corporate bonds, interest rate swaps, or lightly traded securities often fall into this category. They require interpolation and adjustment, but they are still anchored in market activity.

Level 3 inputs are the most judgment-heavy. They are unobservable and based on internal models, assumptions, or estimates. Private equity investments, illiquid warrants, complex derivatives, and contingent liabilities all tend to fall into Level 3. These instruments do not have a clear market price. Instead, the company builds a valuation model based on cash flow projections, discount rates, and probabilities — all of which can be hard to audit and harder to validate.

Why it matters to the startup CFO

The fair value hierarchy might sound academic, but its implications are deeply operational. For high-growth and venture-backed companies, many of the instruments issued in the pursuit of capital — warrants, SAFEs, convertible notes with embedded derivatives — require periodic remeasurement at fair value. That remeasurement affects not only the balance sheet but also the income statement, often through non-cash gains or losses that distort bottom-line results.

In one Series B company I supported, the issuance of performance-linked warrants to a strategic investor triggered a quarterly fair value exercise. The warrants had no public analog and were contingent on several operational KPIs. They fell squarely into Level 3. Valuation required a Monte Carlo simulation with assumptions about volatility, exit timing, and milestone probability. The resulting liability swung from $2 million to $5.2 million in two quarters — not because anything had changed fundamentally, but because volatility inputs had shifted and exit timing had shortened. The income statement reflected a $3.2 million loss. Investors were confused. The board asked whether the company was actually losing money or simply revaluing a future dilution. Both were true, in different senses.

The lesson was not about mechanics. It was about narrative. Fair value accounting requires companies to explain their judgments, not just calculate them.

Level 1: The luxury of clarity

Level 1 inputs are rare in early-stage companies, but where they exist, they are a gift. Publicly traded securities, FX balances, and crypto assets held on balance sheet can all be valued easily using market data. The audit trail is clean. The pricing is reproducible. And the reporting impact is minimal.

That said, even Level 1 assets can raise questions. If the market is active but volatile, mark-to-market swings may create noise. CFOs must determine whether to present such changes above the line or adjust for them in non-GAAP presentations. For companies with investment portfolios, treasury assets, or exposure to public equity through strategic holdings, these decisions matter.

Level 2: The realm of judgment

Level 2 valuations live in the space between confidence and interpretation. For example, if a startup holds corporate bonds or has entered into interest rate swaps, the inputs used to value those instruments — such as yield spreads or swap curves — are observable, but not directly quoted. Third-party pricing services are often used. In theory, these services provide consistency. In practice, they rely on market models that require monitoring.

I recall a late-stage fintech platform with several interest rate hedges designed to protect against Federal Reserve movements. The instruments were categorized as Level 2. But when interest rate volatility spiked in early 2022, the hedges swung in value, and the pricing service models diverged. Two services produced values with a 20 percent difference. The auditors asked why management had chosen one over the other. The answer — ease of integration — did not satisfy. We had to justify the selection with a documented evaluation of pricing methodology.

This episode reminded me that Level 2 does not mean “low risk.” It means “observable, but still reliant on modeling.” The diligence must match the complexity.

Level 3: The danger and opportunity of discretion

Level 3 instruments are where financial reporting becomes most subjective. In venture-backed companies, the most common Level 3 instruments include warrants, SAFEs, convertible instruments with embedded derivatives, earnout provisions, and contingent consideration.

These require custom valuation models. Black-Scholes or Monte Carlo simulations are standard for warrants. Discounted cash flow models are common for private equity. Scenario analyses with weighted probabilities are used for contingent liabilities.

The risks here are threefold. First, model inputs such as volatility, exit timing, or revenue growth projections can be highly sensitive. A small change can produce a large revaluation. Second, the models are often managed by external valuation firms, which may not fully understand the operational nuances behind the assumptions. Third, the results can meaningfully affect financial statements, especially when remeasurement is required each reporting period.

The opportunity, however, is equally real. Companies that invest in robust, transparent valuation processes can build trust with auditors, investors, and internal stakeholders. They can also use the exercise to sharpen their strategic thinking. A well-constructed Monte Carlo simulation for performance-linked instruments can serve not only the accountants, but also the board in scenario planning.

In one company I supported, we built a Level 3 valuation model that mapped three product launch scenarios and their corresponding impact on earnout liabilities. The model was used in both the accounting footnotes and the board’s capital allocation review. For once, the reporting exercise became a planning tool.

Disclosure, control, and audit readiness

ASC 820 requires disclosure of the hierarchy level, the valuation techniques used, and any significant unobservable inputs. For Level 3 instruments, a reconciliation of opening and closing balances is required, along with narrative disclosure of changes in inputs or assumptions.

CFOs must ensure that these disclosures are not only accurate but coherent. Audit readiness depends not just on the math, but on the governance. Who approves the assumptions? How frequently are models refreshed? Are changes to volatility or discount rates documented?

Control frameworks around fair value are often underdeveloped in high-growth companies. The finance team may rely on external valuation firms without adequate internal oversight. This is a risk. The external valuation is an input, not a decision. Management retains ultimate responsibility for the assumptions.

Conclusion: From hierarchy to humility

The fair value hierarchy was meant to bring order to the chaos of valuation. In many ways, it has succeeded. It distinguishes between what can be observed and what must be inferred. It compels disclosure and encourages discipline. But it also invites overconfidence. A Level 3 value, rendered to the nearest dollar, may look precise but be built on sand.

For finance leaders, the hierarchy is not a formula. It is a lens. It reminds us to consider where our numbers come from, how solid the ground beneath them is, and what story they tell. In a world where capital is abstract, instruments are engineered, and volatility is the norm, this clarity matters.

Call to action

CFOs should inventory all instruments subject to fair value measurement. Classify them rigorously. Understand the assumptions, challenge the models, and ensure the disclosures tell the whole story. Fair value is not just a number. It is a statement of what we believe today about what something might be worth tomorrow.