Independent Price Verification (IPV)

This is the control that checks the level of the book valuation against independent prices before it enters financial reporting, and investigates the cause if the gap is large.

Why valuations are checked against independent prices

Independent price verification (IPV) is a control in which people or units independent of the front office (sales and trading) recheck the valuation of a transaction against independent prices, rather than leaving it entirely to the front office's estimate. The initial price of a new transaction is usually set by the trader who structures it. In daily valuation as well, traders may be involved in deciding the prices and inputs used¹. The examination manual of the US Federal Reserve Board (the "examination manual" below) says that inadequate valuation policies and procedures raise doubts about a financial institution's trading profits and its ability to manage the risk of its transactions².

The examination manual statements here come from an edition dated February 1998 (manual pages 1 and 2) and one dated April 2003 (manual pages 3 to 8).

The rules of the UK Prudential Regulation Authority (PRA) require financial institutions to perform IPV in addition to daily marking to market. Verification of market prices and valuation inputs is performed by a person or unit independent of the persons or units that benefit from the trading book, at least monthly, or more frequently depending on the nature of the market or trading activity³. The examination manual likewise describes a periodic independent revaluation, separate from the daily valuation, to check that positions reflect fair value and that the valuation inputs are appropriate¹. IPV corresponds to this periodic revaluation.

Daily valuation and P&L sign-off is covered in Daily Valuation and P&L Sign-Off. A unit (product control) is responsible for checking daily valuations and P&L, but IPV is a separate control that periodically checks the level of valuations against independent prices.

Steps of IPV

IPV proceeds in this order: the policy sets the scope and the tolerance, independent prices are collected and compared with the book valuation, differences above the tolerance are investigated, and the results are recorded and reported. The names of the units responsible and how the work is divided differ between financial institutions. The steps below set out what the examination manual and the PRA rules call for.

Figure description

The steps of IPV. An independent person or unit selects the transactions to verify within the policy scope, collects independent prices and compares them with the book valuations. It investigates the causes of differences above the tolerance, decides on financial reporting fixes and, when independent prices are hard to obtain, also considers valuation adjustments. The results report goes to senior management and others.

1. The policy sets the frequency of verification, its scope, the tolerance, and the sources of independent prices and how their appropriateness is checked. The UK rules require the valuation policies and procedures to include the frequency of independent valuation, the sources of market information and the review of their appropriateness, and the procedures for end-of-month and ad hoc verification⁴. The examination manual says the policy should make the scope of verification clear and give a reasonable justification for any transaction excluded from it¹. Frequency and scope are covered in the subsection after these steps.
2. The independent person or unit selects the transactions to verify and collects independent prices. How prices are obtained depends on the liquidity of the transaction (see the later section).
3. The independent person or unit compares the book valuation with the value revalued using independent prices.
4. If the difference exceeds the tolerance in the policy, the cause is investigated.
5. The response is decided according to the cause. The response follows the sections "Handling differences from the book valuation" and "Reflecting results in valuation adjustments."
6. The results of verification are recorded in enough detail to serve as an audit trail. A summary of the results is passed to management units such as senior management. The report includes at least the scope of verification, significant differences between the independent valuation and the reported valuation, and the response to those differences⁵.

Frequency and scope are set by the policy

In the UK, verification is performed at least monthly, or more frequently depending on the nature of the market or trading activity³. The examination manual says independent revaluation is usually monthly, but that for illiquid transactions that are hard to value, and for transactions that are not marked to market daily, it may be prudent to revalue more frequently¹.

Scope varies with the size and sophistication of the institution's transactions. There are three examples of how to set the scope¹. Which one to use is written in the policy and differs between financial institutions.

How independent prices are obtained

Where independent prices come from depends on the liquidity of the transaction. The examination manual says that for products traded on exchanges or in liquid over-the-counter markets, the monthly revaluation may be done by comparing the book valuation with exchange or broker-dealer quotes. It also says the institution should make clear whether the party that supplied the price is a counterparty to the transaction or a paid pricing service. A broker-dealer quote may not be the same as the value the dealer uses internally, and may not represent values in other markets or model values. The financial institution needs to check for itself that valuations obtained from outside are appropriate⁵.

For illiquid over-the-counter transactions, independent third-party prices are usually not available. Broker quotes are sometimes available, but rarely. When there are several quotes, a representative value is set using a simple average or an average that excludes extreme prices, and the method is set in the institution's internal policy. When the middle office (risk management and similar functions) or the back office (operations and settlement) enters broker quotes, the traders are also asked to report their own positions as a check. When brokers enter data directly, the middle or back office checks the accuracy and appropriateness of that data. Price differences here are resolved by reconciliation⁶.

For transactions with complex structures, reliable independent quotes are usually unavailable even occasionally. In that case, methods include valuing at a benchmark rate plus a fixed spread, or comparing with similar transactions that have prices, and the price is usually produced by a model⁶. In the monthly revaluation, the checks are whether prices are consistent with past values and whether actual close-out prices and the results of hedge transactions agree with the model's predictions. For hard-to-value products, the control unit may also run a separate model in parallel to cross-check the trader's model value⁶.

For regulatory capital valuation too, when it is based on market data, the same range of data as in IPV is considered (EU Delegated Regulation 2016/101)⁷.

Handling differences from the book valuation

The examination manual says that when the value from the periodic independent revaluation differs significantly from the book value, further investigation is called for. The amount that triggers an investigation is set concretely in the written policy, for example "all discrepancies above $x thousand must be investigated to determine the source of the difference." Once the cause is known, the institution decides whether financial reporting needs to be corrected. Depending on the size and pattern of the differences, the pricing process or the models are changed¹⁵.

The difference between the book valuation and the independent price is viewed with this formula.

$$\text{Difference} = \text{Book valuation} - \text{Independent price}$$

Consider a hypothetical example in which three transactions are verified in an end-of-month IPV, under a policy that investigates the cause when the difference for a single transaction exceeds ±0.5 million yen.

For Transaction B, the book valuation is 1.5 million yen higher than the independent price, so the valuation and profit may be overstated. If the investigation finds that the volatility used in the valuation was out of date, the book valuation is corrected and the institution decides whether financial reporting needs to be corrected. For Transaction C, the book valuation is 0.8 million yen lower, so the valuation may be understated. If the investigation finds that the independent price is a broker's indicative value and not a level at which the position could actually be closed out, the book valuation may be left unchanged. Even when there is a difference, the independent price is not necessarily the correct one, so the institution checks the appropriateness of the price source before deciding which to use.

Reflecting results in valuation adjustments

The adjustments added to valuation also include those related to counterparty credit and funding costs (CVA, FVA and others).

When independent prices are hard to obtain or the price source is subjective, a valuation adjustment may be appropriate³. The PRA rules require procedures for considering valuation adjustments, and list eight adjustments to be formally considered: unearned credit spreads, close-out costs, operational risks, market price uncertainty, early termination, investing and funding costs, future administrative costs and, where relevant, model risk⁸.

Less liquid positions can arise from market events, concentrated positions, positions held beyond the intended holding period and similar situations. For such positions, procedures are needed for considering a valuation adjustment, separate from the changes in value needed for financial reporting. The rules list factors to consider in making the judgment. They include the time it would take to hedge out the position, the size of the bid-ask spread and the number and identity of the market makers quoting. They also include concentration of positions, the age of positions, and the extent to which valuation relies on marking to model⁹. When third-party valuations or marking to model are used, the institution also considers a valuation adjustment and explicitly assesses whether an adjustment is needed for the uncertainty of the model's input parameters¹⁰.

For the holdback reserve that defers part of a transaction's day-one profit, the examination manual calls for traders and business units not to be able to set the level of the reserve. An independent control unit often reviews whether the level is appropriate, together with the monthly revaluation¹¹.

Units involved and their roles

The PRA rules and the examination manual do not name the unit responsible for IPV. Both make independence from trading a condition. The PRA rules require that the person or unit performing verification be separate from those who benefit from the trading book³. The examination manual says it is done by a control unit that is separate from the trading unit¹. The names of units and how they are divided differ between financial institutions. The front office is the party whose valuations IPV checks. The work of the IPV unit in collecting and comparing prices is as described in the steps section, and the roles that do not appear in the steps are in the table below.

The division of work in the table rests on the PRA rules, the text of the EU regulation, and the statements in the supervisory guidelines of Japan's Financial Services Agency (FSA)^14121314.

Relevant regulation

Documented valuation policies and procedures, a reporting line for the valuation unit that is independent of the front office (ultimately reaching senior management), and IPV in addition to daily valuation are requirements of the UK PRA rules (Article 105)⁴³.

In Japan, the Financial Services Agency's "Comprehensive Guidelines for Supervision of Major Banks, etc." lists points of focus for market-based credit products such as securitized products (including market-based loans and credit default swaps (CDS)). When prices calculated by the front office are used as market values for risk management, the FSA checks whether risk management and similar units verify those prices from an independent standpoint. When valuations are obtained from brokers or external vendors, it checks whether the institution asks, as far as possible, for information on the valuation method and works to verify appropriateness. When model uncertainty or liquidity risk is material, it also checks whether these are taken into account¹⁴.

For regulatory capital valuation there is prudent valuation. EU Delegated Regulation 2016/101 (original text of October 26, 2015) applies a conservative valuation standard to all positions measured at fair value. The difference from the accounting valuation is calculated as an additional valuation adjustment (AVA) and used only to adjust Common Equity Tier 1 capital¹⁷. The AVA is the absolute value of the difference when the prudent value is at or below fair value for an asset, or at or above fair value for a liability, and it is never negative¹⁷¹⁸. There are three links with IPV. The market data considered in calculating AVAs is the same range as in IPV (see the section on how independent prices are obtained)⁷. As a valuation control, the regulation requires a formal IPV process based on prices independent of the trading desk. AVAs are approved by an independent control unit and monitored on an ongoing basis¹³. The details of how AVAs are calculated are not covered on this page.

Notes

1. Board of Governors of the Federal Reserve System, "Trading and Capital-Markets Activities Manual," Section 2100.1 (February 1998), page 2, "Initial Pricing," "Daily Mark-to-Market Process" and "Independent Price Testing and Revaluation." ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

2. Board of Governors of the Federal Reserve System, "Trading and Capital-Markets Activities Manual," Section 2100.1 "Financial Performance" (February 1998), page 1, preamble and "VALUATION." ↩ ↩²

3. Prudential Regulation Authority, "PRA Rulebook," Trading Book (CRR) Part, Article 105 (version of February 8, 2025), paragraph 8. ↩ ↩² ↩³ ↩⁴ ↩⁵

4. Prudential Regulation Authority, "PRA Rulebook," Trading Book (CRR) Part, Article 105 "Requirements for Prudent Valuation" (version of February 8, 2025), paragraph 2(a) and (b). ↩ ↩² ↩³

5. Board of Governors of the Federal Reserve System, "Trading and Capital-Markets Activities Manual," Section 2100.1 (April 2003), page 3, "Liquid Instruments and Transactions" and others. ↩ ↩² ↩³ ↩⁴

6. The same manual, Section 2100.1 (April 2003), page 4, "Illiquid Instruments and Transactions." ↩ ↩² ↩³ ↩⁴

7. Commission Delegated Regulation (EU) 2016/101 (October 26, 2015; original text on legislation.gov.uk), Article 3, paragraph 1. ↩ ↩²

8. Prudential Regulation Authority, "PRA Rulebook," Trading Book (CRR) Part, Article 105 (version of February 8, 2025), paragraphs 9 and 10. ↩

9. Prudential Regulation Authority, "PRA Rulebook," Trading Book (CRR) Part, Article 105 (version of February 8, 2025), paragraph 11(a) to (g). ↩

10. Prudential Regulation Authority, "PRA Rulebook," Trading Book (CRR) Part, Article 105 (version of February 8, 2025), paragraph 12. ↩

11. Board of Governors of the Federal Reserve System, "Trading and Capital-Markets Activities Manual," Section 2100.1 (April 2003), page 7, "HOLDBACK RESERVES" and "Reserve Adequacy." ↩

12. Prudential Regulation Authority, "PRA Rulebook," Trading Book (CRR) Part, Article 105 (version of February 8, 2025), paragraph 7(f). ↩

13. Commission Delegated Regulation (EU) 2016/101 (October 26, 2015; original text on legislation.gov.uk), Article 19, paragraph 1, paragraph 2(b), (d) and (e), and paragraph 3(b) and (e). ↩ ↩²

14. Financial Services Agency, "主要行等向けの総合的な監督指針" (Comprehensive Guidelines for Supervision of Major Banks, etc.), III-2-3-3-2(3), item ① (sub-items ロ, ハ and ニ) (October 2026, page 126) (in Japanese). ↩ ↩²

15. Board of Governors of the Federal Reserve System, "Trading and Capital-Markets Activities Manual," Section 2100.1 (April 2003), page 5, "Pricing Models." ↩

16. The same manual, Section 2100.1 (April 2003), page 6, "Pricing-Model Outputs." ↩

17. Commission Delegated Regulation (EU) 2016/101 (October 26, 2015; original text on legislation.gov.uk), recitals (1), (2) and (4). ↩ ↩²

18. Commission Delegated Regulation (EU) 2016/101 (October 26, 2015; original text on legislation.gov.uk), Article 8, paragraph 4. ↩