Netting Enforceability and Legal Opinions

A bank with a close-out netting contract obtains a legal opinion when it uses the contract's effect in its capital adequacy ratio calculation.

What a legal opinion checks

What the opinion confirms is whether close-out netting, as the contract provides, would be legally recognized if the counterparty failed.

Close-out netting does not work just because a contract says so. It depends on the law of the jurisdiction that handles the counterparty's failure recognizing it. This premise is covered in Close-Out Netting.

The Financial Services Agency (FSA) notice issued under Article 14-2 of the Banking Act (the "notice") sets the standards for Japan's capital adequacy ratio rules. As a precondition for recognizing netting, the notice requires a "legally valid bilateral netting agreement," meaning a contract that is valid under the laws of the relevant countries. The FSA Q&A says this validity is confirmed through a written legal view (a legal opinion). The view reasonably states that the competent court and authorities would limit the bank's exposure to the net amount in a dispute¹. The same Q&A gives as examples of who may prepare an opinion an outside law firm or the bank's own in-house legal function².

The combination of laws to check

What is checked is not a single law but a combination of laws determined by the counterparty and the contract. The FSA Q&A lists three relevant laws¹.

• The law of the country that granted the counterparty its license to be established (together with the law of the country where a branch is located, if the counterparty's overseas branch is included)
• The law governing the individual transactions subject to netting
• The law governing the contracts needed to carry out the netting

Article 296 of the EU Capital Requirements Regulation likewise names four jurisdictions an opinion should refer to³.

• The counterparty's jurisdiction of incorporation
• The jurisdiction of a branch, if there is one
• The jurisdiction whose law governs the individual transactions
• The jurisdiction whose law governs any contract needed to effect the netting

Consider a hypothetical example in which a Japanese bank trades with Company Z, which is incorporated in Country P and has a branch in Country Q. The bank trades with both Z's head office and its Q branch, and the individual transactions and the master agreement are governed by Japanese law.

In this example, the bank would check whether close-out netting is expected to be recognized in each of three jurisdictions: Country P, Country Q, and Japan. If a contract is governed by the law of yet another country, the number of jurisdictions to check increases.

An opinion is not always obtained contract by contract. The EU regulation allows opinions to be prepared for each type of netting⁴. The FSA Q&A also does not necessarily require obtaining one per contract for regulatory purposes. A bank is, however, expected to check each concluded contract against the opinions it holds. If a contract cannot be judged legally valid, the bank supplements the opinion².

Sometimes the content of a contract decides the conclusion. A walk-away clause allows the non-defaulting party to make limited payments, or no payments at all, even when the defaulting party is a net creditor. The Q&A says that for a contract containing this clause, the effect of netting cannot be taken into account when calculating the credit equivalent amount (the amount corresponding to exposure at default)⁵.

Netting across different products (cross-product netting) widens what must be checked. Under the EU regulation, an opinion must address the validity of the contract as a whole and the effect of the netting on the material provisions of the individual master agreements it includes⁴. The US Federal Reserve's rule likewise requires a written legal opinion for cross-product master netting agreements⁶.

What regulators require as confirmation, such as an opinion

From the regulator's side, an opinion is a means of meeting the conditions for using the effect of netting in capital calculations. The table lists the provisions of Japan, the EU, and the United States.

All three rules require validity to be confirmed under the laws of the relevant jurisdictions. They differ in how the confirmation is documented. Japan's notice (the condition for the expected exposure method) requires an appropriate confirmation of validity, the EU regulation requires a written, reasoned legal opinion, and the US rule requires sufficient legal review and a written record of it.

The calculation without an opinion

For a counterparty whose validity cannot be confirmed through an opinion, the effect of netting may not be usable. Under the notice's definition, a set of trades under a legally valid bilateral netting agreement forms one netting set, and every other trade forms a netting set of its own, one per trade⁷. The link between a contract whose validity cannot be confirmed and treatment as individual trades is a reading drawn from this definition and the Q&A's validity test.

The conditions so far can be arranged as a test of whether the effect of netting can be used in the credit equivalent amount.

Figure description

For the second question, a yes means the contract has a walk-away clause and netting cannot be taken into account, and a no leads on. Banks using the expected exposure method also face conditions not shown in this figure, such as building the credit equivalent amount into risk management procedures and retaining documents.

Consider a hypothetical example in which trades with Company Z fall into three situations.

• Example 1: validity is confirmed in Country P, Country Q, and Japan, and the contract has no walk-away clause. As a rule, the effect of netting can be taken into account for the set of trades with Z.
• Example 2: validity is confirmed in Country P and Japan, but not in Country Q, where the branch is located. Country Q is also a relevant jurisdiction, because the bank also trades with Z's branch. The trades may be treated as individual trades. The figure's endpoint shows all the trades as individual, but this reading does not settle whether only the branch trades drop out, or whether the head office trades are also treated individually.
• Example 3: validity is confirmed in all three jurisdictions, but the contract has a walk-away clause. The effect of netting cannot be taken into account in the credit equivalent amount⁵. Confirming validity is not enough on its own.

Next, a numerical comparison shows how the replacement cost changes between two cases. In one, validity is confirmed, as in example 1. In the other, all the trades are treated individually, following the simplified reading of example 2. The credit equivalent amount under SA-CCR is calculated for each netting set as $1.4 \times (RC + PFE)$. Here $RC$ is the replacement cost and $PFE$ is the potential future exposure. With no collateral agreement, $RC$ is given by the following formula¹¹.

$$RC = \max\{V - C,\ 0\}$$

$V$ is the sum of the market values of the trades in the netting set, and $C$ is the net collateral amount after haircut adjustment (as defined in the notice).

Consider a hypothetical example in which three trades between the bank and Company Z have market values, from the bank's side, divided into the three cases below. A positive value is an amount the bank receives and a negative value is an amount it pays. There is no collateral, so $C = 0$. The $PFE$ and the multiplier of 1.4 are left out. Market values are in units of 100 million yen.

For these three cases, $RC$ differs between when validity is confirmed and when the trades are treated individually, as shown below. The unit is 100 million yen.

In case A, if validity is confirmed, the total of 200 million yen becomes $RC$. Without confirmation, trade 1's 900 million yen is its own $RC$, and trades 2 and 3 are 0 because their values are negative. In case B the total is negative, so $RC$ is 0 if validity is confirmed, and trade 1's 300 million yen remains without confirmation. In case C all three values are positive, so confirmation makes no difference. In other words, the effect of netting reduces $RC$ when positive and negative trades are mixed.

An actual credit equivalent amount also includes $PFE$, and the formula changes when there is a collateral agreement. This table is a simple example for seeing how the presence or absence of confirmation changes the inputs to the calculation.

Legal reforms and updating opinions

An opinion is a view given under the law as it stood when the opinion was written. If the law changes, the premises of its conclusion can change. The US Federal Reserve rule requires written procedures for monitoring changes in the relevant laws, so as to confirm that a netting agreement continues to meet the conditions¹⁰.

Resolution regimes are an example of legal reform. In Japan, the 2013 amendment of the Deposit Insurance Act added Article 137-3. When the Prime Minister takes certain measures, such as granting a certification, the Prime Minister can decide that a clause in a contract providing for specified termination (including close-out netting) because of that measure has no effect while the measure is in force¹². According to an account from around 2015, the EU had a regime under which authorities could temporarily stay the exercise of netting, and it was unclear whether the stay would also extend to counterparties in other jurisdictions¹².

Such regimes can bear on the range in which close-out netting works.

What an opinion cannot tell you

The US rule requires a well-founded conclusion on how courts and authorities would decide in a dispute¹⁰. The FSA Q&A also describes the view as one showing that the competent court and authorities "would" decide in a certain way¹. An opinion states how a decision is expected to come out. The actual decision is made by the competent court and authorities when a dispute arises.

An opinion also leaves some matters uncovered. Where a contract cannot be judged valid in light of the opinions held, or in other cases where it is needed, the bank supplements the opinion². The treatment when validity cannot be confirmed is set by the rules. The EU regulation does not recognize that netting agreement as risk-reducing³. The US rule likewise does not recognize an agreement that fails the conditions as a qualifying master netting agreement¹⁰.

Managing opinions inside the firm

To keep the confirmation going, one possible design is to link each counterparty's contract and its opinion to the trade calculations. Japan's notice sets conditions for a bank that uses the expected exposure method and wants to take netting into account. They include building the netted credit equivalent amount into ordinary risk management procedures and keeping all contract documents appropriately⁸.

The content of a counterparty's contract is in the master agreement, such as an ISDA Master Agreement. The structure of the contract documents is covered in The ISDA Master Agreement and Schedule.

What this page does not cover

The kinds of opinion and their assumptions are not covered on this page.

Notes

1. Financial Services Agency, "自己資本比率規制に関するＱ＆Ａ" (Q&A on capital adequacy ratio regulation; in Japanese; unofficial translation of the title), full version updated August 24, 2026, "Article 79-2, Q20" (page 130). The Q was added on July 28, 2006 and revised on March 23, 2018. ↩ ↩² ↩³ ↩⁴

2. The same Q&A, "Article 79-2, Q21" (pages 130–131). ↩ ↩² ↩³ ↩⁴ ↩⁵

3. Regulation (EU) No 575/2013, Article 296, paragraph 2(b) (the version legislation.gov.uk publishes as "Latest available (Revised)", checked on October 5, 2026). Items (i) to (iv) of (b) are the four jurisdictions. The end of paragraph 2 says that if the authority is not satisfied that the agreement is valid in any of the jurisdictions, it does not recognize it as risk-reducing. ↩ ↩² ↩³

4. Regulation (EU) No 575/2013, Article 296, paragraph 3 (same version on legislation.gov.uk). The opening of paragraph 3 allows opinions for each type of netting, and (b) requires cross-product contracts to address the contract as a whole and the effect on the individual master agreements included. ↩ ↩²

5. The same Q&A, "Article 79-2, Q19" (page 130). The walk-away clause is explained in the note to that Q. ↩ ↩²

6. 12 CFR § 217.3(c) (the e-CFR as published by the Legal Information Institute, Cornell Law School, checked on October 5, 2026). A provision on institutions subject to the Federal Reserve Board's regulation. ↩

7. The same notice, Article 79, paragraph 7 (pages 189–190). It defines a netting set as the group of trades under a legally valid bilateral netting agreement, and as a single trade for any other trade. ↩ ↩² ↩³

8. Financial Services Agency, "銀行法第十四条の二の規定に基づき、銀行がその保有する資産等に照らし自己資本の充実の状況が適当であるかどうかを判断するための基準" (the notice setting the criteria under Article 14-2 of the Banking Act for judging whether a bank's capital is adequate in light of its assets; in Japanese; unofficial translation of the title), full version updated August 24, 2026, Article 79-3, paragraph 16, items 1 to 4 (pages 211–212). Item 2 covers confirmation of validity, item 3 building the credit equivalent amount into risk management procedures, and item 4 retention of documents. ↩ ↩²

9. The same notice, Article 79-3, paragraph 3, items 2 to 4 (page 207). ↩ ↩² ↩³ ↩⁴

10. 12 CFR § 217.3(d) (same). Paragraph (1) requires sufficient legal review (a "well-founded basis") and a written record, and paragraph (2) requires written procedures to monitor changes in relevant law. ↩ ↩² ↩³ ↩⁴

11. The same notice, Article 79-2, paragraph 1 and paragraph 2, item 1 (pages 190–191). The definitions of $V$ and $C$ in the worked example are in the same item. The example has no collateral and $C = 0$. ↩

12. Hideki Kanda (神田秀樹), "デリバティブ取引に関する法の動向――ネッティングを中心として――" (Trends in the law on derivatives transactions, focusing on netting; in Japanese; unofficial translation of the title), Chapter 2 of a report of the Financial Law Board, Japanese Bankers Association (the file name in the URL indicates fiscal 2015), page 24. It describes the addition of Article 137-3 to the Deposit Insurance Act, and that it was unclear whether a temporary stay of netting by EU authorities would also reach counterparties in other jurisdictions. ↩ ↩²