A distinctive feature of tax controversy in Japan, compared with other jurisdictions, is the relatively small number of formal disputes, such as tax litigation and administrative appeals. This means that, in Japan, discussions between taxpayers and the tax authorities during tax audits – the stage preceding formal proceedings – serve as the de facto forum for tax controversy.
In other words, formal tax controversy – tax litigation and administrative appeals – begins after the tax authority issues a tax assessment or other administrative disposition, and the taxpayer files an administrative appeal. By contrast, the substantive phase of a tax controversy – namely, discussions between the taxpayer and tax authority during an audit – typically begins when, after the taxpayer has filed a tax return, the authority issues an audit notice.
In Japan, tax controversies most often arise from issues of factual determination rather than statutory interpretation. Disputes often arise between taxpayers and tax authorities over the existence of facts that satisfy the statutory criteria for taxability. This is true regardless of the type of tax or the amount in dispute. However, where the applicable criteria are relatively indeterminate – for example, when the conclusion requires a holistic assessment of the totality of the facts – differences in factual determinations are more likely to emerge and develop into tax controversies.
In Japan, discussions between taxpayers and tax authorities during tax audits often serve as the de facto forum for tax disputes, and differences of opinion frequently arise – particularly over factual determinations. Despite this, there has traditionally been a tendency to avoid controversy too readily by negotiating the tax amount without fully engaging with the merits.
However, the firm does not believe this traditional approach necessarily yields outcomes favourable to taxpayers. While the appropriate course depends on the circumstances, it is generally preferable at the audit stage to submit rebuttal letters or legal opinions akin to briefs filed in tax litigation and to engage in thorough discussions with the tax authorities to resolve differences of opinion.
In Japan, the Base Erosion and Profit Shifting (BEPS) recommendations have, as a general matter, already been incorporated into domestic legislation, and tax administration and enforcement are conducted in line with them. As a result, the archetypal forms of tax avoidance that the BEPS Project sought to curb appear to have diminished.
At the same time, Japan may at times have implemented the BEPS recommendations in a somewhat mechanical manner, without always fully co-ordinating them with existing domestic law. This may have increased the risk of tax issues in certain areas, such as potential double taxation under the controlled foreign company (CFC) regime.
Moreover, as a number of the BEPS recommendations establish requirements that are not necessarily clear – such as the interpretation of payments economically equivalent to interest under the earnings stripping rules – an increase in tax controversies can be expected going forward.
In Japan, taxpayers are not required to prepay the assessed additional tax in order to file an administrative appeal against a tax assessment. However, the tax liability is not suspended until the tax assessment is finally set aside; accordingly, if payment is not made, delinquent tax for late payment will continue to accrue, and there remains a risk of enforced collection measures. For this reason, in practice, taxpayers commonly pay the assessed additional tax and then pursue the appeal.
Corporate Taxpayers
The tax authorities employ AI-driven predictive models to identify entities with a high likelihood of requiring audit. They analyse tax returns together with a wide range of data and materials to determine whether an audit should be conducted. In particular, they conduct rigorous audits of:
Individual Taxpayers
The authorities actively conduct audits of high net worth individuals (HNWIs) – such as large holders of securities and real estate, individuals with particularly high recurring income and those who actively invest overseas – mindful of the increasing diversification and globalisation of asset management. They also actively audit individuals engaged in transactions in emerging economic sectors, such as the sharing economy conducted via online platforms, as well as those trading in crypto-assets.
Initiation of a Tax Audit
Before initiating a tax audit, the tax authorities generally are required to notify a taxpayer of the start date, time and location of the field audit, as well as the tax items, books, documents, and other materials and period to be audited. The taxpayer may request a change to the date and time of the audit if there is a legitimate reason to do so. The tax authorities may commence a tax audit without notification where an advance notice may impede the precise finding of facts or otherwise limit their ability to appropriately conduct the audit.
Duration of a Tax Audit
The duration of a tax audit depends on the taxpayer’s circumstances. Usually, an audit takes between one and six months but may take longer, particularly where a transfer pricing audit is conducted.
The tax authorities generally must issue an assessment notice within five years after the statutory due date for filing the tax return (within six years for gift tax, seven years for transfer pricing matters and ten years for matters in relation to net operating losses regarding corporation tax). The assessment period is seven years where the taxpayer reduces their tax liability or receives a tax refund based on false statements or misconduct. The tax authorities must conduct a tax audit before the assessment deadline expires and, if necessary, issue a tax assessment.
Location of Tax Audits
As a general rule, field audits are conducted on the taxpayer’s premises. An examiner conducting a field audit who visits a taxpayer’s office is required to carry and present official identification displaying their name and title.
Drawing on analysis of various data – such as return information and prior audit history – the tax authorities may, where they determine that a taxpayer is not attempting to evade tax, opt for simpler forms of contact, such as letters or telephone calls, in lieu of a field audit.
Procedure of Tax Audits
The tax examiner is authorised to require the taxpayer to produce relevant books, documents and other materials for inspection, and may question the taxpayer as necessary. Audits are based not only on printed documents but also on electronically stored or submitted data.
Penalties may be imposed if the taxpayer fails to provide information or provides incorrect information in response to the examiner’s questions, refuses to co-operate with the inspection, fails to produce requested books, documents or other materials, or knowingly produces materials containing false information. In the case of an individual taxpayer or individual acting on the taxpayer’s behalf, penalties may include imprisonment.
The tax examiner may retain books, documents or other materials (generally with the taxpayer’s consent). If materials are retained, the examiner must issue a receipt and return them to the taxpayer as soon as they are no longer needed.
If necessary, the tax examiner may also question a counterparty to the taxpayer’s transactions (including an individual taxpayer’s employer) as well as request and inspect the counterparty’s relevant books, documents and other materials.
Explanation of Audit Results
Where the tax authorities find that a tax return was incorrect or that the taxpayer failed to file a return, they will explain their audit findings, including any errors identified and the calculation and amount of any additional tax payable.
The tax authorities usually will encourage the taxpayer to file an amended tax return voluntarily. After doing so, the taxpayer cannot initiate an administrative appeal without first requesting that the tax authorities issue an assessment notice decreasing the amount of tax.
Alternatively, where the tax authorities find that the tax return was not incorrect or that the taxpayer was not required to file a return, they will notify the taxpayer accordingly, in writing.
Assessment Notice
Where the tax authorities find that a tax return was incorrect or that the taxpayer failed to file a return, and the taxpayer does not file an amended return, the tax authorities will issue an assessment notice. When an assessment notice is issued or a request for assessment decreasing the amount of tax from a taxpayer is rejected, the tax authorities must state the reasons for doing so on the relevant notice.
A taxpayer should pay the principal tax due plus any penalty tax assessed within one month after the date the assessment notice is issued, as well as any delinquent tax, regardless of whether the taxpayer intends to appeal the assessment.
However, if the taxpayer is unable to pay the tax at one time, the tax authorities may suspend the tax payment (in full or part) for up to one year after the due date, but only if the tax is finalised one year or more after the statutory filing date of the return and if the taxpayer has made such a request by the due date for payment of the assessed tax.
The tax authorities strive to ensure rigorous and proper administration of the consumption tax. In particular, when taxpayers attempt to obtain refunds improperly by filing false returns, refunds are suspended, and the factual basis for the claim is verified through audits and other procedures.
On the other hand, the authorities’ vigorous efforts to address fraudulent consumption tax refunds may, at times, have led to factual determinations that could be viewed as somewhat expansive and assessments that may not have been entirely appropriate. As a result, there are numerous tax disputes involving the consumption tax, and, in some cases, taxpayers’ positions have been upheld.
The tax authorities make effective use of data – including reports of overseas remittances, information exchanged with foreign authorities under tax treaties and information received under the Common Reporting Standard (CRS) – to establish the facts and conduct in-depth examinations.
In particular, for HNWIs, they aim to ensure appropriate taxation of investment returns generated by increasingly diversified and international asset management, while also accumulating information to facilitate appropriate inheritance tax assessments in the future.
Accordingly, it can be said that traditional forms of tax avoidance – such as hiding assets offshore – are no longer effective.
Most taxpayers want to resolve tax disputes as quickly as possible. While filing an administrative appeal or pursuing tax litigation may ultimately result in the taxpayer’s position being upheld, doing so entails substantial time, effort and expense.
What, then, should be done? The answer is to submit, at the audit stage, written arguments – of a quality and level of detail comparable to those filed in tax litigation – together with supporting evidence and documentation. This approach allows taxpayers to resolve disputes as swiftly as possible. In the past, some may have rushed to negotiate assessment amounts without fully fleshing out the issues with the tax authorities; it may now be necessary to engage in a more thoroughly reasoned discussion at the audit stage.
Initiating Administrative Appeals
After a tax audit, a taxpayer may challenge a tax assessment issued as a result of the audit. Once the assessment notice is issued, the taxpayer may either accept the assessment or challenge it by filing an administrative appeal.
Outside of the audit context, if a taxpayer believes the amount of tax paid was incorrect and should have been lower, the taxpayer must first request that the tax authorities issue an assessment reducing the tax. After reviewing the request, the tax authorities will issue a notice of assessment or other disposition either granting or denying the requested reduction; an adverse notice may be challenged by filing an administrative appeal.
As a general rule, a taxpayer may bring an action in court only after filing a second‑tier administrative appeal with the National Tax Tribunal and receiving an unfavourable decision.
First-Tier Administrative Appeal
When the opinion of a taxpayer under tax audit differs from that of the tax authorities, the optimal solution generally would be to resolve the difference during the tax audit. However, if the difference cannot be resolved in the tax audit, the taxpayer may choose to file a first-tier or second-tier administrative appeal. Filing a first-tier administrative appeal generally will offer the fastest route to resolve any differences or address a contested fact pattern included in the assessment notice.
A first-tier administrative appeal to the tax authorities that issued the assessment notice generally must be initiated within three months after the date on which the notice is received by the taxpayer or the date the taxpayer acknowledges its issuance (if it is not received by the taxpayer).
First-tier administrative appeals are handled by a specific tax division, which often is the same division that internally reviewed the original assessment notice. The first-tier administrative appeal essentially is a quick internal review process. An assessment may be overturned, for example, if it was based on inaccurate information or the taxpayer subsequently obtained favourable evidence that was not available during the tax audit. Even if the taxpayer’s first-tier administrative appeal is denied, the reason for the decision will be disclosed in more detail, which is useful in conducting a second-tier administrative appeal if the reason for the assessment is not necessarily clear.
Where a taxpayer’s first-tier administrative appeal is denied, the taxpayer still may file a second-tier administrative appeal with the National Tax Tribunal. In such a case, the second-tier administrative appeal generally must be filed within one month after the date the tax authorities’ decision is delivered to the taxpayer.
Second-Tier Administrative Appeal
By doing a second-tier administrative appeal, the taxpayer may seek a final decision on the matter by the administrative branch. The National Tax Tribunal hears arguments from both the taxpayer and tax authorities, performs its own review of the evidence and issues a decision. The process is independent, not open to the public, and free of charge and allows taxpayers to address a contested assessment on a timelier basis than if they were to pursue tax litigation. Even if tax litigation becomes necessary (ie, if the National Tax Tribunal issues a decision that is unfavourable to the taxpayer), clarifying contentious points in advance should allow tax litigation to proceed more efficiently.
Where a taxpayer appeals directly to the National Tax Tribunal without making a first-tier administrative appeal, the second-tier administrative appeal generally must be initiated within three months after the date the assessment notice is received by the taxpayer or the date the taxpayer acknowledges its issuance (if it is not received by the taxpayer).
Upon receiving the appeal, the National Tax Tribunal notifies the tax authorities that issued the assessment that an appeal has been initiated, and the tax authorities file a response with the National Tax Tribunal. The taxpayer then may present further arguments and evidence. After the National Tax Tribunal considers the relevant arguments and evidence presented, it ends its examination and issues a decision in writing.
If the National Tax Tribunal renders a decision adverse to the taxpayer, the taxpayer may bring an action in court. Conversely, if the decision is adverse to the tax authorities, they have no right of appeal to the courts.
More than half of the National Tax Tribunal’s staff are seconded from the tax authorities as part of their job rotations. However, it is mostly headed by civil court judges, and attorneys, certified public tax accountants and certified public accountants are usually hired as fixed-term appeal judges.
First-Tier Administrative Appeal
The tax authorities generally aim to render a decision within three months of the filing of the appeal. If no decision is rendered within three months of the date the first‑tier administrative appeal was filed, the taxpayer may initiate a second‑tier administrative appeal.
Second-Tier Administrative Appeal
The National Tax Tribunal generally renders a decision within one year of the date the appeal was filed. If the Tribunal does not render a decision within three months of the date the appeal to the Tribunal was filed, the taxpayer may bring an action in court.
If the taxpayer wishes to continue challenging a tax assessment or other disposition after the National Tax Tribunal has rendered its decision, the taxpayer may bring an action in court within six months after the date the taxpayer acknowledged the issuance of the Tribunal’s decision.
When initiating tax litigation, the taxpayer first files a complaint with the district court. After a formal review of the complaint, the court serves it on the tax authorities and directs them to submit a response. The court then schedules hearing dates at which both the taxpayer and tax authorities can appear, and the proceedings move forward. In principle, the case is adjudicated on the basis of the parties’ briefs and documentary evidence, though witnesses may sometimes be examined. Once the court considers the case ready for decision, it closes the proceedings and renders judgment.
In tax litigation, it is common for both the taxpayer and tax authorities to actively submit documentary evidence together with their pleadings – the complaint or response – and, in many cases, such evidence proves decisive.
However, where the dispute cannot readily be resolved on written submissions alone, witness testimony is taken upon a motion by either the taxpayer or tax authorities, with both sides examining the witness. Witness examination is typically conducted after the proceedings have progressed to a certain stage.
The burden of proof generally lies with the tax authorities issuing the assessment, who must provide evidence to support their position. However, where a taxpayer files a request for a reduced assessment and the tax authorities issue a notice denying the request, the burden of proof lies with the taxpayer to show that a reduction is warranted.
In tax litigation, the cardinal rule is to present your strongest arguments and evidence at the outset. Courts attach the greatest weight to the initial submissions, and at times the tax authorities may accept the taxpayer’s position, and the matter will be resolved. By contrast, settlements within litigation are not provided for under the Japanese procedure. Accordingly, unless one party fully accepts the other side’s claims, the dispute will continue until a final judgment is rendered.
Where the principal issue is the interpretation of tax law, it can be effective to submit an expert opinion from a tax-law scholar. As a general matter, whether the taxpayer pays the tax assessed under the assessment does not affect the outcome.
Although Japan is a civil-law jurisdiction, judicial precedents – particularly Supreme Court decisions – carry significant weight in adjudication and are frequently cited in judgments. In the field of international taxation, courts are expected to refer to the OECD Model Tax Convention Commentary and BEPS-related materials when reaching their decisions.
While relevant precedents from other countries are not directly binding, they may, in some cases, be consulted as persuasive authority.
Japan has a three-tiered judicial system comprised of district courts (the first level), High Courts (the second level), and the Supreme Court (the final level). A taxpayer must file their initial lawsuit in the district court and may appeal an unfavourable decision to a High Court. In general, the Supreme Court deals only with legal matters and has the discretion to refuse to consider cases.
At first instance, both the taxpayer and tax authorities fully develop their arguments and evidence, and the court advances the proceedings over an appropriate period of time. By contrast, on appeal, the losing party files a notice of appeal and statement of reasons for appeal together with supporting evidence, and the opposing party files a response to the appeal with its evidence; however, if the court considers that no further proceedings are necessary, it is not uncommon for the case to be concluded after a single hearing.
At the final appeal stage before the Supreme Court, the losing party typically files a petition for acceptance of final appeal and statement of reasons, with supporting evidence. The opposing party is not required to submit papers. Even if the Supreme Court accepts the appeal, the parties generally do not exchange briefs, although each may submit briefs separately to the Court. If the Supreme Court intends to alter the appellate court’s judgment, it will hold a hearing and, after formally confirming the parties’ positions, render judgment; if it finds no grounds for the final appeal, it will generally dismiss the appeal without holding a hearing.
The timeframe for tax litigation depends on the individual case. It usually takes around one and a half years at the district court and within one year at the High Court. It may take longer than one year at the Supreme Court (if it accepts the case).
At first instance and on appeal, cases are decided after collegial deliberation by panels of three judges, while at the Supreme Court they are typically decided by a five-justice panel. At first instance, depending on the district court, cases may be heard by judges with substantial experience in tax litigation; however, at the appellate level and in the Supreme Court, there is generally no particular specialisation.
Although the successful tax litigation ratio is relatively lower compared with most other countries, Japanese courts generally are considered reliable and free from bias, as it appears that taxpayers’ reasonable arguments tend to be actually accepted by the court.
In Japan, there is no official settlement procedure between a taxpayer and the tax authorities during tax controversy procedures, other than the mutual agreement procedure (MAP) in an applicable tax treaty. However, the tax authorities often recommend that a taxpayer file an amended tax return during the tax audit in accordance with a de facto agreement between the taxpayer and tax authorities, which may function as an unofficial settlement.
There is no applicable information in this jurisdiction.
There is no applicable information in this jurisdiction.
A taxpayer may seek an advance ruling from the tax authorities in relation to a transaction that either has taken place or will take place before the due date for filing the tax return reflecting the transaction. An advance ruling may not be requested where:
The tax authorities generally try to issue an advance ruling within three months after the date the request is filed, but it may often take longer. Taxpayers, therefore, tend to seek an informal response from the tax authorities rather than seeking an advance ruling, or tend to adopt a position without any response or ruling from the tax authorities.
Another potential challenge for a taxpayer is disclosure. An advance ruling generally will be published within two months after its issue date, although the taxpayer may request the tax authorities delay publication for up to one year after the decision is made. Where a taxpayer does not want details of the transaction to be disclosed, they should not seek an advance ruling.
There is no applicable information in this jurisdiction.
When an assessment notice results in international double taxation for a taxpayer, the taxpayer may request that the tax authorities enter into a MAP with the relevant foreign tax authorities under provisions to resolve such double taxation contained in the applicable tax treaty.
Criminal tax proceedings are separate and independent from tax assessments. Criminal tax penalties are imposed only when a taxpayer evades taxes or obtains tax refunds through intentional acts, such as fraudulent conduct, and are not imposed in cases of violations resulting from negligence or differences of opinion between taxpayers and the tax authorities.
Therefore, when the tax authorities identify errors in a taxpayer’s return in the course of an ordinary tax audit, they issue an assessment but do not immediately impose criminal penalties. However, where a taxpayer has wilfully evaded tax by fraudulent means, the authorities conduct a compulsory investigation using coercive powers, akin to a criminal investigation, and, based on the results, file a criminal complaint with the public prosecutor and request indictment.
On the other hand, administrative penalty proceedings are conducted in conjunction with tax assessments. Administrative penalties are automatically imposed when tax auditors identify underreporting, failure to file tax returns by the deadline or non-payment of withholding taxes (WHTs) during a tax audit, unless the taxpayer has a legitimate reason for such conduct.
The appeal process, including tax litigation challenging an assessment, and the process for imposing criminal penalties for tax evasion are separate and independent. Hence, in theory, criminal proceedings are not stayed even if the assessment has not yet become final and binding in court. However, in practice, appeal proceedings before the National Tax Tribunal regarding the assessments are suspended until the criminal proceedings have concluded. This is because the evidence of the tax offences is in the possession of the public prosecutor’s office, and the tax authorities cannot access it.
When a suspected tax evader is identified, tax examiners first conduct a covert preliminary inquiry to verify more concretely the scale of the evasion and methods used. After this covert investigation, where substantial evasion is anticipated and the methods are deemed egregious, the tax authorities present the facts supporting the suspicion of tax evasion to a judge and obtain a warrant. Based on that warrant, they commence a compulsory investigation.
In general, ordinary tax audits do not directly lead to criminal investigations. Tax examiners handling criminal cases are assigned to departments specialising in criminal proceedings and do not conduct tax audits for assessment purposes.
It remains an open question whether evidence obtained through ordinary tax audits may be used in criminal proceedings in relation to constitutional provisions concerning the privilege against self-incrimination. One eminent scholar argues that evidence from tax audits should be inadmissible in criminal proceedings and that tax officers’ duty of confidentiality should prevail over their obligation, as public servants, to file a criminal complaint when they discover a criminal tax offence during a tax audit. However, judicial precedent on these issues is unclear.
By contrast, Supreme Court precedent permits the use of evidence obtained through criminal investigations for assessment purposes.
Tax examiners ascertain the facts underlying the suspicion through a criminal tax investigation and, where the results indicate egregious tax evasion warranting criminal penalties, file a criminal complaint with the public prosecutor and seek prosecution. The public prosecutor then conducts a fresh investigation and determines whether to indict. After indictment, the case proceeds through the ordinary criminal trial process, and a judgment of conviction or acquittal is rendered.
The trials of legality of the corresponding tax assessments are held by other courts, and the trial procedures are independent and separated from the criminal proceedings. In contrast, the courts try the cases on tax assessments and the relevant administrative penalties at the same time, because the validity of the administrative penalties depends on that of the corresponding assessments.
Even if the taxpayer promptly pays the tax assessed, they generally cannot benefit from any reduction in potential fines for the corresponding tax offence; although such payment may count against them for purposes of mitigation, it merely reflects the taxpayer’s fulfilment of their legal obligation to pay tax.
In general, it is not possible to pay the tax assessed, plus interest and penalties, or enter into an agreement with the tax authorities or public prosecutor to prevent or stop a criminal tax trial. However, if the taxpayer demonstrates sincere remorse, then depending on the circumstances, the case may be disposed of without prosecution, or a lighter penalty may be imposed as a matter of mitigation. The public prosecutor has discretion to decide whether to indict, taking into account the suspect’s personal background and circumstances.
If the taxpayer is dissatisfied with the first-instance judgment, they may appeal to the court of second instance. If they are dissatisfied with the appellate judgment, they may further appeal to the Supreme Court. However, the grounds for a final appeal are limited to constitutional issues or conflicts with judicial precedent.
Most transactions or arrangements challenged under transfer-pricing or anti-avoidance rules do not generally give rise to criminal tax proceedings, because taxpayers engaging in such conduct are not typically deemed to have wilfully evaded tax by fraudulent means. In such cases, taxpayers are usually subject only to administrative penalties, the amounts of which are the same as those imposed for negligence or differences of opinion between taxpayers and the tax authorities.
If double taxation arises from an additional assessment in a cross-border case, it is common to pursue the MAP alongside domestic administrative appeal procedures, although the domestic process may be suspended while MAP negotiations are ongoing. When the competent authorities of the relevant countries reach a mutual agreement under the MAP, the Japanese tax authorities must make corresponding adjustments in accordance with the agreement to eliminate double taxation. If the MAP fails, domestic litigation will resume.
The Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting (MLI) entered into force in Japan in 2019.
In Japan, there is no general or specific anti-avoidance rule – ie, general anti-avoidance rule (GAAR) or specific anti-avoidance rule (SAAR) – that applies to cross-border situations covered by bilateral tax treaties. The principal purpose test (PPT) introduced under the MLI may apply in cross-border cases; however, because the criteria are not clearly defined, it is expected that there will be increased tax controversy over the application of the PPT in the near future. In 2025, the media reported a case in which the tax authorities sought to challenge a tax scheme under the PPT in the tax treaty between Japan and Singapore, but they eventually abandoned the challenge.
Before the OECD/G20 BEPS Project was launched, the tax authorities challenged a cross-border transaction covered by a tax treaty between Japan and Ireland – which did not include the PPT at the time – based on a purposive interpretation of the treaty. Although the authorities argued that they could refuse to apply tax treaties to abusive transactions, the court rejected that argument.
Japan is a proponent of the MAP. Where a treaty contains MAP provisions, an affected taxpayer should consider filing a MAP request in addition to an administrative appeal to the tax authorities or National Tax Tribunal, because the appeal periods are short and strictly enforced. The statute of limitations is not tolled while the MAP is ongoing. The domestic tax controversy procedure may be suspended while the MAP negotiation is ongoing. If the MAP is not available or practical for any reason, Japan’s domestic tax controversy procedures would be the last resort for the taxpayer.
Japan advocates the conclusion of advance pricing arrangements (APAs). To ensure transfer pricing certainty, taxpayers may request the tax authorities to enter into a unilateral or bilateral APA with the relevant foreign tax authority. More than 100 bilateral APAs are concluded each year.
The tax authorities’ APA team is separate from the tax examination team, and, in practice, no information is exchanged between the two teams. However, the APA team will rigorously investigate profits and losses in respect of transactions with foreign related parties.
In respect of a unilateral APA, the APA team applies a more theoretical approach, although they usually request a premium in the range of 1–2%. In bilateral APA negotiations, the APA team tends to adopt a stronger negotiating stance and opening position.
In the field of international taxation in Japan, issues such as WHT, permanent establishment (PE) and transfer pricing have been the subject of tax litigation, but cases involving the CFC rules are the most numerous. This is partly because the tax authorities treat the CFC rules as a priority audit focus with respect to corporations and HNWIs; more importantly, taxpayers often regard themselves as conducting legitimate overseas business activities, yet if the formal conditions for taxation under the CFC rules are met, substantial assessments may be issued, resulting in outcomes that taxpayers find difficult to accept.
Many disputes also involve WHT issues or HNWIs. Issues of individual residence often arise in such controversies. Although individual tax residence is determined under the Civil Code, that code does not provide precise rules or guidance. Tax authorities tend to strictly require payers to verify the recipients’ residence status for WHT purposes, and the courts often uphold the authorities in such cases. Notably, the Supreme Court has held that intent to avoid tax does not affect the determination of a recipient’s residence in a gift tax case. By contrast, residence rarely matters for corporate tax, because a corporation’s residence is determined by the location of its registered office.
To avoid tax litigation, the most important step is to prevent an assessment at the audit stage by fully developing the kinds of arguments and evidentiary showings one would make in court.
There is no applicable information in this jurisdiction.
There is no applicable information in this jurisdiction.
There is no applicable information in this jurisdiction.
There is no applicable information in this jurisdiction.
Some of Japan’s tax treaties include an arbitration clause. Japan has also opted for mandatory binding arbitration under Part VI of the Multilateral Instrument (MLI). However, the domestic legal framework for implementing arbitration remains underdeveloped in many respects.
Japan’s tax treaties do not limit arbitration to specific matters in general. On the other hand, Japan reserves the right to exclude from the scope of Part VI of the MLI with respect to Covered Tax Agreement of Japan cases falling within the provisions of that Covered Tax Agreement, which provide rules for determining whether a person other than an individual shall be treated as a resident of one of the contracting jurisdictions in cases in which that person would otherwise be treated as a resident of both contracting jurisdictions (as they may be modified by the MLI).
In addition, where a reservation made by the other contracting jurisdiction to a Covered Tax Agreement of Japan pursuant to Article 28(2)(a) of the MLI exclusively excludes, whether or not by referring to its domestic law, from the scope of Part VI of the MLI cases of taxation in that other contracting jurisdiction, Japan reserves the right to exclude from the scope of Part VI of the MLI, with respect to that Covered Tax Agreement, cases of taxation in Japan that are analogous to the cases referred to in that other contracting jurisdiction’s reservation.
Japan has traditionally adopted, as a general rule, the independent opinion approach, under which the panel issues a reasoned decision based on legal principles, allowing for a fair, quasi-judicial process, and it has entered a reservation under the MLI against adopting the baseball approach. That said, there are instances – such as under the Japan–USA tax treaty – where arbitration is conducted using the baseball approach.
Japan has long adopted mandatory binding arbitration in line with the OECD approach.
To our knowledge, the arbitration procedure has not been used in practice to date.
The global minimum tax rules under Pillar Two entered into force in Japan in 2024. Initially, only the Income Inclusion Rule (IIR) was enacted; the Undertaxed Profits Rule (UTPR) took effect in 2025, and the qualified domestic minimum top‑up tax (QDMTT) came into effect in April 2026. While the tax authorities have established consultation services for the global minimum tax rules, there are, at present, no specific measures or procedures to enhance certainty or resolve disputes.
Although Japan is working to conclude the multilateral convention on Pillar One Amount A, the outlook remains unclear. Separately, Japan has decided not to implement Pillar One Amount B for the time being.
Japan has concluded implementing arrangements with several jurisdictions under the MLI, and each arrangement includes a provision on confidentiality obligations. These provisions oblige the arbitrators and competent authority officials to maintain confidentiality to the same standard as that required under domestic law and the treaty provisions on the exchange of information.
It appears that the MAP under double tax treaties (DTTs), as modified by the MLI, will be the most common mechanism for resolving tax disputes in Japan. Japan’s major MAP counterparties, such as China, India, South Korea and Germany, have ratified the MLI. However, pre‑MLI DTTs remain important because the United States, Japan’s largest MAP counterparty, has not ratified the MLI.
It is common for taxpayers to retain independent professionals (such as tax accountants and attorneys‑at‑law) when filing a request for the MAP. In later stages of the MAP, competent authority officials conduct the proceedings directly, and independent professionals are generally no longer involved.
There are no filing fees payable to the tax authorities or National Tax Tribunal for first and second-tier administrative appeals.
When initiating tax litigation, taxpayers must pay court filing fees to the court. The amount varies depending on the amount in dispute. Separate filing fees are also required when filing an appeal and when filing a final appeal. As a general rule, court costs are borne by the losing party. Accordingly, if a taxpayer brings tax litigation and prevails, the filing fees the taxpayer paid are generally refunded, but no interest is paid.
If an assessment is revoked, the taxpayer is entitled to a refund of the additional tax paid together with refund interest. The refund interest rate varies by year; for 2026 it is 1.3% per annum.
In Japan, no fees are charged for filing a MAP request or for pre-filing consultations.
In FY 2025, 1,447 first-tier administrative appeals and 3,537 second-tier administrative appeals were commenced, and 196 tax litigation cases were filed (97 at first instance, 57 on appeal and 42 on final appeal).
Of the 196 tax litigation cases commenced in FY 2025, 67 were related to income tax, 37 to corporation tax, 29 to asset taxes, 25 to consumption tax, eight to other taxes, 22 to collection matters and eight to tribunal-related matters.
Of the 1,752 first-tier administrative appeal cases concluded in FY 2025, the taxpayer’s claims were upheld in whole or in part in 91 cases (5.2%).
Of the 3,872 second-tier administrative appeal cases concluded in FY 2025, the taxpayer’s claims were upheld in whole or in part in 693 cases (17.9%).
Of the 168 tax litigation cases concluded in FY 2025, the taxpayer’s claims were upheld in whole or in part in eight cases (4.8%).
There are two major categories of issues in tax controversy: fact-finding and legal interpretation. In most administrative appeal cases where the National Tax Tribunal reversed an assessment, it did so because it disagreed with the facts assumed and presented by the tax authorities. Fact-finding is an objective exercise and, by relying on objective facts, the National Tax Tribunal functions as a neutral arbiter of disputes between taxpayers and the tax authorities.
However, the National Tax Tribunal rarely rejects the tax authorities’ legal interpretations, possibly because it is mindful of the wider impact an interpretation may have on other taxpayers. In such cases, if it upholds the assessment, the taxpayer may bring the case to court. By engaging in tax litigation, the taxpayer also may address any legal interpretations they disagree with, in the hope that the court will reject these interpretations.
That said, tax disputes take time, money, and effort. No taxpayer welcomes a tax controversy. Even if one cannot avoid a dispute, one will want to resolve it as quickly as possible. Accordingly, during the tax audit – the de facto forum in which tax controversies are fought in Japan – the best strategy is not merely to take a passive stance and answer the examiner’s questions, but rather to proactively present the taxpayer’s position with supporting evidence, persuade the tax authorities and prevent the issuance of a tax assessment.
Yodoyabashi Mitsui Building 5F
4-1-1 Imabashi
Chuo-ku
Osaka
Japan
+81 6 7711 2540
+81 6 7669 9640
dtlegal@tohmatsu.co.jp www.deloitte.com/jp/dtlegal
Introduction
In Japan, despite the introduction of the global minimum tax, complex controlled foreign company (CFC) rules remain in place, and no sweeping reforms to fundamentally simplify the CFC rules have yet been undertaken.
Japanese CFC rules are primarily based on an entity approach, under which the income, in full, of foreign subsidiaries is deemed the income of the Japanese parent company on an entity-wide basis. Following the 2017 tax reform undertaken in response to the BEPS Project Final Report (Action 3), two categories have been adopted within the entity approach:
However, inclusion is not required where the foreign subsidiary’s tax burden ratio is 20% or higher for category (i) or 27% or higher for category (ii). Roughly speaking, the tax burden ratio is the ratio of the amount of corporate income tax actually imposed (numerator) to income (denominator), with the income in the denominator including tax‑exempt income. Therefore, when tax‑exempt income is substantial, the tax burden ratio may be significantly lower than that country’s effective tax rate.
Meanwhile, even where full entity-level inclusion under the entity approach does not apply, passive income – such as dividends, interests and royalties meeting specified conditions – is subject to partial item-of-income inclusion (the threshold for exemption from partial inclusion is a tax burden ratio of 20% or higher), making Japanese CFC rules a complex system that combines the entity approach and the income approach.
Japanese CFC rules present a number of issues, chief among them being:
Ambiguous Statutory Language
Overview
As noted above, foreign subsidiaries that fail to satisfy even one of the economic activity tests become subject to company-level income inclusion. The economic activity tests consist of the following individual tests: the business test, the substance test, the management and control test, and either the unrelated-party test or the location-country test.
If the company’s principal business falls within one of eight enumerated businesses, namely wholesale, banking, trust, financial instruments, insurance, water transportation, air transportation or goods lending business – limited to those whose principal business is the lending of aircraft – the unrelated-party test applies; otherwise, the location-country test applies.
In this regard, the relevant legislation (the Act on Special Measures Concerning Taxation) sets out highly abstract statutory language for the economic activity test to be satisfied by the foreign subsidiary, as follows:
As each of the economic activity tests includes statutory language whose scope and reach are not clear, differences in interpretation between the taxpayers and the tax authorities easily arise, and numerous tax cases have been litigated over whether the individual tests are satisfied. It should be noted that prior to the FY 2017 tax reform, the economic activity test was referred to as the “exemption test”, but the contents of each test under the exemption test were carried over to the current economic activity test with little change.
Supreme Court ruling in 2017 on regional headquarters case
By way of example, in relation to the business test, how to interpret and apply highly abstract statutory terms such as “principal business” and “holding of shares” can be more difficult than one might expect. In this regard, there is a well-known tax case involving a subsidiary in Singapore that, while holding the shares of its second-tier subsidiaries in various Association of Southeast Asian Nations (ASEAN) countries, also performed regional headquarters functions for those subsidiaries. The dispute, relating to the period when the exemption test was in effect, centred on whether the foreign subsidiary’s principal business was the regional headquarters business or shareholding business.
In the appellate decision, the court held that regional headquarters operations were subsumed within the shareholding business and concluded that the foreign subsidiary engaged solely in the shareholding business, ruling against the taxpayer by upholding income inclusion under the CFC rules.
The Supreme Court, however, fundamentally overturned that approach in 2017. First, on the premise that regional headquarters operations have purposes, content and functions distinct from activities related to the exercise of shareholder rights or investment management, it held that such operations are not encompassed within the shareholding business. It then proceeded on the basis that the foreign subsidiary engaged in multiple businesses – namely, a regional headquarters business and shareholding business – and, upon comprehensive consideration of various factors such as the revenue and income attributable to each business and the actual state of its business activities, concluded that the company’s principal business was the regional headquarters business. Consequently, it found that the business test was satisfied and, reversing the appellate court ruling, ruled in favour of the taxpayer.
Issues and lessons arising from the Supreme Court ruling
There are at least two lessons to be drawn from this Supreme Court decision. First, it should be kept in mind that it is not an easy task to ascertain the scope of highly abstract statutory terms of tax law such as “holding of shares” by relying on their wording alone. In fact, the Supreme Court, without attempting to define the term “holding of shares”, narrowed its focus to whether the regional headquarters operations are encompassed within the shareholding business, and – taking into account purposes, content, functions of regional headquarters and consistency with the purpose of the business test – merely concluded that they are not encompassed.
Second, where a foreign subsidiary engages in multiple businesses, determining which is its principal business is likewise not straightforward. The settled standard – also applied in this Supreme Court decision – is to make that determination holistically, taking into account such factors as the amounts of revenue or income derived from each business activity at the subsidiary, the number of employees required for those activities and the state of its offices, stores, factories and other fixed facilities.
That said, even under a holistic assessment approach, deciding how to balance quantitative factors against operational factors is a difficult judgment. The Supreme Court, based on detailed findings of fact, recognised that the subsidiary’s regional headquarters business was of substantial scale and substance and constituted a significant portion of its business activities, and the taxpayer prevailed. However, in cases unlike this one where the revenue attributable to businesses other than the shareholding business is relatively small, it is difficult for taxpayers to predict in advance whether the business test is satisfied, even if most operational factors such as employees and fixed facilities are put into such active business.
Moreover, as a result of tax reform in 2010, a special rule for headquarters companies has been established within the business test. Consequently, even if a company’s principal business is shareholding, it can still satisfy the business test provided it meets the requirements of the Headquarters Company Special Rule (the Supreme Court case discussed above involved fiscal years prior to the application of this special rule).
In light of the FY 2010 tax reform and the aforementioned Supreme Court decision, there are theoretically two avenues for satisfying the business test:
It is undeniable that this situation makes practical decision-making somewhat complex.
Other related issues and key practical points
Ambiguity in the statutory language is not limited to the business test; for other components of the economic activity test as well, it is not always possible to determine the interpretive scope of statutory language solely from its wording. For instance, wording such as “offices considered necessary to conduct the principal business” used in the substance test has unclear parameters.
A lower court ruling in another case showed that whether the fixed facilities used by a foreign subsidiary meet the requisite scale under the substance test should be assessed in light of the industry and form of its principal business, and further clarified that even rental office space can meet the requisite scale. This ruling is helpful in that it clarifies the criteria for interpreting the above wording to some extent. However, the mere presence of a rental office does not always satisfy the substance test. It should be noted that interpretive ambiguity still remains because the relevant facts must be assessed in detail on a case-by-case basis, taking into account the nature and form of the principal business.
In this regard, the National Tax Agency (NTA) has published a Q&A compiling questions and typical scenarios concerning Japanese CFC rules and setting out its views on interpreting the relevant requirements, such as the substance test and the management and control test. For example, the above Q&A explains the scope of “the fixed facilities considered necessary to conduct the principal business” in the substance test by referring to specific scenarios with clear answers. Furthermore, with respect to the management and control test, it explains, with reference to specific situations, what is meant by “undertaking the management, control, and operation of the business itself”. These explanations can be useful in practice to understand how the tax authorities interpret the relevant statutory language of CFC rules.
Formalistic Application of the Text
Overview
There is risk in adopting practices that clearly depart from the text and structure of the statutes and regulations governing the CFC rules. In other words, even in cases where, in light of the legislative purpose and the substance of the facts, it feels counterintuitive from a common-sense perspective to subject the taxpayer to entity-level inclusion, the tax authorities tend to impose inclusion by applying the statutory language of the CFC rules in an extremely formalistic manner.
By contrast, taxpayers sometimes defend by arguing – departing from the statutory text – that a purely literal application would be inconsistent with the legislative purpose of the CFC rules. In this regard, while courts may, as noted above, construe ambiguous wording in a manner consistent with the legislative purpose (which falls within a form of textual interpretation), they tend to reject taxpayer’s arguments that rely on the legislative purpose alone to deviate from the statutory text.
Supreme Court ruling in 2023 on Megabank case
For example, in one case, Cayman subsidiaries received repayment of a subordinated loan from its Japanese parent and used those funds to redeem its preferred equity securities held by a related group company. As a result, at the end of the relevant fiscal year, the only outstanding shares of Cayman subsidiaries were the common shares held by the Japanese parent, which did not expect to receive any dividends from them during that fiscal year. The tax authorities, taking the parent’s year-end shareholding ratio to be 100%, issued an assessment under the CFC rules, and the taxpayer brought a suit seeking its revocation. The appellate court, taking into account the fact that the Japanese parent lacked control over the foreign subsidiaries’ current-year net income, concluded there was no amount subject to inclusion under the basic structure and principle of the CFC rules and ruled for the taxpayer.
In 2023, the Supreme Court, however, held that the relevant provisions of the Order for Enforcement of the Act on Special Measures Concerning Taxation, which uniformly set the timing for measuring the shareholding ratio as the close of the foreign subsidiary’s fiscal year, are reasonable from the standpoint of ensuring stability in tax administration and, by applying the statutory language in a formalistic manner, reversed the appellate court decision. In this case, there is an intuitive sense that it is unduly harsh to impose CFC inclusion where the parent had no control over the subsidiary’s net income, and the 100% shareholding ratio occurred merely because the measuring point happened to be the end of year.
Nevertheless, the Supreme Court deigned to indicate that there was room for tax planning, such as setting the foreign subsidiary’s fiscal year to end on the day before the redemption date of the preferred equity securities so that the amount subject to inclusion would be zero. Thus, the Supreme Court suggested an approach premised on formal application of the statutory text. In cases like this, where the statutory wording is not ambiguous and is facially clear, it is often not easy in tax litigation to overcome the text by appealing to legislative purpose.
Unreasonably strict procedural requirements
Another issue with the CFC rules is that certain procedural requirements are unreasonably strict in light of the original legislative purpose and impose an excessive burden on taxpayers. For example, even when a foreign subsidiary’s income is subject to company-level inclusion under the CFC rules, the dividends it receives from its second-tier subsidiaries are deducted from the inclusion amount. This mechanism is intended to align with the introduction, in the 2009 tax reform, of a system under which dividends paid by qualifying foreign subsidiaries to their Japanese parent company are excluded from taxable income, and it has played an important role in curbing the amount subject to inclusion under the CFC regime. To claim this deduction, however, the Order for Enforcement of the Act on Special Measures Concerning Taxation requires that prescribed documents be attached to the final tax return, unless there are unavoidable circumstances surrounding the failure to attach them.
In practice, at the filing stage, taxpayers may conclude that the CFC rules do not apply because they believe the economic activity test is satisfied, and therefore they may omit the documentation for the dividend deduction. If a subsequent tax audit determines that the economic activity test is in fact not satisfied, a strictly formalistic application of the law would lead to full company-level inclusion without deducting those dividends from second-tier subsidiaries, simply because the prescribed documents were not attached to the return.
It can hardly be called a reasonable system to deny the aforesaid deduction from the inclusion amount solely because the taxpayer’s initial self‑assessment at the time of filing was found to be mistaken. However, taxpayers should be mindful of the current risk that procedural requirements with little policy justification may be applied mechanically. It is to be hoped that tax litigation or legislative amendment will improve the situation.
Practical Measures Under the Absence of Fundamental Overhaul of the CFC Rules
As noted above, the statutory wording of each element of the economic activity test is highly abstract, making the task of determining whether the requirements are satisfied burdensome. In a 2022 study group report on the global minimum tax and the CFC rules, which was led by the Ministry of Economy, Trade and Industry, some participants argued that the economic activity test should be abolished or significantly simplified. Unfortunately, that approach has not been realised through tax reform.
Meanwhile, in addition to the Headquarters Company Special Rule mentioned above, the tax reform introduced various other special rules – such as the Aircraft Leasing Company Special Rule related to the business test and the Manufacturing Special Rule related to the location‑country test. Depending on the case, it is important to make effective use of these special rules to satisfy the economic activity test. In particular, the Manufacturing Special Rule deems the location‑country test satisfied where a company is actively involved in manufacturing through significant functions performed in the country where its head office is located.
Given that, prior to the introduction of this rule, there were numerous tax cases related to “processing with supplied materials” in which taxpayers lost on the grounds that their foreign subsidiaries did not actually conduct manufacturing in the head‑office country (Hong Kong) and therefore did not satisfy the location‑country test, this was a groundbreaking reform. However, it should also be noted that such exceptions to the location‑country test are applied only to the manufacturing sector.
Where a foreign subsidiary’s business realities are complex and it is not easy to determine whether the CFC rules apply, one potential strategy is to submit an advance confirmation request to the tax authorities to first obtain their official views. In practice, there have been cases where, prior to executing M&A or corporate reorganisations, taxpayers sought such confirmation as to whether Japanese CFC rules would trigger taxation.
In one recently reported case, the tax authorities responded to a confirmation request that, if the planned acquisition went ahead, inclusion would be required under the CFC rules, and the acquisition scheme was finally abandoned by the taxpayer. This approach can be commended insofar as it proactively manages CFC tax risk in advance. At the same time, it reveals a facet of the CFC rules that could impede cross‑border M&A.
To avoid being blindsided by an unexpected CFC assessment, there is an urgent need to clarify the relevant statutory language (including the introduction of clear rules tailored to cross‑border M&A) and to further enhance guidelines such as the NTA’s Q&A. In any event, a key task going forward is to build a tax framework that does not unduly chill cross‑border transactions, including M&A.
Lastly, if the global minimum tax and the CFC rules are to continue to coexist, appropriate legislative action will be necessary to harmonise the two rules and reduce taxpayers’ overall compliance burden. This has not yet been achieved and remains an issue for future reform.
Yodoyabashi Mitsui Building 5F
4-1-1 Imabashi
Chuo-ku
Osaka
Japan
+81 6 7711 2540
+81 6 7669 9640
dtlegal@tohmatsu.co.jp www.deloitte.com/jp/dtlegal