全球最低税经济影响评估
2026年更新

ECONOMIC IMPACT
ASSESSMENT OF THE
GLOBAL MINIMUM TAX
2026 UPDATE

2026年7月15日
15 July 2026

刘剑波2026年7月16日编译（OECD ECONOMIC IMPACT ASSESSMENT OF THE GMT-2026-update OECD的网络研讨会材料）

今日网络研讨会——注意事项

Today’s webinar – Housekeeping

记录

Recording

• 本次网络研讨会将被录制，英文版将在OECD网站上发布。

• This webinar is being recorded and will be made available in English on the OECD website.

问答

Q&As

• 聊天功能已被禁用，请使用问答功能提交您的问题。

• The chat function has been disabled, please submit your questions using the Q&A function.

文件

Documents

• 如需获取全球最低税经济影响评估的内容，请访问OECD网站。

• To access the content of the Global Minimum Tax Economic Impact Assessment, visit the OECD website.

今日网络研讨会——演讲嘉宾

Today’s webinar – Speakers

Kurt Van Dender——税收政策与统计司司长，税收政策与管理中心（CTP）

Kurt Van Dender — Head of Tax Policy and Statistics Division, CTP

Pierce O’Reilly——税收政策与统计司副司长，企业与国际税务部门负责人，CTP

Pierce O’Reilly — Deputy Head of Tax Policy and Statistics Division, Head of Business and International Taxes Unit, CTP

Felix Hugger——经济学家，企业与国际税务部门，CTP

Felix Hugger — Economist, Business and International Taxes Unit, CTP

Laura Arnemann——初级经济学家，企业与国际税务部门，CTP

Laura Arnemann — Junior Economist, Business and International Taxes Unit, CTP

今日网络研讨会——议程

Today’s webinar – Agenda

1. 全球最低税（GMT）的GloBE规则概述、并行方案及迄今为止的影响评估工作

1. Overview of GloBE Rules, Side-by-Side Package and impact assessment work of the Global Minimum Tax (GMT) to date

2. 并行方案后经济影响评估（EIA）模型的更新

2. Update of the Economic Impact Assessment (EIA) Model after the Side-by-Side Package

3. 实施后数据的早期结果

3. Early results from post-implementation data

4. 问答

4. Q&A

主要结果

Key results

EIA模型更新

Update to the EIA model

据估计，在辖区层面，有效税率（ETR）平均将上升2.8–3.7个百分点，投资中心的税率将上升5.5–6.9个百分点。

Effective tax rates (ETR) are estimated to rise by 2.8 – 3.7 percentage points on average at the jurisdiction level, rates in investment hubs rise by 5.5 – 6.9 percentage points.

据估计，ETR差异将下降19.0–25.0%，有利于改善资本配置。

ETR differentials are estimated to fall by 19.0 – 25.0%, which is conducive to improved allocation of capital.

据估计，全球最低税将大幅减少利润转移，预计减少幅度在22.6%–44.6%之间。

The GMT is estimated to reduce profit-shifting substantially with an estimated reduction of between 22.6 – 44.6%.

据估计，全球税收收入将增长3.2%–5.4%，即每年910亿至1550亿美元。

Global tax revenues are estimated to rise by 3.2 – 5.4% or USD 91 – 155 bn per year.

早期事后分析

Early ex-post analysis

2024年财务数据的早期事后分析表明，全球最低税对ETR产生了积极影响，没有证据表明对投资和就业产生负面影响。

Early ex-post 2024 financial data suggest positive ETR impacts of the GMT with no evidence of negative impact on investment and employment.

背景

BACKGROUND

全球最低税的主要要素

Main elements of the GMT

由GloBE规则确立，该规则基于调整后的财务会计数据定义了共同的税基和有效税率。

Established by the GloBE Rules defining a common tax base and effective tax rate based on adjusted financial accounting.

适用于年收入超过7.5亿欧元的跨国企业（MNEs）。

Applies to multinational enterprises (MNEs) with revenues above EUR 750 million.

作为补足税运作，按辖区逐个达到15%的最低税率，并对某些类型的税收优惠进行特殊处理。

Works as a top-up tax to reach a 15% minimum tax rate at a jurisdiction-by-jurisdiction basis, with specific treatment of certain kinds of tax incentives.

适用于扣除基于工资和有形资产的实质性收入排除（SBIE）以及附加安全港后的超额利润。

Applies to excess profits after applying the substance-based income exclusion (SBIE) based on payroll and tangible assets and additional safe harbours.

确立了征收补足税的规则顺序：1. QDMTT（本地征收）；2. IIR（最终母公司）；3. UTPR（其他辖区）。

Establishes a rule order to collect top-up taxes: 1. QDMTT (locally); 2. IIR (Ultimate Parent Entity); 3. UTPR (other jurisdictions)

评估全球最低税：时间线

Assessing the GMT: a timeline

2019年1月：提出作为解决数字经济更广泛税收挑战的共识方案的双支柱方案。

01/2019: Proposed approach as consensus-based solution to broader digital tax challenges

2015年10月：发布BEPS最终报告。

10/2015: Release of Final BEPS Action Reports

2020年10月：发布第二支柱蓝图报告及2020年经济影响评估报告。

10/2020: Release of Pillar Two Blueprint Report and 2020 EIA Report

2021年10月-12月：发布双支柱方案声明，随后发布GloBE示范规则；更新全球EIA结果。

10/2021-12/2021: Statement on Two-Pillar Solution and subsequent publication of GloBE Model Rules; Updated global EIA results published

2023年11月-2024年1月：发布全球低税利润论文，以及2024年EIA模型（网络研讨会）。

11/2023-01/2024: Paper on global low-taxed profit, GMT and Webinar (2024 EIA Model)

2023年1月：全球最低税开始实施。

01/2023: Start of GMT implementation

2024年1月：发布更新后的全球结果（网络研讨会）。

01/2024: Updated global results published (Webinar)

2026年1月：就并行方案达成一致。

01/2026: Agreement on the Side-by-Side Package

2026年7月：纳入并行方案后的更新结果。

07/2026: Updated results incorporating Side-by-Side Package

注：全球最低税实施过程中选定里程碑的高层次总结。

Note: High-level summary of selected milestones towards the implementation of the GMT.

全球最低税最新进展

Latest GMT developments

实施情况

Implementation

约55个辖区将在2024-2025年实施。

~55 implementing jurisdictions for 2024-2025

• 列入中央记录。

• Listed in the Central Record

• 实施可涉及QDMTT、IIR和/或UTPR。

• Implementation can involve QDMTTs, IIR and/or UTPR

其他进展

Other developments

• 部分辖区在2026年实施了DMTT。

• Some jurisdictions implemented a DMTT in 2026

• 部分辖区选择根据欧盟指令推迟实施。

• Some jurisdictions opted for deferred implementation under EU directive

• 部分辖区选择了其他改革形式（如重新审视税收优惠或15%企业所得税率）。

• Some jurisdictions chose other forms of reform (e.g. revisit tax incentives or 15% CIT rate)

全球最低税最新进展

Latest GMT developments

并行方案

The Side-by-Side Package

并行方案及UPE安全港

Side-by-Side and UPE Safe Harbour

• 容纳实现等效结果的不同最低税制度。

• Accommodates different systems of minimum taxation that achieve equivalent outcomes

• 限制IIR/UTPR对总部位于合资格辖区MNEs的适用。

• Limits the application of IIR/UTPR to MNEs headquartered in eligible jurisdictions

• 合资格辖区由包容性框架协商一致确定。

• Eligible jurisdictions determined by Inclusive Framework consensus

• QDMTT不受影响，在合资格并行方案制度下予以抵免。

• QDMTTs remain unaffected and are credited under eligible Side-by-Side regimes

简化ETR安全港

Simplified ETR Safe Harbour

• 辖区层面计算，模块化设计。

• Jurisdictional level computations with modular design

• 基于当前税收/收入的简化ETR计算——仅有限调整。

• Simplified ETR calculation based on current taxes / income – with limited set of adjustments

基于实质性税收优惠（SBTI）安全港

Substance-based Tax Incentive (SBTI) Safe Harbour

• 将合格税收优惠加回已覆盖税款，金额等于财政年度使用的合格税收优惠。

• An add-back to covered taxes equal to qualified tax incentives used in the fiscal year

• 合格税收优惠（QTIs）一般为普惠性支出型或生产型抵免/扣除。

• Qualified tax incentives (QTIs) are generally available expenditure or production-based credits / deductions

• 上限等于工资或折旧的5.5%（或可折旧资产的1%）。

• Cap equal to 5.5% of payroll or depreciation (or 1% of depreciable assets)

盘点

Stocktake

• 基于证据的盘点将于2028-2029年进行。

• Evidence-based stocktake to take place in 2028-2029

EIA模型更新

UPDATE TO THE EIA MODEL

与前一版EIA模型发布的比较

Comparison to previous EIA Model release

2024年EIA模型发布版：数据2017-2020；SBIE第1年；当前实施（并行方案前）；仅纳入GILTI收入；利润转移未考虑美国CFC制度。

2024 EIA Model Release: Data 2017-2020; SBIE Year 1; Current implementation; Before Side-by-Side Package (i.e. no SBTI and Side-by-Side Safe Harbour modelled); Only GILTI revenues included; Profit-shifting not accounting for US CFC regime.

2026年EIA模型更新版：数据2019-2022*；SBIE第10年；当前实施（2026年6月）；纳入并行方案安全港和SBTI安全港；纳入NCTI、Subpart F和分支税收；更新利润转移建模以考虑美国CFC制度。

2026 EIA Model Update: Data 2019-2022*; SBIE Year 10; Current implementation (June 2026); Side-by-Side Safe Harbour; SBTI Safe Harbour; NCTI, Subpart F and Branch taxation included; Updated modelling of profit-shifting to account for US CFC regime.

* 由于更新了外推和CbCR数据，当前数据与之前影响评估中使用的同年份数据不完全相同。

* Due to updated extrapolations and updated CbCR data, current data is not identical to data used in previous impact assessment for the same years

估算全球最低税影响的更新步骤

Updated steps to estimate GMT impacts

步骤1：基于汇总的国别报告（CbCR）数据，辅以其他数据源（”数据矩阵”，参见Hugger等，2024），估算各辖区内的ETR分布。

Step 1: Estimate ETR distributions within jurisdictions based on aggregated Country-by-Country Reporting (CbCR) data, complemented with additional data sources (“data matrices”, see Hugger et al., 2024)

更新：整合SBTI安全港对ETR的影响。

Updates: Integrate impact of SBTI Safe Harbour on ETRs

更新：基于QTI可用性和使用情况的不同假设估算情景（关于QTI可用性和使用的信息有限，增加了不确定性）。

Updates: Estimate scenarios with varying assumptions on QTI availability and use (limited information on QTI availability and use increases uncertainty)

步骤2：基于汇总的CbCR数据，辅以其他数据源，估算辖区对层面的全球利润、资产和工资分布。

Step 2: Estimate global distribution of profit, assets and payroll at the jurisdiction pair level based on aggregated CbCR data, complemented with additional data sources

更新：更新数据以整合最新可用的CbCR和其他数据（2019-2022）。

Updates: Update data to integrate latest available CbCR and other data (2019-2022)

更新：考虑对SBTI和并行方案安全港的行为反应，以估算利润转移的变化。

Updates: Account for behavioural responses to the SBTI and Side-by-Side Safe Harbour to estimate changes in profit-shifting

步骤3：在辖区层面估算需缴纳补足税的利润份额。

Step 3: Estimate share of profits subject to top-up tax at a jurisdictional level

更新：在估算中应用考虑SBTI安全港的调整后GloBE ETR。

Updates: Apply adjusted GloBE ETRs accounting for SBTI Safe Harbour in the estimation

更新：在补足税计算中考虑并行方案安全港。

Updates: Account for Side-by-Side Safe Harbour on the calculation of top-up taxes

步骤4：估算各实施辖区的补足税分布。

Step 4: Estimate distribution of top-up taxes across implementing jurisdictions

更新：更新实施假设以反映当前全球最低税实施情况——不考虑潜在的未来实施或撤销；为说明目的建模了”全球实施”情景（并行方案制度除外）。

Updates: Update implementation assumptions to reflect current GMT implementation – No potential future implementation or rollback considered; Scenario of ‘global implementation’ (except for Side-by-Side regimes) modelled for illustrative purposes

并行方案安全港建模

Modelling the Side-by-Side Safe Harbour

• 迄今为止，美国是中央记录中唯一被列为拥有合资格并行方案制度的国家。

• To date, the United States is the only country listed on the Central Record as having an eligible Side-by-Side regime

• 美国MNEs的利润仍受美国补足税制度（NCTI、Subpart F、分支税收）约束，但不受UTPR或IIR约束。

• Profits of US MNEs remain subject to the US system of top-up taxation (NCTI, Subpart F, Branch Taxation), but are not subject to the UTPR or IIR

• QDMTT继续对美国MNEs的外国利润征收，并基于当前实施进行建模。

• QDMTTs continue to be levied on foreign profits of US MNEs and are modelled based on current implementation

• 并行方案安全港可能影响利润转移激励。

• Side-by-Side Safe Harbour may affect profit-shifting incentives

– 假设美国MNEs减少向已实施QDMTT的投资中心转移利润。

– US MNEs are assumed to reduce profit-shifting towards investment hubs that have implemented a QDMTT

– 建模了从QDMTT中心向非QDMTT中心转移利润的部分重新路由。

– Some re-routing of shifted profit from QDMTT hubs towards non-QDMTT hubs modelled

– 对于非美国MNEs，利润转移反应与之前建模保持不变。

– For non-US MNEs, profit-shifting responses unchanged compared to previous modelling

SBTI安全港建模

Modelling the SBTI Safe Harbour

• 在SBTI安全港下，QTIs可加回已覆盖税款，上限为工资或折旧的5.5%（或可折旧资产的1%）。

• Under the SBTI Safe Harbour, QTIs can be added back to covered taxes up to a cap (5.5% of payroll or depreciation or 1% of depreciable assets)

• 没有关于QTI可用性和QTI使用的综合数据。

• No comprehensive data on QTI availability and QTI use

– 案头研究提供了各辖区QTI可用性的部分信息。

– Some information on QTI availability per jurisdiction informed by desk-research

– OECD及伙伴国家的研发税收优惠使用数据有限。

– Limited data on use of R&D tax incentives available for OECD and partner countries

• 鉴于不确定性，基于上限建模了两种情景。

• Given the uncertainty, two scenarios are modelled based on the cap

1. 上限情景：在已识别QTIs的辖区（222个辖区中的114个），已覆盖税款增加上限的10%。

1. Upper Bound Scenario: Covered taxes increase by 10% of cap in jurisdictions with QTIs identified (114 out of 222 jurisdictions)

2. 下限情景：在所有辖区，已覆盖税款增加上限的60%。

2. Lower Bound Scenario: Covered taxes increase by 60% of cap in all jurisdictions

• 建模考虑修订后QTI处理的利润转移反应。

• Model profit-shifting responses accounting for the revised QTI treatment

注意事项

Caveats

一般注意事项

General caveats

• EIA模型高度汇总，因此掩盖了MNE特定结果。

• EIA Model highly aggregated and therefore obscures MNE-specific outcomes

• EIA模型主要基于CbCR数据，该数据与GloBE税基存在重要差异。

• EIA Model largely based on CbCR data which has important differences to the GloBE tax base

• 预期会有各种动态效应，但仅部分建模。

• Various dynamic effects are to be expected, but only partially modelled

与并行方案安全港相关的注意事项

Caveats relating to the Side-by-Side Safe Harbour

• 建模现有美国制度非常具有挑战性，因为仅有汇总数据可用，且与潜在GloBE税收的确切互动在数据中不明确。

• Modelling the existing US system very challenging as only aggregated data available and exact interaction with potential GloBE taxes unclear in the data

• GloBE和NCTI/Subpart F/分支税基不同；在EIA模型中只能近似评估这些详细差异。

• GloBE and NCTI/Subpart F/Branch tax bases are different; assessing these detailed differences can only be approximated in the EIA Model

与SBTI安全港相关的注意事项

Caveats relating to the SBTI Safe Harbour

• 模型不反映企业异质性。

• Model does not reflect firm heterogeneity

– 需要对QTI使用做出强假设，因为在CbCR数据中无法观察到。

– Strong assumptions on use of QTIs need to be made, since it is not observed in CbCR data

– 如果实质和QTI使用集中于低税企业，则收入收益被高估。

– If substance and QTI use are concentrated amongst low-taxed firms, revenue gains are overestimated

– 如果实质和QTI使用集中于高税企业，则收入收益被低估。

– If substance and QTI use are concentrated amongst high-taxed firms, revenue gains are underestimated

• 数据无法区分低ETR的原因（抵免、补贴、其他）。

• Data cannot distinguish the cause of the low ETRs (credits, allowances, other)

• 建模无法捕捉QTI上限是否被用尽。

• Modelling cannot capture whether the QTI cap is exhausted

• 由于建模所用的CbCR数据不包含折旧信息，假设资产账面折旧率为12.5%。

• Since CbCR data used for the modelling does not contain information on depreciation, it is assumed that book depreciation of assets is equal to 12.5%

– 该关系由Orbis的企业级汇总数据提供依据。

– This relationship is informed by consolidated firm-level data from Orbis

 

考虑的关键结果

Key outcomes considered

1. 对有效税率的影响

1. Impact on effective tax rates

2. 对ETR差异的影响

2. Impact on ETR differentials

3. 对利润转移的影响

3. Impact on profit-shifting

4. 对税收收入的影响

4. Impact on tax revenue

 

1. 对有效税率的影响

1. Impact on effective tax rates

按全球最低税前ETR估算的ETR增长。

Estimated increase in ETRs by pre-GMT ETR

• 估算表明大多数辖区的ETR有所上升。

• Estimates suggest increases in ETRs in most jurisdictions

• 全球最低税前ETR较低的辖区ETR平均上升幅度更大。

• ETRs increase more in jurisdictions with lower pre-GMT ETRs on average

• 全球最低税前平均ETR超过15%的辖区影响有限。

• Limited effects in jurisdictions with pre-GMT average ETRs exceeding 15%

• SBTI安全港使ETR增长更集中于低实质辖区。

• SBTI Safe Harbour concentrates ETR increases more in jurisdictions with low substance

注：估算基于2019-2022年数据。估算考虑了利润转移敏感性和QTI可用性及使用的变化。估算基于GloBE规则和QDMTT的当前实施，DMTT或其他形式的实施。估算以第10年SBIE情景（工资5%和有形资产5%）呈现。辖区分组基于世界银行分类。投资中心是FDI与GDP之比超过150%的辖区。

Note: The estimates are presented based on 2019-2022 data. The estimates account for the variation in the sensitivity of profit-shifting and QTI availability and use. Estimates are based on the current implementation of the GloBE Rules and QDMTTs, implementation of DMTTs or other forms of implementation. Estimates are presented for the year-ten SBIE scenario (5% on payroll and 5% on tangible assets). Jurisdiction groups are based on World Bank classifications. Investment hubs are jurisdictions with an FDI-to-GDP ratio of 150%.

2. 对ETR差异的影响

2. Impact on ETR differentials

• 由于ETR在全球最低税前ETR较低的辖区增长最多，全球最低税减少了ETR的离散程度。

• As ETRs increase most in jurisdictions with low pre-GMT ETRs, the GMT reduces the dispersion of ETRs

• 全球最低税后的ETR分布受并行方案（SBTI安全港和并行方案安全港）影响有限。

• Post-GMT distribution of ETR changes modestly impacted by Side-by-Side Package (SBTI SH and SbS SH)

– 全球最低税后ETR集中度低于之前估算中的15%。

– Post-GMT ETRs are less concentrated at 15% than in previous estimates

– 这主要是由于引入了SBTI安全港。

– This is largely due to the introduction of the SBTI Safe Harbour

• 税率差异缩小可能促进全球资本更有效配置。

• Reduced tax rate differentials may lead to a more efficient allocation of capital globally

注：估算基于2019-2022年数据。估算考虑了利润转移敏感性和QTI可用性及使用的变化。估算基于GloBE规则和QDMTT的当前实施。估算以第10年SBIE情景呈现。辖区分组基于世界银行分类。投资中心是FDI与GDP之比超过150%的辖区。

Note: The estimates are presented based on 2019-2022 data. The estimates account for the variation in the sensitivity of profit-shifting and QTI availability and use. Estimates are based on the current implementation of the GloBE Rules and QDMTTs. Estimates are presented for the year-ten SBIE scenario. Jurisdiction groups are based on World Bank classifications. Investment hubs are jurisdictions with an FDI-to-GDP ratio of 150%.

3. 对利润转移的影响

3. Impact on profit-shifting

• 估算表明，全球最低税继续显著减少利润转移激励。

• Estimates suggest that the GMT continues to significantly reduce profit-shifting incentives

• 由于利润转移减少，投资中心预计将税基流失给非中心。

• Investment hubs are estimated to lose tax base to non-hubs due to reduced profit-shifting

• 利润转移也受SBTI安全港影响，该安全港在EIA模型中提高了非中心相对于中心的竞争力。

• Profit-shifting impacted also by SBTI Safe Harbour, which increases the competitiveness of non-hubs relative to hubs in the EIA model

注：估算基于2019-2022年数据。估算考虑了利润转移敏感性和QTI可用性及使用的变化。估算基于GloBE规则和QDMTT的当前实施。估算以第10年SBIE情景呈现。辖区分组基于世界银行分类。投资中心是FDI与GDP之比超过150%的辖区。

Note: The estimates are presented based on 2019-2022 data. The estimates account for the variation in the sensitivity of profit-shifting and QTI availability and use. Estimates are based on the current implementation of the GloBE Rules and QDMTTs. Estimates are presented for the year-ten SBIE scenario. Jurisdiction groups are based on World Bank classifications. Investment hubs are jurisdictions with an FDI-to-GDP ratio of 150%.

4. 对税收收入的影响

4. Impact on tax revenue

• 全球最低税补足税和利润转移减少导致企业所得税收入增加。

• GMT top-up taxation and reduced profit-shifting lead to increased CIT revenue

• 收入收益预计达当前平均企业所得税收入的3.2%–5.4%，即910亿至1550亿美元。

• Revenue gains are estimated to amount to 3.2 – 5.4% of current average CIT revenues or USD 91 – 155 bn

• 估算表明全球最低税为所有辖区组别带来收入收益。

• Estimates suggest GMT yields revenue gains for all jurisdiction groups

• 由于SBTI安全港的不确定性，结果相对于早期分析有更大的误差范围。

• Larger margins of error in the results relative to earlier analysis due to uncertainty surrounding SBTI Safe Harbour

注：估算基于2019-2022年数据。估算考虑了利润转移敏感性和QTI可用性及使用的变化。估算基于GloBE规则和QDMTT的当前实施。估算以第10年SBIE情景呈现。辖区分组基于世界银行分类。投资中心是FDI与GDP之比超过150%的辖区。

Note: The estimates are presented based on 2019-2022 data. The estimates account for the variation in the sensitivity of profit-shifting and QTI availability and use. Estimates are based on the current implementation of the GloBE Rules and QDMTTs. Estimates are presented for the year-ten SBIE scenario. Jurisdiction groups are based on World Bank classifications. Investment hubs are jurisdictions with an FDI-to-GDP ratio of 150%.

4. 对税收收入的影响（续）

4. Impact on tax revenue (continued)

• 收入收益的分布取决于实施情况。

• Distribution of revenue gains depends on implementation

• 正在开展大量工作支持发展中国家实施全球最低税。

• Significant ongoing work to support developing countries in implementing the GMT

• 包括双边支持、与伙伴组织合作以及多边方法。

• Including bilateral support, work with partner organisations and multilateral approaches

• 多边方法包括通过OECD全球关系计划、税务管理论坛以及指导说明、工具包和外联活动开展的工作。

• Multilateral approaches include the work through the OECD Global Relations Programme, the Forum on Tax Administration, as well as guidance notes, toolkits and outreach events

4. 对税收收入的影响——包含投资中心

4. Impact on tax revenue – Including investment hubs

• 相对于企业所得税，投资中心的收入收益较大，由两个因素驱动：

• Large revenue gains relative to CIT in hubs driven by two factors:

– 中心存在大量低税利润。

– Large amount of low-taxed profit located in hubs

– 企业所得税收入水平较低。

– Low level of CIT revenues

• 在全球实施下，中心因利润转移减少带来的损失被QDMTT实施带来的收入增加所抵消。

• Under global implementation, losses due to reduced profit-shifting in hubs are offset by increases in revenues through QDMTT implementation

• EIA模型不考虑实际活动的迁移，这可能也会影响投资中心的收入收益和经济活动。

• EIA Model does not account for relocation of real activity which could also affect revenue gains and economic activity in investment hubs

实施后的早期证据

EARLY POST-IMPLEMENTATION EVIDENCE

实施后证据

Post-implementation evidence

• OECD正在进行的分析估算全球最低税的实施后影响，超越模拟分析。

• Ongoing OECD analysis estimates post-implementation impacts of the GMT, moving beyond simulations

– 新发布的合并MNE财务数据允许分析第一个实施后年度（即2024年）。

– Newly published consolidated MNE financials allow analysis of first post-implementation year (i.e. 2024)

• 还可基于2022年和2023年数据研究潜在的公告效应。

• Potential announcement effects can also be studied based on 2022 & 2023 data

– 相对于之前税收与投资研究中常研究的单边税收变化，多边方法可能改变改革效应。

– Multilateral approach may change reform effects relative to unilateral tax changes often studied in prior research on tax and investment

– 方法论使用双重差分法（DiD），比较范围内MNEs与范围外MNEs在ETR、投资和就业方面的变化。

– Methodology compares changes for in-scope MNEs to those of out-of-scope MNEs focusing on ETRs, investment and employment using a Difference-in-Differences (DiD) approach

识别策略

Identification strategy

• 分析利用2016-2024财政年度的Orbis合并数据。

• Analysis draws on consolidated Orbis data for fiscal years 2016-2024

• 识别基于利用7.5亿欧元收入门槛的双重差分法。

• Identification based on DiD approach exploiting the EUR 750m revenue threshold

• 对照组：收入低于门槛的MNEs；最低收入要求3亿欧元以确保与范围内MNEs的可比性。

• Control group: MNEs below the revenue threshold; Minimum revenue of EUR 300m required to ensure comparability to in-scope MNEs

• 处理组：收入高于门槛的MNEs；适用20亿欧元规模上限。

• Treatment group: MNEs above the revenue threshold; Size cap of EUR 2bn applied

• 基于2017-2019年平均收入和MNE状态进行分组，排除收入接近门槛的MNEs以避免进入/退出处理的选择问题并改善分组分配。

• Assignment based on 2017-2019 average revenues and MNE status and exclusion of MNEs with revenues close to the cutoff to avoid issues related to selection into/out of treatment and improve group assignment

识别策略（续）

Identification strategy (continued)

• 基线样本：非金融MNEs；排除航运业（鉴于全球最低税范围规则）；2017-2019年ETR低于中位数以关注受暴露的MNEs。

• Baseline sample: Non-financial MNEs; Exclusion of shipping sector given GMT scoping rules; Below-median ETRs in 2017-2019 to focus on exposed MNEs

• 虽然两组在构造上规模不同，但在全球最低税前ETR、增长率和行业构成方面具有可比性。

• While groups differ in size by construction, they are comparable in terms of pre-GMT ETRs, growth rates and industry composition

• 回归分析包含MNE和年份固定效应以及额外的控制变量，以考虑潜在的混淆因素。

• Regression analysis contains MNE and year fixed effects, and additional control variables to account for potential confounding factors

• 在稳健性检验中变化收入边界和样本定义。

• Revenue bounds and sample definition are varied in robustness checks

全球最低税对ETR的影响

GMT impacts on ETRs

• 全球最低税公告前平行趋势成立。

• Parallel trends before GMT-announcement

• 没有显著的估算公告效应。

• No significant estimated announcement effects

• 实施后，合并ETR估算增加约1-2个百分点（取决于样本）。

• Post-implementation, estimated increase of consolidated ETRs by around 1-2 percentage points depending on the sample

• 基线1.74个百分点相当于相对于改革前处理组均值ETR增加了12%。

• Baseline of 1.74 ppts represents a 12% increase in ETRs compared to pre-reform treated mean

• ETR影响由税收支付增加而非税前利润变化驱动。

• ETR impacts driven by increase in tax payments rather than changes in pre-tax profit

注：图报告了ETR的估算DiD系数。处理组和对照组的定义基于7.5亿欧元收入门槛。样本由2017-2019年ETR低于中位数的MNEs组成。竖线表示系数估计的95%置信区间。虚红线表示2021年全球最低税公告，实红线表示2024年全球最低税实施。结果变量在2018年标准化为零。

Note: Figure reports estimated DiD coefficients for ETRs. The definition of the treatment and control group is based on the EUR 750m revenue threshold. The sample consists of MNEs with below median ETRs in the 2017-2019 period. The vertical lines indicate 95 percent confidence intervals for the coefficient estimates. The dashed red lines indicate the announcement of the GMT in 2021, the solid red lines indicate the implementation of the GMT in 2024. Outcome variables are normalised to zero in 2018.

全球最低税对投资和就业的影响

GMT impacts on investment & employment

• 没有证据表明全球最低税公告或实施后投资或就业显著减少。

• No evidence of significant reduction in investment or employment after GMT announcement or implementation

• 附加分析表明，短期内范围内MNEs的结构没有变化（如从低税辖区迁出）。

• Additional analysis shows no changes to the structures of in-scope MNEs (e.g. relocation away from low-tax jurisdictions) in the short run

稳健性和异质性

Robustness and heterogeneity

结果稳健

Results are robust

• 纳入额外控制变量。

• Inclusion of additional controls

• 替代收入边界，包括无上限。

• Alternative revenue bounds, including no upper bound

• 使用平衡样本。

• Use of balanced sample

方法适用性确认

Suitability of approach confirmed

• 安慰剂处理年份产生无效结果。

• Placebo treatment years yield null-results

• 使用范围外企业的安慰剂检验产生无效结果。

• Placebo tests using out-of-scope firms yield null-results

部分MNE类型ETR反应更强

Stronger ETR response for some MNE types

• 税务激进度较高的MNEs。

• MNEs with greater propensity for higher tax-aggressiveness

• 盈利能力较高的MNEs。

• MNEs with higher profitability

• 总部位于实施IIR辖区的MNEs。

• MNEs headquartered in a jurisdiction with IIR

结论

CONCLUSION

2026年EIA模型更新的结论

Conclusions from 2026 EIA Model update

ETR预计上升2.8–3.7个百分点。

ETRs are estimated to increase by 2.8 – 3.7 percentage points

ETR差异预计减少19–25%。

ETR differentials are estimated to be reduced by 19 – 25%

利润转移预计减少22.6–44.6%。

Profit-shifting is estimated to be reduced by 22.6 – 44.6%

全球税收收入预计增加企业所得税收入的3.2–5.4%。

Global tax revenues are estimated to increase by 3.2 – 5.4% of CIT revenue

早期事后分析的结论

Conclusions from early ex-post analysis

• 分析评估了MNE层面对全球最低税的实际反应。

• Analysis assesses realised MNE-level responses to the GMT

– 没有预期效应的证据。

– No evidence for anticipation effects

– 2024年全球最低税实施后，合并ETR统计显著上升1-2个百分点，对于更可能从事利润转移的企业效应更强。

– Statistically significant rise in consolidated ETRs of 1-2 ppts after GMT-implementation in 2024 with stronger effects for firms more likely to be engaged in profit-shifting

– 2024年未发现对应的MNE层面投资或就业统计显著减少。

– No corresponding statistically significant reduction in MNE-level investment or employment identified for 2024

• 结果为短期效应。

• Results are short-term effects

– MNEs可能逐步调整。

– MNEs might adjust gradually

– 全球最低税设计参数在后续年份变化。

– GMT design parameters change in later years

• UTPR随时间逐步生效。

• UTPR comes into force over time

• SBIE比率随时间降低，增大全球最低税税基规模。

• SBIE rate reduces over time, increasing the size of the GMT tax base

• 2026年达成的并行方案调整了GloBE规则。

• Side-by-Side Package agreed in 2026 adjusts GloBE rules

• 工作论文与本网络研讨会同步发布。

• Working paper released alongside this webinar

有任何问题吗？

ANY QUESTIONS?

关注OECD税务

Keep up with OECD Tax

在社交媒体上关注我们：OECD Tax / OECDtax

FOLLOW US ON SOCIALS: OECD Tax / OECDtax

订阅我们的通讯：https://www.oecd.org/en/about/newsletters.html

SUBSCRIBE TO OUR NEWSLETTER: https://www.oecd.org/en/about/newsletters.html

访问我们的网站：www.oecd.org/tax

VISIT OUR WEBSITE: www.oecd.org/tax

联系我们：ctp.contact@oecd.org

CONTACT US: ctp.contact@oecd.org