CARF Global Implementation Timeline Overview: What Are the Commitments of Mainland China and Hong Kong?

marsbitPublished on 2026-01-28Last updated on 2026-01-28

Abstract

CARF (Crypto-Asset Reporting Framework) is a global framework for the automatic exchange of tax-related data on crypto-assets, targeting crypto-asset service providers as reporting entities. As of the end of 2025, 76 jurisdictions have committed to implementing CARF, with a phased rollout plan. The first group, including the UK and EU member states, will begin automatic information exchange in 2027. The second group, which includes Singapore, the United Arab Emirates, and Hong Kong, is scheduled to fully implement the framework in 2028. Data collection for reportable transactions will begin one year prior, starting in 2026. Hong Kong has explicitly committed to implementing CARF. It plans to start collecting crypto-asset transaction data in 2027 and commence automatic tax information exchange with partner jurisdictions in 2028. Service providers operating in Hong Kong must establish compliance and reporting mechanisms. In contrast, Mainland China has not yet committed to CARF and is not included in any of the implementation batches. It is also not listed by the OECD as a jurisdiction with relevance that has yet to commit. Under its current regulatory framework, which imposes strict limitations on crypto-asset activities, there are no legal crypto-asset service providers that could be integrated into the CARF system. Therefore, in the short term, Mainland China does not meet the conditions for participating in CARF's routine information exchange. It is noted that Mainland C...

Author: Fintax

Basic Positioning of CARF

CARF is a framework for the automatic exchange of tax-related information on crypto-assets, with crypto-asset service providers as the reporting entities, used to support tax authorities in various jurisdictions in obtaining information related to transactions involving their taxpayers.

Global Implementation Progress and Timeline

According to information released by the OECD Global Forum, as of the end of 2025, 76 countries and regions have committed to implementing CARF and will advance its implementation in batches.

The first group of jurisdictions plans to commence the first automatic information exchange in 2027, primarily including the UK and EU member states, etc.; the second group of jurisdictions plans to fully implement in 2028, including Singapore, the United Arab Emirates, and China Hong Kong, etc.

According to the institutional arrangement, the collection of relevant transaction data will commence one year in advance, starting from 2026, crypto-asset service providers will need to systematically organize reportable transaction information.

Figure 1: Overview of CARF Implementation Progress in Major Jurisdictions

China Hong Kong: Clear Participation and Timetable Advancement

In the aforementioned arrangement, China Hong Kong has clearly committed to implementing CARF and will advance related work according to the international timetable.

Hong Kong plans to start collecting crypto-asset transaction data from 2027 and commence automatic tax information exchange with other cooperating jurisdictions in 2028.

Crypto-asset service providers operating under Hong Kong's regulatory framework need to establish corresponding data compliance and reporting mechanisms, and relevant reportable transactions will be included in the cross-border information exchange process.

Mainland China: Not Yet Committed and Outside the Implementation Scope

In comparison, Mainland China has not yet made an implementation commitment to CARF.

As of the current stage, Mainland China is not included in any implementation batch of CARF, nor has it been listed by the OECD as a jurisdiction that is relevant but has not yet committed to participation.

Under the current regulatory framework, the mainland adopts a strict restrictive attitude towards cryptocurrency trading activities, and there are no legal crypto-asset service providers within the country that can be纳入 the CARF reporting system; therefore, it does not currently meet the institutional conditions for participating in CARF's regularized information exchange.

Future Possibilities and Realistic Judgment

It should be pointed out that Mainland China has fully implemented CRS since 2018 and has mature experience in financial account information exchange.

If future crypto-asset regulatory policies are adjusted, the mainland has the institutional and technical conditions to对接 CARF.

However, given the current policy environment, the possibility of Mainland China joining this framework around the 2027 launch of CARF and in the following years remains relatively low.

Related Questions

QWhat is the CARF and what is its primary purpose?

AThe CARF (Crypto-Asset Reporting Framework) is an international framework for the automatic exchange of tax-related information on crypto-assets. It targets crypto-asset service providers as the reporting entities, aiming to support tax authorities in various jurisdictions in obtaining information related to transactions involving their taxpayers.

QHow many jurisdictions have committed to implementing the CARF as of the end of 2025, and what is the implementation timeline?

AAs of the end of 2025, 76 countries and regions have committed to implementing the CARF. The first group of jurisdictions, including the UK and EU member states, plan to commence the first automatic information exchange in 2027. The second group, which includes Singapore, the United Arab Emirates, and Hong Kong, China, is scheduled to fully implement the framework in 2028.

QWhat is the commitment and implementation plan of Hong Kong, China regarding the CARF?

AHong Kong, China has explicitly committed to implementing the CARF. It plans to start collecting crypto-asset transaction data from 2027 and begin automatic exchange of tax information with other cooperating jurisdictions in 2028. Crypto-asset service providers operating under Hong Kong's regulatory framework will need to establish corresponding data compliance and reporting mechanisms.

QWhat is the current status of Mainland China's commitment to the CARF?

AMainland China has not yet made a commitment to implement the CARF. It is not included in any of the implementation batches listed by the OECD and is not currently classified as a relevant jurisdiction that has yet to commit. Under its current regulatory framework, which imposes strict restrictions on crypto-asset activities, there are no legitimate crypto-asset service providers that could be incorporated into the CARF reporting system, making short-term participation unlikely.

QDoes Mainland China have the foundational experience to potentially join the CARF in the future?

AYes, Mainland China has comprehensive experience in financial account information exchange, having fully implemented the CRS (Common Reporting Standard) since 2018. This provides it with the technical and institutional foundation to potentially adapt to the CARF requirements if its crypto-asset regulatory policy were to change in the future. However, given the current policy environment, the possibility of joining the CARF around its 2027 launch or in the subsequent few years remains relatively low.

Related Reads

Arthur Hayes' New Article: It's 'No-Trade Zone' Time

Arthur Hayes argues that the current market is in a "no-trade zone," a period of high uncertainty created by two converging forces: the deflationary shock from AI and the inflationary shock from geopolitics. AI agents are rapidly displacing knowledge workers, eroding their incomes and creditworthiness, which will eventually trigger a deflationary financial crisis in consumer credit-dependent Western economies. Simultaneously, the war in the Middle East, particularly the potential disruption to shipping through the Strait of Hormuz, threatens global energy supplies and could force nations to abandon the dollar system. Hayes outlines three main scenarios: 1) A return to normalcy, where the deflationary AI shock remains the primary concern; 2) The "Tehran Toll Booth," where Iran controls the Strait and demands payment in gold or yuan, accelerating the end of dollar hegemony; and 3) "Empire Strikes Back," where the US destroys Iran's capabilities but risks a catastrophic regional war that sends commodity prices soaring. In all but the most extreme scenarios, Hayes posits that the key driver for Bitcoin's price will be the *quantity* of money, not its price (interest rates). He expects that governments, forced to fund wars and stockpile resources, will have to print money, expanding the money supply. This would be bullish for fixed-supply assets like Bitcoin, even if it occurs alongside rising rates. However, he cautions that until this liquidity is explicitly unleashed (e.g., when bond market volatility spikes), the risk/reward for new long positions is poor. His current strategy is to wait for a clear signal of monetary expansion before deploying capital, preferring to hold gold and select crypto assets in the meantime.

marsbit1h ago

Arthur Hayes' New Article: It's 'No-Trade Zone' Time

marsbit1h ago

a16z Founder: In the Agent Era, What Truly Matters Has Changed

Marc Andreessen, co-founder of a16z, argues that the current AI boom is not an overnight success but the culmination of 80 years of research, now delivering practical results. He emphasizes that this era is defined by the convergence of four key capabilities: large language models (LLMs), reasoning, coding, and agents capable of recursive self-improvement. Andreessen describes the agent architecture—combining an LLM with a shell, file system, markdown, and cron/loop—as a fundamental shift beyond chatbots. This structure leverages existing software components, allowing agents to maintain state, introspect, and extend their own functionality. He predicts a move away from traditional GUI and browser-based interactions toward an "agent-first" world where software is primarily operated by bots, not humans, with people simply stating their goals. He draws parallels to the 2000 internet bubble but notes key differences: current AI infrastructure investments are led by cash-rich giants and quickly monetized. He highlights that scaling constraints involve not just GPUs but the entire chip ecosystem. Open source and edge inference are crucial for democratizing knowledge and enabling low-latency, cost-effective applications on local hardware. Finally, Andreessen identifies significant non-technical challenges: potential short-term cybersecurity crises, the need for "proof of human" identity solutions, financial infrastructure for agents, and institutional resistance from sectors like education and healthcare. He cautions that societal adoption will be slower than technological change.

marsbit1h ago

a16z Founder: In the Agent Era, What Truly Matters Has Changed

marsbit1h ago

Trading

Spot
Futures
活动图片