How to design RWA for data assets?

Source: Ye Kaiwen

Last week, at the Data Cross-border Symposium at the Hong Kong-China Financial Elite Exchange Center in Admiralty, Hong Kong, many elites from the Ministry of Commerce Research Institute, Hong Kong University of Science and Technology, enterprises, associations, etc. discussed cutting-edge issues such as cross-border data flows. I combined the speech with the minutes and supplemented it into an article on digital assets and RWA for reference.

In the past two years, there have been many policies and discussions around data elements and digital asset transactions. The establishment of the National Big Data Bureau and relevant data bureaus in various places means that data elements have become an important direction, and discussions around data asset exchanges have also begun to increase. RuiHe Capital mainly focuses on the establishment of RWA investment banks, and is in close communication and cooperation with licensed exchanges and licensed Hong Kong-funded financial institutions such as securities companies, asset management, and trusts. RWA is Web2.5, a combination of traditional real-world assets and tokenization, and a fusion of traditional assets and finance with Web3.0 and encrypted assets. It is equivalent to a transitional and compromising intermediate state. At the symposium, President Wang of Hong Kong University of Science and Technology recommended Hong Kong RWA, which is very suitable for data assets.

The inclusion of data assets in the balance sheet and data mortgage financing promoted in the mainland are basically a new financing channel for state-owned enterprises, because most of the enterprises with large-scale personal data or industrial data are state-owned enterprises. The inclusion of data assets in the balance sheet and the loan are nothing more than the bank releasing money to state-owned enterprises, which are basically "assets" with no liquidity.

Is data a trade or a transaction? If it is a trade, it means that data is a commodity, and if it is a transaction, it means data assets. Assets are not data itself, and data cannot be bought and sold directly, especially when it comes to data privacy protection regulations; we rarely say what data I buy, and there are often indirect purposes, such as reaching consumers with certain attributes, in order to have a more accurate credit assessment for loans or contracts. Moreover, these data are divided into 2C and 2B data: personal data, such as data similar to banks, WeChat, medical health, etc.; industrial data, such as data similar to enterprises, complete sets of accessories, production and manufacturing, market sales inventory, etc.

From the perspective of asset securitization, data assets are more financial assets based on the income or cash flow brought by this purpose. If we analyze it from the perspective of the ADF industry analysis framework (ADF: Assets-Transactions-Finance), it will be very clear.
How to assetize data? RWA is a good direction and model. The RWA model is not a direct transaction of physical assets in the real world, but an assetization and tokenization based on the cash flow or expected returns brought by the underlying assets, and there is liquidity in the secondary market. Therefore, RWA is particularly suitable for the "transaction" of data assets.

There are many introductions to the relevant policies of Hong Kong RWA. Regarding data, Hong Kong has the "Hong Kong Policy Declaration on Promoting Data Circulation and Ensuring Data Security", which mentions several points: one is to ensure the anonymity of access assets, and the other is to use blockchain technology to build infrastructure for data circulation.

How can data become a valuable financial "asset"?

First, it is an application scenario with a relatively high degree of digitization. Data can realize the value of the chain confirmation and value isolation (SPV), and realize a data asset of "SPV+smart contract+cash flow". For example, the core business of Fubao Group - streaming content copyright services, because streaming content is completely online, and cash flow and income distribution are also digital and online, this can be designed as a typical RWA data asset.

Secondly, the payment scenarios derived indirectly from data. The aforementioned data indirectly generates credit or enhances credit enhancement. For example, the DePIN project generates consensus data based on distributed networks and accounting, which generates credit or credit enhancement value from the perspective of financial assets, and institutions are willing to pay. For example, the Domo project, the BOM distributed network of automobiles, transforms personal related data, driving habits, etc. into data assets that are valuable to personal credit and insurance pricing algorithms, and insurance companies pay.

Also, the intermediary value scenario of data. Experts mentioned the barter of data. In fact, there was a state-owned enterprise quota of the State-owned Assets Supervision and Administration Commission in overseas trade before, which is equivalent to a virtual large asset pool of overseas state-owned enterprises. Instead of complex and expensive foreign exchange settlement and then going out for procurement, it can be directly bartered, which reduces capital costs and improves procurement efficiency. This intermediary value comes from the electronic quota generated by detailed data and pricing algorithms, which is also a similar RWA product.

The RWA of data assets requires several steps:
The first step is to package data assets as financial products, the second step is asset tokenization, and the third step is trading. In the future, tokenized cash flow and second-tier financial derivatives can be further expanded.

For data assets in the Mainland, there may be such a path:
Mainland data assets, get a road permit to set up a VIE structure to a Hong Kong entity, the Hong Kong entity issues data asset RWA, trades and invests in Hong Kong licensed exchanges, and connects with mainland enterprises through the WFOE of the VIE structure, forming a cycle.

Data assets are not just data itself, but a digital asset ecosystem: from data desensitization, labeling and asset confirmation, application coordination, pricing algorithms, transactions and liquidity pools, etc. Compared with personal data, industrial data may be easier to assetize. Because industrial data, often combined with the digitalization level of Industry 4.0 in the industry, can not only generate industrial credit value, but also provide value for trade, supply chain finance and industrial capital, so the scenarios and cash flow sources of data assetization will be richer.

The complexity of industrial data assets requires the implementation of a dynamic data asset pricing algorithm based on the data asset pool, combined with technologies such as AIGC, so that different industrial chains and industrial data can reasonably and dynamically form the value pricing of assets and intermediaries.

In this way, the final data asset ecosystem will be very rich, not only buyers and sellers of digital asset transactions, but also LPs that provide data asset liquidity, funds that incubate and invest in data assets, speculators and arbitrage institutions, RWA investment banks for data assets, and so on.

A friend at the scene asked which industries are suitable for data assets? Ye Kai summarized several industries here: 1) Cultural content streaming media, the core is online streaming content, not the traditional film and television box office, but those content smart boxes, online video platforms, short dramas of cultural exports, and Tiktok, etc. These streaming content are already completely online subscription recharge and payment; 2) New energy photovoltaic storage and charging distributed network, China's photovoltaic storage and charging capacity accounts for 80% of the world, but it is mainly hardware, and is relatively weak in software. The green electricity data asset space generated by fully market-oriented distributed network equipment is very large. Don't make the hardware manufacturing capacity in China, and the soft asset confirmation pricing transaction finance in the United States and Europe; 3) The AI ​​computing power mentioned by President Wang, AI computing power is mainly used for computing and processing data. We are currently the largest purchaser of AI computing power. We have both centralized training of large models and the need for reasoning and rendering of a large number of application scenarios. These can form effective AI computing power data assets based on the scale of procurement needs; 4) Medical care and health care. With the popularization of digitalization and electronic prescriptions, smart wearable devices for diagnosis and care, smart devices for chronic diseases, etc., data assets generated by distributed networks are formed, which can be combined with personal health assets and service agency assets; 5) Manufacturing industries with a high degree of industrial 4.0, such as smart home appliances, mobile phones, smart robots, etc., these data that are deeply integrated with family individuals and specific application scenarios can also be combined and designed into valuable data assets. In summary, data assets are very suitable for RWA. Data asset RWA can realize the digitization, securitization and globalization of data.