Hydra- Decentralized Credit Reporting 1 Hydra - Decentralized Credit Reporting Abhishek K Agarwal Ko Ohashi Credit Hydra Author Note Abhishek Agarwal, [email protected] ​ Ko Ohashi, [email protected] ​ Hydra- Decentralized Credit Reporting 2 Abstract This paper defines a credit data tracking protocol that will collect & store consumer credit information in a decentralized manner on blockchain1. The decentralization will give consumers the ownership of their own credit data and will enable them to start tracking their histories irrespective of location. This will allow institutions to give loans to billions of people in developing world. Individual data ownership will also eliminate data and identity theft issues. The protocol consists of three parts: HydraID: A unique ID assigned to every individual that will provide unique identity to the participant in the ecosystem. HydraSTATS: A standard data repository of past credit data that tracks all debits and credits for an individual. HydraRISK: A credit risk score for the individual on a predefined scale to provide credit worthiness of a borrower. Keywords: Credit reporting, Identity Thefts, Security, decentralization, blockchain ​ 1 http://www.ifc.org/wps/wcm/connect/2867f3804958602ba222b719583b6d16/FI-CB-KnowledgeGuide-E.pdf?MOD =AJPERES&CACHEID=2867f3804958602ba222b719583b6d16 Hydra- Decentralized Credit Reporting 3 Overview Hydra is a credit data tracking and reporting protocol that decentralizes credit data collection and reporting to blockchain. Hydra stores data using IPFS/SWARM protocol that can reside on individual’s devices giving them complete control of their data thus enabling them to become their own credit bureau and enabling lending to billions of underbanked. Hydra is building an integral part of the loan ecosystem that will integrate with traditional Fiat and Digital lenders to create a seamless risk assessment and data reporting system. Hydra will also integrate with traditional and new identity management systems to store and provide identity verification so that lenders can issue fully compliant loans on the platforms. The protocol is designed to provide various benefits: ● Faster and Easier onboarding for new countries: Building decentralized credit ​ tracking on consumer devices is easier and can be done with a stepwise approach (by segmenting the population). Whereas, building a centralized system requires standardization that is expensive and takes decades to implement. ● Identity Protection: Individual ownership of data that requires signing with a 256 bit ​ private key for each transaction secures lending systems while eliminating data & Identity theft threats2. 2 https://www.consumer.ftc.gov/blog/2017/09/equifax-data-breach-what-do ​ Hydra- Decentralized Credit Reporting 4 ● Cross Border Credit Identity: Individual ownership of data allows them to carry their ​ data in standard format across border, thus eliminating cold start issues that a lot of people face while traveling between countries. ● Real time data: Decentralized data with inbuilt tracking of transaction will be updated in ​ real time3. (Unlike a centralized system takes anywhere from 1 month to 3 months to update) ● Reporting Error Elimination: Decentralized tracking at the individual level will ​ eliminate the tracking and reporting errors that creep into centralized credit reporting4. This will save billions of dollars in operational costs5. 3 http://documents.worldbank.org/curated/en/662161468147557554/pdf/70193-2014-CR-General-Principles-Web-Re ady.pdf 4 https://www.federalreserve.gov/pubs/bulletin/2004/summer04_credit.pdf ​ 5 http://www.gao.gov/new.items/d031036t.pdf ​ Hydra- Decentralized Credit Reporting 5 Background For any credit market to work well the tracking of past borrower behavior is of utmost importance. It gives a lender an indication of the creditworthiness of a borrower, enabling them to sell customer the right product. Lack of this data hinders credit growth and economic growth, depriving consumers and countries a significant source of wealth creation6. Centralized credit reporting agencies like Equifax and Transunion have served this purpose in various countries over last few decades. Earlier centralization was the only solution, there was no way to decentralize the information and still maintain its sanctity. Blockchain technology has revolutionized this. A system built on blockchain can be decentralized and can still ensure that data is secured, immutable and unaltered. In this paper we are proposing a credit data tracking protocol that is built on the blockchain that will replace traditional systems with a decentralized approach. This change will make data tracking cheaper, easier, more secure, and will also improve the accessibility of data. Hydra Protocol Stack Hydra Protocol creates a new ecosystem that promotes and facilitates the building of credit in new territories and improves on current approach in an exponential manner. By adopting Hydra countries can launch credit tracking to build a lending ecosystem in 1-2 months (unlike traditional centralized tracking systems that takes years if not decades to build and cost hundreds 6 https://mitpress.mit.edu/books/credit-reporting-systems-and-international-economy Hydra- Decentralized Credit Reporting 6 of millions of dollars). Hydra as a tracking mechanism is as effective if adopted by a single person or by the majority of population7. The core of the protocol is a decentralized database built on IPFS/SWARM protocol called STATS. STATS can reside on a consumer device or on a IPFS/SWARM node in the network and is equally secure at both locations. In addition to STATS, HYDRA ID and HYDRA RISK provide the other necessary building blocks to launch the ecosystem. Hydra STATS (Immutable and Secured Credit data repository) STATS creates a standard that allows institutions to store and retrieve credit data to track and retrieve past borrower behavior. Unlike a database that can be altered anytime, Blockchain 7 https://openknowledge.worldbank.org/bitstream/handle/10986/11848/Global%20Financial%20Development%20Re port%202013.pdf Hydra- Decentralized Credit Reporting 7 provides a immutable data registry eliminating the need for centralization for data authenticity checks. For STATS to be effective, authenticated consumer information (credit + identity + transaction) will be required to be imported to STATS. This can be done in an automated manner using API data uploads or using manual data imports. For manually imported data, secondary checks (authentication institutions or social reputation scores) would be required. Automated Data Imports (centralized validated data) Populating STATS from data sources that provide automated way to import the data inside the blockchain provides the best and easiest way to bootstrap the ecosystem . For automated pulls it will be assumed that data is authenticated and no secondary validation is required. The structure of this data import will be function createRecord { id: 00983fCw39 ​ dataType: PastCreditInformation ​ debt: Y ​ debtType: Credit Card1 ​ startDate: 06/21/2001 ​ Active: Y ​ Open: Y ​ currentStatus: Current ​ frequency: monthly ​ everDelinquent: N ​ beginingAmount: 0 ​ currentAmount: 234.5 ​ totalCapacity: 10000 ​ currency: US$ ​ dateOfRecord: 09/21/2017 ​ Jurisdiction: USA ​ } Hydra- Decentralized Credit Reporting 8 The data can be past credit data or other consumer identifiable data like their salary information or their ID or their legal documents. For example, if a consumer imports a salary slip, a national ID and a credit report that has 35 different lines. A total of 39 entries will be created in STATS chain on IPFS/SWARM and the data will be signed with the private key of the consumer and the project that has implemented the wallet. Manual Data Imports (Using Vouching, Reputation & Attestation) For data sources that cannot be imported directly into the STATS a manual import and verification process is required. The process will be completed in two steps: 1)Data Import, 2) Data validation. Data import can be done either through the API’s built by the team implementing the project or using photoscan of the documents. Data validation is more complicated and needs an authenticated source to validate the data. This can can be achieved in two manners: 1) Institutional Attestation 2) Social Attestation Institutional Attestation Institutional attestation (Attestation by institutions that have existing reputation in the current ecosystem) will be a big part of the project at the launch of the ecosystem. Various established institutions that supply personal data in a manual manner will be the certifier of this data. These certifiers can be large institutions including banks, employment companies or government Hydra- Decentralized Credit Reporting 9 agencies. To attest this data, these institutions can either sign the data with their private key on blockchain or do a manual certification. A manual certification (like institution ink seal) will be imported using photoscan and csr validation. Social Attestation Social attestation (Attestation by members who have high social reputation) is the future of Hydra, it will decentralize the authority and enable the ecosystem to self-sustain. A Social attestation is vouching system that drives its credibility from the network effects. A member’s connectivity in the network represents by their reputation score (Range: 0-100) and this determines their reliability. This idea is similar to a combination of Google Page Rank (more links higher rank) and Grameen Bank (social connections and peer pressure has motivated people to act ethically). The reputation score will be based on what percentage of the network can vouch for the member's identity and documents. If a member can get 1% of the network to vouch for them their score will be 100. (In theory your score can go above 100 but that would be significantly hard to achieve once the networks become larger and geographically diversified.) To seed the system Hydra will send out initial invitations to the community members who will be whitelisted using a verification system. Once these community members have joined the ecosystem they can send out invitations to their friends and family whom they know and can vouch for them. Hydra- Decentralized Credit Reporting 10 To get started a participant can get their friends and family who have reputations in the system to vouch for the validity of their data. As more and more people vouch for participant’s data their reputation points go up and make them credible players in the ecosystem. For example: Tom is new user in the system and wants his data/documents to be validated and get established. When he starts he is represented in the system as below: Tom has two friends Jim and George who are already in the system. Jim has 3 direct connections and 32 second connections. George has 15 direct connections and 123 second connections. If both of them agree to vouch for Tom he will get a primary score of 2 and a secondary score of 18. The network diagram for this process will look something like below: