How does R3 prevent snoopers from front-running transactions on its blockchain?

As soon as the issue of building a blockchain is brought up, it immediately attracts people's attention.

If you pay close attention to the latest developments in Internet finance, you have probably heard that many banks want to use blockchain to transform record keeping. Millions of transactions such as bank settlements, transfers, swaps, etc. will be recorded in a shared distributed ledger.

Banks have built a working prototype and should be able to adopt blockchain technology next year. But before it can actually be rolled out, the financial industry must overcome a major challenge: How can banks ensure that the new tool doesn’t let competitors discover sensitive transactions of the bank or its customers?

The reason for this embarrassing situation is that the essence of blockchain is a permanent record that can be verified by users. The transparency of blockchain is one of its greatest features; however, this feature also has the risk of exposing confidential bank information.

To understand how banks plan to prevent competitors or others from snooping on transactions , I spoke with executives from R3, a company that works with the financial industry to build blockchain applications.

Secret public chain

David Rutter, who once worked on Wall Street, is the CEO of R3 and has been responsible for developing the underlying software technology for the blockchain project for R3's bank members. If the project is successful, transaction settlement speed will be greatly improved, while the cost of record keeping will be greatly reduced.

R3 isn’t the only company developing blockchain applications for banks — Digital Assets is a major competitor — but R3 has by far the most members.

Sitting in a glass conference room with a stony-faced man named after the movie Top Gun, Rutter explained the problems blockchain could pose for privacy.

Hypothetically, "if I have a deal with Goldman Sachs and I don't want JPMorgan to know about it and front-load the deal," he said, suggesting that the banking industry needs to forge a conflicting relationship between openness and secrecy.

To understand the dilemma Root describes, it helps to recall the main principle behind how blockchains work: They use software to generate tamper-proof records of transactions and rely on multiple parties to verify that transactions actually took place.

The most famous of all blockchains is the underlying technology of Bitcoin. This ledger is completely public and the records are kept by miners. Miners can be anyone with a computer. They can get a certain amount of Bitcoin rewards by running software to verify whether the information in the new block is valid.

The blockchain built by R3 is different. Banks that are committed to and benefit from the development of blockchain want to store transactions on the ledger, but do not want everyone to participate in the verification of transactions. Therefore, they build a "private chain" that only allows trusted members to write and save transactions on the blockchain . The members here, of course, refer to banks and financial regulators.

The problem, however, is that banks need to rely on each other in the process of verifying the information in the blockchain. But like Rutter's hypothetical example above, this process involves sharing information between competitors. Fortunately, they have found a solution. "We try to add maximum flexibility to the verification layer. If the verifier we need is also a competitor, we can use encryption to hide the relevant information."

But what does this mean in practice?

Enough data

Charley Cooper, a former regulator and lawyer, now works with R3 engineers and is responsible for explaining internal developments to the media and the public.

When it comes to shielding sensitive transactions from a bank’s blockchain, Cooper noted that similar discussions have been taking place regarding the bitcoin ledger. In that case, he said, the solution that has been proposed is encryption: specifically, that the person who records transactions in the blockchain can modify them with a private key that can only be unlocked by authorized parties.

However, this plan makes no sense for banks. Cooper said there are two reasons: the first is the issue of capacity expansion.

Although the volume of transactions on the Bitcoin ledger is relatively small, the blockchains that banks build will reduce the volume even further. In particular, the purpose of a bank blockchain is to record billions or even trillions of transactions, including stock trades, interest rate swaps, and consumers using credit cards. It is simply not practical to encrypt such a large amount of data and put it on a blockchain.

Cooper explained that the second reason is that banks that are extremely concerned about security do not want data to be out there, even if it is encrypted. "Bitcoin hardliners say security cannot be broken, but I have not found anything that cannot be broken in all this time. Even if it is possible, why share information with people who are not involved in the transaction? Why should Pimco see the transactions between JPMorgan Chase and BlackRock, even if the transactions are encrypted."

Another solution is what R3 is doing, which is developing a new type of blockchain in which public transactions used for verification contain only a subset of the information .

Cooper used the example of interest rate swaps, saying that Dodd-Frank required both parties to record 140 data fields. But in the blockchain, other banks can only see specific data points, which may just be codes showing the nature of the transaction and the price range.

At the same time, the blockchain will store complete transactions, which can only be seen by relevant parties and regulators with monitoring obligations.

Cooper summed up R3’s software, known as Corda, this way: “Our platform, Corda, is derived from blockchain technology, but it is not a blockchain in the full technical sense. We want to extract the best features of blockchain (transaction data storage, immutability, asset representation, encryption technology, decentralization, consensus) and integrate them into financial services. Pure public blockchains cannot do this.”

The result is that by flexibly stipulating who and what information the blockchain reveals, banks can replace many traditional systems with appropriate new ledgers. In the future, it will also become a template for many other industries to integrate blockchain technology - land title registration, mortgage companies, etc.