Meet Alex at X-Tech 2019: Financial Services and Technology (Get 15% discount on the Convention using presenter code: XTECH33)
Before we delve further, we need to agree on what blockchain is. Blockchain was first described in the same paper that introduced Bitcoin. Blockchain was created to enable Bitcoin, a currency not controlled by any authority, by permitting transactions without centralized control or trusted agents. This is a rather difficult requirement to satisfy and the genius of blockchain is that it does so in a fairly simple way.
What is Blockchain? Each block in a blockchain is just a recording of a transaction such as “X gave Y $100”. This transaction is encrypted to a “hash value”. A hash value is a fixed length string of letters and numbers that is extraordinarily unlikely to be reproduced by applying the same encryption to a different underlying transaction. When the next transaction comes along and is also “hashed” it is given as input not just the new transaction but the hash value of the preceding block. Thus, the hash value of the new block is reflecting both the transaction it is representing and the encrypted form of the previous transaction. This creates a sequential dependency between the blocks. As more transactions come along, the same steps are applied, resulting in a sequential chain of blocks or “blockchain”.
How is this secure? Blockchain provides for security in two ways. First, because the blocks are dependent on each other, attempting to modify a block somewhere in the middle of the chain requires one to recompute every subsequent block in the chain, which would require a lot of computation. The really innovative feature of blockchain is that it assumes a world where every transaction is seen by every participant in the market, each of which maintains their own version of the blockchain and votes on the validity of any new blocks being added. Thus, a market participant who wanted to change history (to perform some illicit action) would not only have to recompute every subsequent block but would need to overpower the voting of every other market participant. To do this, our illicit actor must possess more compute power than the rest of the market combined.
Under what conditions is this useful? Now that we have defined what blockchain is, let’s list some criteria that would make it useful: (1) the properties of the market in question make it beneficial to not have a central controlling or trusted party, and (2) there are many market participants none of which is likely to ever own more than 50% of the market (and the underlying computing power). If the market does not meet these criteria, then Blockchain is a poor solution and other solutions may be simpler and more appropriate.
Everything is a nail. There is the old saying that if the only tool you have is a hammer then everything starts to look like a nail. Blockchain evangelists have claimed that it will solve food distribution problems in the developing world, prevent the flow of “blood diamonds”, eliminate money laundering, and so on. Unfortunately, the claimants hardly ever explain how Blockchain is going to address these issues in a way that is fundamentally different and better than traditional solutions. If what is needed is a database to store a series of transactions that is provably accurate then far simpler and cheaper ways already exist.
A June 2018 study by Greenwich Associates estimated an annual blockchain-related spend of $1.7B in Capital Markets and Banking. They state “Early feedback from proofs of concept (PoCs) indicated that the technology indeed showed promise in solving some of the big problems in financial services … Indeed, 14% now claim to have successfully deployed a production blockchain solution. These successes notwithstanding, blockchain technology has yet to achieve those lofty expectations.”
We can conclude from this that with a large enough investment it is possible to get working financial markets functions using blockchain. Can you hammer in a screw? Sure you can. It just might not be the best way to solve the problem and may take a great deal more effort than the simpler approach of just using a screwdriver.
Is there a meaningful use case? To answer this, we need to go back to the principles of what would make a blockchain useful and identify problems that cannot be solved effectively or efficiently using traditional centralized technologies.
For example, much of global communication between banks uses one of several centralized communication mechanisms, with SWIFT being arguably the most important. These mechanisms are centralized simply because it is not practical to have the ~11,000 SWIFT bank members establish bilateral communication mechanisms (this would result in an unmanageable >60 million relationships). But what would happen if a SWIFT member was compromised and was forced off the network? This has already occurred in the 2016 attack on the Bank of Bangladesh and several other less publicized instances. What would happen if one or more of the top 10 global banks were similarly affected? The answer is trillions of dollars in daily business would not be able to move with potentially catastrophic consequences to the global economy. A practical alternative does not currently exist but must be found that allows at least the most important flows to continue.
Interestingly, SWIFT itself has conducted a pilot program to explore the use of blockchain. While the early results appear to be mixed, it is perhaps a harbinger of a promising approach, where a blockchain-like distributed ledger capability is offered by a central provider. Done right, this has the potential of solving problems that have proven very difficult using traditional approaches while reducing the technical burden of implementing a complex technology to a single central provider. In other words, a hybrid between the traditional centralized and fully decentralized models may have the best opportunity of providing practical solutions to currently unsolved problems.