Blockchains, shipping and messages

Bloomberg’s Matt Levine published an interesting take today on the Maersk shipping blockchain trial. Ever the blockchain sceptic (insightfully so, in my opinion), he does see value in the application for the shipping industry, and by correlation, supply chain management.

Some background: Danish shipping giant Maersk has built a blockchain platform, in partnership with IBM, to streamline the management of shipments. A trial was run on a live shipment from Rotterdam to Newark, with the relevant documentation registered on the platform, and the participating parties given access to the parts that they needed to see.

Here’s why Matt finds this interesting:

“Whereas finance already has trusted central intermediaries, shipping doesn’t. Maersk has no trouble keeping track of who owns its containers. The problem that blockchain is solving here is not so much a database problem, as it is a messaging problem. The question is not “how do we keep a list of who owns what,” but “how do we communicate all our approvals efficiently and in the same format?””

I disagree.

The problem that blockchain is solving is a database problem. Even if the only advantage were improved handling of documents, various databases would be involved. Shipping documents are more than just messages. They are, in effect, contracts. And each carries specific information related to the material shipment.

Managing the process involves not only lots of documents, but lots of different databases. What the blockchain can do is bring all those databases together.

You can’t do that without the blockchain, because who would control the new, merged source of information? Even if that answer was obvious, why should the parties trust the central authority? They would end up keeping separate databases anyway, just in case.

He also points out that the teddy bears could be swapped for cocaine and the blockchain wouldn’t know.

“The blockchain is about taming all of the virtual attributes of the container, all of the paperwork that accompanies it. But the boundary between the physical and virtual worlds will always be a bit more lawless.”

Again, I disagree.

Perhaps the Maersk/IBM platform is not content-sensitive, but that’s by design rather than necessity. Other trials have shown that sensors can connect with the blockchain. And sensors can detect changes in content, temperature, humidity, etc. So, theoretically, a DLT supply chain platform could detect fraud and theft.

While I normally applaud blockchain scepticism (the sector needs it), in this case it’s missing the opportunity by focusing on a narrow application. And while Matt Levine has the advantage of deep well of financial sector knowledge as well as a compelling writing style, it has been fun disagreeing with him, just this once.


Leave a Reply

Your email address will not be published. Required fields are marked *