Blockchain, big data and the value of global trade

Most data turns out to have a greater value than the sum of the parts. There's a story about a global courier firm that said it saw a large drop off in its monthly orders at some point in 2007, not too long before the bottom fell out of the global economy. Traditional economic forecasting did not see an issue, but had there been some visibility into trade finance data at that time it would have shown many contracts had been cancelled, affording us some warning of what was ahead.

Macro-economic forecasters and statistical analysts know that trade data provides the most precise window into the global economy there is. Trade finance data has always been a notoriously opaque part in the supply chain, but we are now seeing end-to-end digitisation, as a multitude of banks and software providers test out trade/supply blockchains and other digital platforms.

Supply chains have traditionally transacted using antiquated processes and physical letters of credit, bills of lading and purchase orders. This is now being rapidly digitalised, making it easy to aggregate and also access.

Christian Lanng, CEO Tradeshift (a trade finance and supply chain solution working with HSBC and Santander), said: "Over the next two to five years we are going to see massive digitalisation of that global trade dataset. It's moving from offline to online. It also means you can trade on it in a whole different way, leverage it in different ways. You can use it for research and in other contexts."

Tradeshift supports over 100 different transacting types, like bills of lading, letters of credit, invoices, purchase orders and more. Lanng said one of the main reasons you are seeing lots of data aggregation is because of cloud computing, which has driven down the cost of the procedure. "All this is built on open powerful platforms. We have open APIs. We can build third party applications that can leverage this data."

Today there's whole industry that revolves around alternative data that can be cleaned and readied for use by alpha-hungry asset managers and hedge funds. But while a company like Visa has one of the most valuable datasets on the planet, it would never reveal transaction data about its customers. The same could be said about Maersk, which has at its fingertips a snapshot of more than half the world's supply chains.

So while the platforming of supply chains and the use of shared ledgers will undoubtedly make this data more immediate and more granular – the question is will it ultimately become more accessible?

Tony Nash, CEO at data analytics firm Complete Intelligence, says trade datasets are a very good proxy of global supply/demand, but trade finance remains a missing piece.

"Trade lets us know exactly what's happening on supply chains, you know exactly what's happening with finished goods. The problem with trade finance today is that there is huge amount of opacity. Individual banks and trade finance firms aren't really that willing to surrender their trade finance data, so anything around trade finance that's public is really estimated.

"I think blockchain will bring some transparency to trade finance and actual aggregate values, and will do it in a much more real time sense. A lot of this information is so far delayed that you don't really know what's going on."

"There isn't a global source for this stuff. We've got goods data; we've got some services data, and all this currency commodity macro data. Trade finance adds a very interesting layer in terms of the timing of impacts and how they impact the cyclical nature of trade."

Those who follow the enterprise blockchain space know that data privacy is sine qua non – as much between participant banks running nodes as anyone else. Professional data collection and curation players are well aware of this fact. But they see the potential and are watching the space closely.

Tammer Kamel, CEO of Quandl, a well-known provider of data insights to the asset management industry, said: "Almost all our hedge fund customers covet supply chain insights. While Quandl has some data now that illuminates part of the picture, gaps remain. Blockchain adoption could well be what throws more light on things.

"That said, most supply chain participants have strong incentives to keep the details of their business operations confidential. In this space, as in others, we are watching blockchain adoptions closely to see what powerful data will emerge as the by-product."

Blockchain technology is very much in the offing and there are likely to be many variations in design as different use cases are fully explored. Amber Baldet, blockchain programme lead at JP Morgan, has overseen the creation of an extremely security-conscious modification of the Ethereum public blockchain called Quorum, which is aimed at enterprise uses. However, she said some blockchain uses – and she used global trade as an example – will likely see lots of value in the future by being spread across very large networks of users.

"Everybody is looking at supply chain on blockchain. If you only want to internalise receivables flows or letters of credit – markets between a relatively small number of banks – then you are fine using an enterprise blockchain solution purpose built for a small group of semi-trusted parties," said Baldet.

"But if you think that over time you want to add thousands or millions of end point suppliers to this thing; not only corporates, but perhaps actual vendors making fabrics and very little shops ... any sufficiently adopted sort of permissioned chain starts to look a lot more like a public chain."

This type of blockchain, while not open to the entire world, might become so adopted that we will gradually see security/utility trade-offs. At some point this seems to bleed into the old internet/intranet argument often invoked by opponents of private blockchains.

Jeremy Epstein, CEO of Never Stop Marketing has spent some time thinking about the intersection of big data and blockchains and how this might play out. He sees this phase of blockchain construction as similar to the big excitement in the late 90s around corporate intranets. "Over time, there will be a realisation that the real value is in permissionless, or the internet version of the blockchain. I think what you'll start to see is new supply chains that are being built around these sort of public blockchains," said Epstein.

He pointed out that companies like Google and Facebook lead the way in artificial intelligence and machine learning at the moment partly because they have access to the most data. But that might change with a more open, decentralised internet.

"In the short term, big companies with massive amounts of data are going to have an advantage. But I think eventually we will see the value of the public blockchain; the ease of innovation on top of it, that's the real differentiator.

"Collaboration is going to happen as more and more of these small players start jumping on, creating nodes and micro-niche apps. It could be a Kenyan coffee supply chain node on this global coffee supply chain.

"Big companies can't innovate at the micro level because it's not valuable or profitable for them. But some entrepreneur in Nairobi can. And all of a sudden the data on public blockchains starts exploding. So if I'm going to analyse what's happening in the global coffee market, I can look at this public blockchain; I can put my AI and machine learning algorithms on top of it," he said.

"The more data wins, and eventually there's going to be more data in the public blockchains. No private organisation or even private ecosystem is going to be able to compete with billions of people."

Newsweek's AI and Data Science in Capital Markets conference on December 6-7 in New York is the most important gathering of experts in Artificial Intelligence and Machine Learning in trading. Join us for two days of talks, workshops and networking sessions with key industry players. With many of the world's leading capital market firms represented don't miss out - register now.