Unpacking upheaval: Interpreting 2022’s blockchain failures
Projects sometimes fail. But when there are lots of headline collapses in a short period, it’s time to look for patterns. Was it all hanky panky or was there a more legitimate reason for failure?
We look at five projects in the blockchain space whose collapse made the headlines over the last 6 months to consider what common things went wrong and what this signals for the sector in 2023. We cover: FTX, ASX, Terra/Luna, we.trade, and TradeLens.
Decentralization sure seems like the culprit
Blockchain is an obvious uniting thread in each of these projects. Each of these projects used decentralized protocols to deliver services. And there is certainly no shortage of known problems in the blockchain space.
Blaming the technology certainly resonates with a skeptical public (like Paul Krugman) who are not aware of where blockchain is being used already.
There’s one big problem with this explanation. The diversity of protocols used in these five projects suggests that blockchain alone cannot be the whole story. Design issues don’t affect all decentralized technologies to the same extent. The technologies used in our 5-project sample were a mix of public and private and include: Hyperledger Fabric (HLF), DAML on VMWare, Terra, and Ethereum. Some of these protocols have had successful deployments in other sectors and cases, which further casts doubt on this angle.
Bespoke solutions are costly
According to both ASX and we.trade, the cost of running the projects was a reason for their decision to shut them down.
Initial costs of designing a solution using a new technology can be daunting. For the banks and exchanges behind these projects, they needed to train up their engineering team, build an ongoing service relationship with the provider and maybe a consultancy, and constantly iterate on the solution, which in the case of ASX meant the final product looked very different from what was initially proposed. This underlines a related challenge – the return on investment can be difficult to estimate in new deployments. Part of the reason is that costs change as deployments begin. Another is that for sector-wide projects, cost recovery depends on adoption. This brings us to the third problem.
Network building is tricky business
When Maersk began collecting members of the TradeLens network, an entirely new network of container operators (GSBN) was formed of operators who did not want to join TradeLens but who were interested in the types of solutions they were creating. Politics still matters.
Both we.trade and TradeLens created and deployed full solutions. Both were available and used among their members (12 banks for we.trade and 15 ocean carriers for TradeLens) and other entities that engaged with the service. But both also struggled mightily from the reluctance of others in the sector to adopt the solution.
FTX and Terra/Luna also suffered from network-building problems but from a different angle. The interconnectedness of both cryptocurrencies and exchanges meant that when their projects failed, they took down others with them. This is a good reminder that networks are important for the adoption of projects, but can also serve as a source of fragility. The way that this can be avoided is regulation.
Lack of regulation trips up deployments
While blockchain as a technology doesn’t require specific regulations, it does interact with certain regulations in a new way. Cryptocurrencies don’t have a clear regulatory structure, DeFi is a fully separate ecosystem from traditional financial services, and electronic documents are often not yet legal in many jurisdictions.
For Terra/Luna and FTX, design decisions were made about their native tokens that were terrible from a technical and business perspective, but did not violate any regulations. This doesn’t define the business as illegal, but it creates some uncertainty and an inability for regulators to intervene once there is a problem. This was the case with Terra and FTX, which both failed far too fast for traditional stability measures to have been addressed.
For we.trade and TradeLens, the solution was to create a rulebook. This is a contract for all parties that introduces some legal structure to a process that isn’t always aligned to the regulatory environment of all jurisdictions. This business decision creates another hurdle to adoption as potential members must agree to a new, unfamiliar set of rules. Of course, plenty of very successful projects use rulebooks. However it underscores the headwinds that regulation creates simply by not being present. We.trade could technically solve a problem that legislation prevented a solution for. The only way to do this was a rulebook.
Interoperability with legacy systems is a key indicator of success
An important lesson from all of these projects is that integration with existing infrastructure is a must. For ASX, we.trade and Tradelens – and honestly any blockchain project – integration contributes to their high cost. For ASX, this is where the problems came to a head. The project had envisioned a complete system overhaul. This required linking into existing systems of record and systems of partners and clients, which is a significant lift and ultimately proved to be too costly.
On the other hand, for FTX, integration was a reason for its initial success and scale. Centralized exchanges sit at the gateway of defi and traditional finance. For terra/luna its complete lack of interoperability led to overall changes in the industry to support projects like circle that were backed by liquid assets. This is a particular challenge for projects because it generally isn’t part of the design process and only happens once deployment has begun.
In the end, scaffolding matters
There is another thing that all of these projects had in common – they failed to build the corporate scaffolding needed for a new business to survive. Every single one of these projects was a start-up. We.trade was spun up from a consortium of banks, Terra/Luna was created by a 30 year old computer scientist, and FTX was designed by a former trader. ASX stood somewhat apart by working with established consultancies, but the project was being built on a relatively untested programming language..
This meant that while each company operated to some extent in the traditional financial sector, none of them managed to build the buttresses that are needed for any company – digital or other – to thrive.
Certainly the choice of technology affected the cost and difficulty of deployment. But in the end the success of a business depends on the ability of its creators to attract other participants and operate within the constraints of the regulatory environment.
If there is one lesson for companies still operating in the blockchain space, it is that as larger and more established institutions continue to adopt this technology, they expect blockchain companies to build their own scaffolding.