While I respect David Richards, the author of this opinionpiece, my reasons for why banks should be hesitant to adopt blockchaintechnology differs from his:
“Onpaper, blockchain seems to have vast potential. But, in common with many otherfintech fads, its current popularity owes more to speculation than genuinepotential, and the chances of it realizing that apparent promise are slim.
That’sbecause the fundamental issues with blockchains aren’t going to go away. First,the whole proposition is built on distrust and on anonymity, shrouded inparanoia. And since blockchains are also highly hackable, disaster isfrequently just round the corner. Last year, the DAO, a venture capital fundusing a decentralized blockchain, lost more than $60 million worth of Etherdigital currency—around a third of its value, when its code was breached. Butsuch is the protection awarded to the parties involved in blockchain-based transactions;it’s impossible to trace the culprits and recover the lost amounts. It’s quiteludicrous.
Thereis a social engineering argument around blockchain that does not stack upeither—that if someone has enough computer power to hack blockchain, they couldmake more by minting bitcoin than by wreaking havoc on users. This is based ona misguided assumption that criminals’ aim isn’t to bring down financialsystems and cause chaos. Where there’s a will, there’s a way.
Theregulated financial industry is founded on trust, not distrust, which relies onparties being known to each other. Unless some kind of shady dealing is goingon, it makes no sense to push transactions underground. It’s certainly no wayto manage risk.
Afurther major sticking point for a financial institution is throughput.Blockchain reconciliations typically happen in periodic batches, which is notgood enough for real-time banking. To keep up with the speed of digitalbusiness, true values need to be reflected at all times.
So,again we have to come back to what banks are trying to do that makes them thinkthey need blockchain. What is the use case in banking-as-usual? What is itsolving?
Real-timereconciliation
Diga little deeper, and what most institutions are really looking for is a robust,trust-based coordinated agreement mechanism for managing very high volumes ofreal-time transactions digitally across networks. Strategic goals might includemaking payments more immediate, easier to automate and cheaper to manage. Or itcould be facilitating new innovation as the Internet of Things takes off,paving the way for all manner of new distributed transactions along a valuechain.
Allof which relies on reliability, speed and trust/transparency. Blockchain doesnot provide that—they weren’t designed to.
IfI’m a goods or service provider that needs to fulfill a purchase, I need to besure of cash settlement before activating delivery, and that’s somethingblockchain will never be able to offer. You can’t see who you’re dealing with,and ledger reconciliations are unpredictable, depending on batch processes andcomputing queues. The technology was never meant to be used in this way.
So,our message to banks is forget blockchain. Unless the fundamental flaws areaddressed, it is wrong for just about anything you might want to do. It’s abubble you don’t want to invest in. Whatever your intentions, there’s a muchmore appropriate way to deliver them.”
Technology companies, including IBM and Microsoft, areclaiming that they can rip the blockchain out of Bitcoin and implement ageneral solution that delivers trust via algorithms that are also capable ofrecording transactions at the speed of traditional databases, while beingimplemented across multiple independently operated nodes over unreliablenetworks. But this comes close to defying the laws of physics.
.png "Pick Twob")
Any two of these three features can be easily implemented,but the problems associated with implementing all three simultaneously areprofound. Some have “solved” the problem by having all nodes operate on thesame system or in the same cloud, but that could be done on a database and leavesa single point of failure. Some solve the trust problem by arguing that allparticipants are banks that can trust each other; but that ignores known cases ofcollusion such as the Libor incident. Some offer nothing more than an immutableledger maintained across multiple nodes, but distributed databases can beeasily configured to perform that function. I would argue that if this problemcan be solved, it will likely be for a specific use case and not as a generalplatform that supports multiple use cases as with a database.
A Nobel Prize might have gone to Bitcoin had SatoshiNakamoto not been anonymous and it is likely a Nobel Prize awaits the scientiststhat can enable fast transactions, over a distributed and unreliable network,that also implements algorithmically enforced trust for . Until they receivethe award in Stockholm, I remain skeptical. Perhaps quantum computing can crackthis nut but that’s cheating!
Overview by Tim Sloane, VP, Payments Innovation at Mercator Advisory Group
Read the full story here
