There has been a spate of property crimes in London, apparently, with houses being sold by fraudsters who rent houses and then pretend to be the lawful owner or acting on behalf of the lawful owner. Here’s an example. Note that in all of these cases it is the purchasers who are defrauded (the real owner of the property retains the title).
This young woman — who paid the full amount with no mortgage — had been duped into handing over £1.35 million to the alleged vendor. The money was last seen on its way to a bank in Dubai.[From Max Hastings on the thieves who stole his wife’s house and sold it for £1.3m | Daily Mail Online]
Well, that should be no problem then. Since banks have to comply with expensive and stringent KYC procedures, the bank in Dubai will of course know who the account belongs to and will be able to direct the police as to who to arrest. So there’s really nothing to worry about. That’s the good news. However, I couldn’t help but wonder how this situation had come about.
Obviously the agents and the solicitor acting for the purchaser have to carry out appropriate identity checks. Or, if they couldn’t be bothered to do that, they could at least have looked on LinkedIn, where there is only one Kevin Hafter listed and he works for TfL, so I would imagine he is reasonably easy to track down. At which point it would have been obvious that the person renting the house is not Mr. Hafter at all but a fraudster. But the dopey sods seemed to have missed this and also missed another rather obvious aspect of the renter’s behaviour that would naturally attract opprobrium in polite society.
However, one aspect of the deal that would have rung alarm bells — if we had known of it — is that ‘Mr Hafter’ made all payments to the agent in cash.[From Max Hastings on the thieves who stole his wife’s house and sold it for £1.3m | Daily Mail Online]
Bizarre, isn’t it? He might as well have been wearing a striped jersey, carrying a bag marked swag and wearing a black face mask. Gazillions wasted on KYC, but Mister X can rent a £1.3m house for cash and no-one bats an eyelid. Paying rent in cash should be illegal, obviously, and that would also have the additional benefit of generating more tax revenues, I’m sure. But it’s not the point of this post.
The most interesting part of this story comes further down. The house belongs to a woman called “Penelope Hastings”. Now, presumably, at some point the buyer’s solicitors had asked for “proof” that she had instructed Mister X to act on her behalf. So…
We learned that a woman from Catford, South-East London, had changed her name to ‘Penelope Hastings’ by deed poll. Having done this, she secured a passport declaring her to be ‘Penelope Hastings’.[From Max Hastings on the thieves who stole his wife’s house and sold it for £1.3m | Daily Mail Online]
Brilliant. So, because someone with a passport in the name of Penelope Hastings turned up, the solicitors naturally assumed that this must be the Penelope Hastings listed as the property owner by the Land Registry. Well, it’s a simple mistake to make in a country that has no identity infrastructure. Had the solicitors phoned up to ask if the passport was valid, they would have been told that it was. End of. They would not have been told that it had just been issued to someone who had just changed their name. The passport office aren’t running some sort decentralised and shared data structure that shows all state changes and the history of all credentials like a… well, you know.
So, as someone asked me, could the blockchain solve problems like this? Is there an actual use case here, for providing some form of security in places with no reliable identity infrastructure?
Well, in essence, it might. The problem comes about because the personal name on the land registry is an attribute, not an identifier. There is nothing to connect the ledger entry on the house with the individual, just a limited attribute correlation that you probably wouldn’t accept for an online bingo game, much less a house sale.
Now, of course, a replicated distributed shared ledger could fix all of this. Consider a blockchain version of a registry. Imagine that the house is essentially a “coin” in my wallet. The ledger will tell you which wallet the coin is in. So to prove I am the owner of the house, I have to prove ownership of the wallet. Assume for the time being that this means that, essentially, I have to prove ownership of the private key. Personally, I have all of my private keys tattooed on my scrotum, but other people may prefer alternative approaches. Perhaps the key could be held not by me, since I do not want the responsibility, but by a regulated financial institution that I trust. My bank, for example. Then when the solicitors need proof of instruction by the owner of the property, they are really asking to see something that has been signed with that private key. My bank asks me for permission, I give it to them, they use the key to sign a message to the solicitor. When solicitors are ready to proceed, they ask the bank to transfer the house to their client, which the bank does by sending the coin the address specified by the buyer’s solicitors.
Under this kind of scheme, everyone can see who the house belongs to at any one time, but only the owner or owner’s “key holder” (which in my scheme could only be a regulated institution of some kind) can actually sell it. And what’s more, if the bank allows someone other than me to instruct the transfer, then they are liable.
Why would you do this? Why doesn’t the Land Registry just store a public key certificate against each entry, the public key of the title owner signed by the private key of the Land Registry? Or maybe just a picture of the title holder? Or a pointer to their LinkedIn profile (after all, it’s much harder to create a fake LinkedIn profile than to change your name by deed poll) ?
Broadly speaking, I’d say there are three reasons for looking at a shared ledger solution here. The first is robustness, because every estate agent could have a complete copy of the ledger (in fact, taking part in the consensus-forming process might become a pre-requisite for being licensed to practice as an estate agent). The second is transparency, because all parties (and regulators) can see the state of the ledger at any time. The third, and potentially most important, is that the ledger would then become a platform for innovation, moving into the world of APIs, smart contracts and apps.
Let’s be clear. I don’t know whether a shared ledger is the best way to implement a Land Registry or not, but anyone who says that it is not worth considering as an implementation option is almost certainly wrong.