Why Fiat Peer-to-Peer will Ultimately Fail

Prerequisite

In my earlier articles I've proved that fiat and blockchain technology are fundamentally incompatible with each other. In this article I'll show that fiat and non-blockchain technology are doomed to fail as well.

The Fundamental Problem

All computers systems are vulnerable. If not by external threat, then definitely by a dishonest insider. If you don't believe me, please consult articles written by Bruce Schneier on this topic to get the facts.

Personally, I've been working 7 years (1997-2003) in the computer security industry (encryption.) The actual algorithms, the cryptography itself, etc. are the least problems to worry about. The fundamental problem is that computer security is based on trust. If that trust is misused, then computer security is breached.

The Legacy

Given these facts, how come that our financial industry, ultimately based on computer security, can be reasonably secure?

This is an interesting topic. First of all, these computer systems are not as secure as you think, but the reasons why the current system "works" are due to multiple factors:

  • Slow capital flows
  • Disjoint accounting systems
  • Capital controls through algorithmic surveillance
  • Difficult to get reward for a dishonest insider to game the system

The current financial system works by having each bank manage their own accounting so all the deposits are internal (presumably backed up in multiple places.) Then each bank has an account of their total owned capital at the central bank. This is the "real" money the bank has (which is typically only a fraction of the amount the bank owes through their deposits.)

Money that flows out from the bank to IBAN accounts are relatively atypical and monitored by internal algorithmic machinery. Credit card companies use similar technology and anyone who've been abroad and traveling a lot knows that this is a real hassle.

Should the worst case scenario happen (which has happened, e.g. Bank hackers steal millions via malware) the loss is local to that particular bank.

Let's move on to the central bank. What happens if a dishonest insider tries to game the system? Well, nothing actually, because the only thing an employee can do at the central bank is to move money between banks. The central bank has no control over individual deposits. So to summarize:

The slow capital flows means that there's time to react when something goes wrong. The disjoint accounting systems mean: 1) possibility to localize abnormal patterns and especially money outflow (e.g. through IBAN accounts, which are typically rare) 2) as a dishonest insider it's difficult to transfer money to yourself.

Integrating Properties from Bitcoin to Fiat

Now suppose that we create a "universal" accounting system, allowing individuals/companies to transfer money instantly peer-to-peer in the entire world. Then we have:

  • Fast capital flows
  • Homogenized accounting system (at least we should be able to access accounts from anywhere to anywhere)
  • Hard to apply algorithmic surveillance (likely to get many false positives)
  • A dishonest insider could transfer enormous amounts of money to himself (or by proxy.)

If the central bank can create money and have access to specific accounts/deposits, then that would be an absolute disaster as a dishonest insider would be able to create money for himself. Furthermore, he can then quickly move that money to many different accounts to launder them and making it practically infeasible to reverse the transactions.

If you sit down and ponder deeply on these issues it becomes evident that a digital money peer-to-peer system is an intrinsically hard problem to solve. It is rather surprising that it can be solved at all (bitcoin being the solution.)

Ripple - The Competitor

There's one "competitor" against the blockchain technology, and that is the consensus model used by Ripple. It's important to understand that even a consensus model does not allow a fiat digital currency.

The exact problem Ripple tries to solve is the money remittance problem. Here Ripple is only meant to be an intermediary form of "currency" to be hold for the risk of transferring fiat money. Here each node in the network chooses the peers to trust and only if more than 50% of its immediate peers agree on something it will confirm the proposed transaction.

In consensus based networks you still cannot create a fiat based digital currency. That's why Ripple, the currency, has a strict fixed supply. With a capped supply Ripple can be relatively confident that even in the worst case scenario (their payment system gets blocked) the currency itself (Ripple) will still be intact.

Conclusion

So the good news (or the bad, depending on which side of the table you sit) is that there's simply no way out for the legacy banking industry. They'll never be able to come up with something that could even remotely compete with bitcoin.

Computer security is a very delicate and tricky issue. It doesn't matter how much research they'll put into blockchain technology, because I'm sure an academic institution will come to the same conclusion as I have, and maybe Bruce Schneier himself will raise this topic. When that happens you'll be lucky if you just own one whole bitcoin while banks will be fighting to build their bitcoin reserves.

Therefore, what remains is how to scale bitcoin so it can handle 10000 transactions per second and beyond. It turns out that there's actually a solution (albeit not as elegant as you wish) to this problem, but I'll save that to my next article.


Datavetaren