- We will continue to have a decentralized, global chain powered by ETH, but most of the economic activities using smart contracts will happen on “national" chains, which are regulated by governments and interconnected through gateways
- Each national chain will place different constraints on the activity according to the law of the land. E.g. a Chinese chain will have different constraints compared to UK’s chain, etc.
- Similar to trade area, we’ll see chains being consolidated amongst economic regions (EU, NAFTA, ASEAN, TTIP), and smaller countries will use the chain of regional players
- Crucially, the role of the judiciary and human intervention continues - smart contracts will have an interface to allow judiciary and regulators to intervene. Humans remain to be the ultimate arbiter of the contract system.
- The biggest contribution of smart contract system will be the increase in efficiency to conduct economic activity, NOT the decentralized enforcement of contract system
Why the global (truly decentralized) chain can’t carry the bulk of economic activity
We *like* human intervention
In short, we want human intervention, and the crux of decentralization in smart contract system is enforcement through lack of such interventions. Once the contract is signed, it’s mechanically executed, and parties cannot back out or change it. It’s an ultimate form of “your are on your own”. Forgot your private key? Tough luck because you’ll never see that money again. Somebody mislead you to sign a fraudulent contract? Well, tough luck - you shouldn't have made a mistake of signing it.
This mode of operation is simply unsuitable for the vast majority of daily economic activities (just count how many times you lost a cash card, or appealed to a redress scheme etc). It’s great not having to worry about losing a key to your entire net worth. It’s great that you have a remedy in case of injustice. Being able to retroactively intervene on contract execution is essential to the smooth functioning of the economy.
Having these on a truly decentralized chain is not practical. Imagine we had a "hook" for smart contracts to allow governments to intervene - this would defeat the point of decentralization. You might argue that blockchain *has* a way to retroactively intervene (which has happened in Ethereum before to stop a "fraud"). However, this mechanism is far from practical to handle day to day problems. First, it's supposed to be reserved for truly extraordinary situations, not for your daily grievances. Second, it's akin to a global plebiscite where votes are weighted by the voters' asset. I think history shows this isn’t the form of arbitration most people would prefer.
Of course *some* economic activity are well suited for the de-centralized chain. For example…
- If there is no alternative enforcement available anyways
- E.g. if you are doing something illegal, and hence can’t call the police
- The contract is so trivial that parties are willing to abandon it in case of issues
- E.g. internet points
- Parties prefer the arbitration mechanism of the de-centralized chain
- There will be a significant number, but I’d still think it would be a minority
Not enough liquidity to read the world
One of the most difficult question in smart contract system is how you "read" from the world, as in injecting what's going on in the real-world, into the contract system. Suppose I create a car insurance using smart contracts. Now an accident happens. You may have encoded all the rules precisely and neatly in the smart contract, but we still somehow have to let the contract know what happened in the real world - like the fact that the accident happend, in what situation etc.
Decentralized prediction markets like Augur is a theoretical solution to this. In this model, economic incentive is used to turn humans into sensors. Essentially, witnesses are invited to make money by offering what they see, and because they most probably lose money if they lie, they will tell the truth.
This will work for things that can be observed by enough number of people - like stock market indices, weather, election outcome etc. But it will not work for events that are too small (like a car accident), because there won’t be enough witnesses (liquidity) that can be incentivised.
Moreover (at least for the foreseeable future), these "reads" will have to involve a lot of interpretation. For example, did someone violate the non-compete clause in an employment contract? You could, of course, apply the same mechanism as Augur to this and basically hold jury trials, which is unsuited for a LOT of situations, especially so if only a low number of "sensors/juries" can participate.
For this reason, centralized institutions like the judiciary, banks, insurance companies, regulators etc. will most likely continue to be instrumental in the contract system, which again erodes the point of decentralization.
What will be the point then?
If the system isn’t decentralized, then what’s the point? Isn’t that what blockchain is about? Yes. But that said there is an additional, significant (IMO world changing) contribution that smart contract puts on the table. In essence, smart contract system is API on steroids (my previous article on this).
Today we are just starting to witness how APIs change the world - companies, banks and governments are opening up their functions through API, which allows third parties to innovate and build on them. The nature of smart contract will be similar, but far more drastic.
For example, imagine how moving works right now. You find a suitable flat. You get proof of income, scan it and send it to the agent. You exchange contract, notify the government, utility company etc. that you have changed your address, etc. etc.
There are services that help with this process, and some parts have even APIs available that could be used (e.g. setting up rent payment). If more parts of the process becomes available through APIs (for example obtaining credit report, reporting change of address to the government), these services will work better and more can be automated.
This is already a big improvement, but smart contract can go further. For example, I can write software that analyses the contract - say my utility contracts, and suggest optimization. I can easily create a variation of contract - e.g. I can ask a company to validate the level of income of my tenants (while not revealing other irrelevant parts of the contract), without having to phone them up. I can just ask the tenant to sign such a smart contract. I can also assign the contract easily - suppose I have the right to assign my tenancy to someone, subject to credit check. I just need to sign such a contract with the landlord at the beginning - once I initiate the assignment, the assignment will happen automatically if the credit check succeeds.
None of this is impossible without smart contracts, in the same way none of this is impossible without APIs either. But of course it makes it a hell of a lot easier, and this is crucial. API bring changes to the world by being a catalyst - it makes things happen that could have happened but didn't because it took too much energy to make it happen. Smart contract will be the same.
So, how will the future look like then?
Few, but not a single chain
The biggest advantage of API was the fact that the internet connected everybody. Just imagine it were fragmented instead - it would be far less useful. This will be the same for Smart Contracts. However, it will take a long time until the legal framework can be converged. Unless that happens in 30 years, which is unlikely, each legal platform (like a nation or entities like EU) will have to have their own chain. In the meanwhile, the chains could be connected via gateways (APIs).