At Token Summit last month in New York City, Union Square Ventures co-founder Fred Wilson and Polychain Capital’s Olaf Carlson-Wee discussed the state of staking, a cryptocurrency concept that states a person can mine or validate block transactions based on how many coins they hold. The more Bitcoin or altcoin owned by a miner, the more mining power he or she has.
Mr. Wilson prompted the discussion by asking Mr. Carlson-Wee about the most exciting stuff going on in the space currently. Carlson-Wee, who was Coinbase’s first employee, did not hesitate.
“We’ve seen the launch of some of the first really sophisticated, large scale proof of stake protocols,” he said. “Proof of stake has obviously been something people have been talking about for a really, really long time. And there have always been prototype experiments on the fringes trying to figure out how to do this appropriately.”
While people tend to think of Proof-of-Stake as simply a technical consensus mechanism, Carlson-Wee points out another powerful aspect of this consensus method.
“It’s the way that it unifies the people who are actually validating blocks and sort of doing the mining and the actual token holders,” says Carlson-Wee. “In systems like Bitcoin and Ethereum, these are two different groups of people, and there’s a lot of interesting social effects that come with the token holders actually validating all the blocks. For one, everyone is talking constantly about how to get the best returns, what the best set up is, who to delegate to, and more.”
According to Carlson-Wee, this makes participants on a Proof-of-Stake blockchain more engaged in the process of actually coming to consensus than on Proof-of-Work blockchains.
“Whereas, say, mining in Bitcoin is sort of this invisible thing that happens somewhere,” said Carlson-Wee, “[in Proof-of-stake], it’s a sort of social dynamic that shifts among the token holders. That’s really powerful.”
To Olaf’s knowledge, Polychain is running the largest staking operation in the world. “As a large holder, the really interesting thing is now we have a responsibility not to sort of passively sit on the sidelines, but actually build a cloud safety infrastructure that can operate at scale and in a secure manner and participate not just in block validation, but actually launching governance proposals which are going to over the long term dramatically affect the value of the actual tokens. And, so, I find that we end up being way more engaged as a holder of these assets.”
Wilson notes the many companies providing staking services, and that Polychain seems to – instead of using such a service – be building its own infrastructure in order to do this. Olaf says he views this as a long-term choice, and that Polychain has no intention of outsourcing this to a third party.
Its like Google, Facebook, Amazon don’t run on third-party cloud services. They’re so big that it makes no sense for them.
“What we realized is that over time, if you look at the Proof-of-Stake protocols as a percentage of the general market capitalization of all crypto assets, it’s pretty small today,” he says. “It’s probably under 5%.” He thinks this number will rise to be closer to 50%.
“I think it will grow about 10 times from where it is today,” he says. “When we looked at our portfolio in early 2018, we realized that probably most of our fund is going to be in Proof-of-Stake protocols.”
Wilson’s heard the argument that proof of stake systems are not as secure as proof of work systems.
“The fact that people are spending real resources and again, I have issue with that concept on proof of work, like real power or energy consumption is somehow more secure.
Olaf takes issue with the concept of proof of work is somehow more secure, thinking the argument too vague.
“The other thing about proof of stake that is a really big deal to me is if you want one of these Internet mining platforms to truly support the global financial ecosystem, and also reach the scale where you’re replacing things like gold and fiat currency, which is what I am in this to do, you have to be prepared for actors who will attack these networks, not with a profit motive at all, but actually solely to disrupt the network,” says Carlson-Wee.
He adds: “Most of the time when we think about cryptographic and economic models, we think about rational for profit actors, and how they can attack the network. But it’s a different model when it’s like, I have $100 million and I just want to screw up the system. One thing about Proof-of-Stake protocols is if a 51% attack – or often in Proof-of-Stake it is a 34% attack – happens, then after the attack happens, the honest people can simply fork the chain and delete the attackers clients.”
The Proof-of-Stake ledger can just remove them from the ledger and continue forward. “If a 51% attack happens in a Proof-of-Work system, you have to actually change the hashing algorithm of the protocol, which burns down not just the bad guys, but also the good guys.”
When a 51% attack happens in a Proof-of-Work system, you have to destroy the hardware investment of everyone, according to Carlson-Wee.
“Whereas when it happens to Proof-of-Stake, you can just delete the bad guys,” he says. “And so, when you’re trying to prepare for attacks from people that have a lot of money to attack the protocol and resources, and aren’t financially motivated, Proof-of-Stake is a better security model after the attack happens.”
Wilson notices that most of Polychain’s bets have been placed on staking protocols. Wilson wonders why that is. Carlson-Wee notes that the vast majority of blockchain models out there today are Proof-of-State, so its more the market than Polychain.
“The proof of stake design space is a lot richer than the proof of work design space,” says Carlson-Wee. With Proof-of-Work, you basically design a hashing algorithm and that’s it. And, it determines the kind of hardware people should use…But, in Proof-of-Stake, you have much more complex economic rules that you need to design, and these will have rippling long-term effects on security and economics of the protocol.”
For instance, Carlson-Wee points to the Tezos governance model, where token holders can vote on code level changes.
“I like this idea of voting on very formal code level specifications, where people know exactly precisely the code that is going to be merged in if they vote yes or no on this,” says Carlson-Wee, whose Polychain capital participates in such votes.
Upgrades on Tezos take three months to implement – in other words, the system is designed for slow and steady upgrades.
“It’s not like an overnight thing, and it’s not designed for irregular state changes or anything like that,” says Carlson-Wee. “It’s not fast.”
The other thing about governance is in protocols we’ve seen to date, there is a lot of incentive for developers to build applications on top of protocols and they can raise a lot of capital.
“So, people that do token sales, people that launch tokens, they can raise money from venture people,” observes Carlson-Wee. “They can raise money from crowdsourcing everything. Whereas core protocol development really isn’t formally funded. There’s often like an informal foundation, but it’s not super formal. One thing I’m really excited to be on is the ability for token holders to say, okay, I’m willing to take a 1% dilution in order to pay developers to merge in this new upgrade.”
He compels the audience to imagine what would happen if Bitcoin were merging Segregated Witness (Segwit), a proposed bitcoin protocol change designed to.
“How much does that increase the value of bitcoin?” Carlson-Wee asks. “Does it increase it by 1%? If I think so, then I should be willing to agree to a 1% dilution of $10 million, which is a lot of money. It means that dev teams for large scale protocol can have code bounties to contribute to low-level protocol to the tune of millions of dollars.”
Wilson wonders how the price for such a contribution is determined.
“[A developer could say] ‘I have this improvement’ or it could be a bounty,” says Carlson-Wee. “We’re looking for this improvement, if [your solution qualifies], you earn this money.”
It’s not always going to be as clean like, ‘Oh, this is an amazing upgrade, [We’ll dilute] 1%, no problem,” admits Carlson-Wee. “I do believe in [token holder’s] ability to be profiteering, rational actors. And, if they think the token value will rise more than the dilution you’re taking on, rationally they should do that. In these major upgrades, say, a 1% dilution doesn’t feel like that big a deal. Even as a big token holder, any upgrades are probably worth at least a 1% rise in the value of this token. So, that just ends up at scale being a lot of money and then these huge protocol bounties for these teams.” There’s other benefits, too, in such a model.
“It means that you reduce or take away that Foundation layer that’s loosely tethered to the legal and geographic world that basically is just sitting there to have a bank account and rules for how [funds are allocated],” says Carlson-Wee. “There’s a lot these protocols can do where they bake in the ability to incentivize continued improvements to that protocol, which is something they could benefit from greatly.”
There is game theory and behavioral economics to consider in such a model.
“How will people behave when we give them all the right to vote on a capital pool?” Carlson-Wee asks.
Wilson ripostes: “…The best way to solve game theory is to play the game.”
“There’s this weird intersection of game theory and behavioral economics,” says Carlson-Wee. “Like when it comes out that Tether is not as backed as much as people thought, Tether stayed at a dollar. It’s hard to analyze why that is. It’s already interesting that Tether is a dollar to me, because there’s sort of zero probability that it goes up [in price]. But, there’s some probability that it goes down. So, whatever that probability is, I’m surprised it’s not sort of baked into the [price].”