Co-written with Collin Myers, research lead for staking as a service at Token Foundry. Edited by Chris Battenfield, Luke Loreti, and Nathan Chen.
The cryptosphere has been talking about staking as a service for the better part of a year, with protocols like Tezos and EOS hitting mainnet using various configurations like PoS, DPoS, or hybrid PoW/PoS. The market demand for such a service is obvious: hedge funds, VC’s, and whales are holding tons of illiquid tokens that could be supporting networks and generating additional revenue, but spinning up and managing nodes is not something they can or want to handle themselves. Setting up dedicated machines, ensuring physical security, and being on the hook to troubleshoot nodes on immature networks while risking stakes getting slashed should be done by dedicated professionals, not money managers.
Services are popping up to fill this void such as Cryptium and Rocketpool, destined to be highly profitable in the near term before succumbing to commoditization and margin compression in the long term (excellently covered by Arianna Simpson here).
However, the exact staking mechanism, service providers, and business pressures don’t matter nearly as much as the inevitable conclusion of what staking consensus algorithms will lead to.
The problem with staking as a service
Imagine there’s a box.
Any money you put in it grows. The longer you leave it in there, the more it grows.
What percentage of the time do you want your money to be in that box? 100%.
How much of your money do you want to be in that box? 100%
What if a large, trusted financial institution were to come to you and say: “you can leave your money (tokens) with us and we’ll put it in the box for you (stake it). You get a nice return and in the meantime you can have these deposit slips that will always be redeemable for the tokens you’ve left with us.”
Doesn’t that sound like a good deal? You get to eat your cake and have it too. And because this institution is so trusted, these deposit slips trade for par value with the underlying asset and are accepted by all the usual vendors you normally interact with.
Congratulations! We’ve come full circle to reinventing fractional banking! You now have an asset AND a financial instrument that’s a claim on that asset.
In the not so distant future, these deposit slips may be issued on a 1 for 1 basis at first, but once hundreds of millions are participating, it is natural for institutions to think: “we have huge reserves, a bank run is unlikely, fractional banking will boost the economy and create jobs, so we’ll now issue 1.1 notes for 1 token and lend out the additional .1 to entrepreneurs.” No one will rise up in arms. The increase to 1.2 and 1.3 will hardly trigger a news cycle. We will be right back where we started, easily susceptible to economic crises with no hard money and no faith in our financial institutions.
The lending of value to generate more value is one of the oldest tricks in the book. There is a reason that global debt markets are the most deep and liquid markets in the world. All that’s needed is a store of value that is widely accepted by the masses and then the same practices can be applied to it that humans invented many moons ago.
Fractional banking is actually just the first step, because once you have deposit slips, you’re able to unleash the full marvels of financial wizardry and all its accompanying innovation. For reference, global OTC foreign exchange (FX) derivatives markets alone had notional amounts rise to a record high of $95.7 trillion at the end-June 2018. In fact, being digitally native will only expand the product selection as smart contract capabilities add a previously unavailable tool to the arsenal.
This problem persists as long as you can pool or delegate, regardless of whether the tokens arrangement is custodial or not. And once staked, switching costs will prevent most people from changing institutions, just like they don’t bother changing banks for their checking accounts.
This is fundamentally different from proof-of-work consensus algorithms (technically Nakamoto consensus) because you aren’t missing on returns by hodling bitcoins. The definition of “putting your money to work” is different: PoW only earns through lending while PoS can also earn through staking.
There is only one way to prevent this future: by empowering all individuals to be truly self-sovereign not just in how they spend tokens, but also in how they hodl them and stake them.
The biggest challenge we face is building self-sovereignty into the system from the start. People do not want to work to be self-sovereign; they will not choose it. It is too hard and too stressful and has way too much weight to look down at a USB that holds your life savings.
Users need to be able to have control over their money that is easy and convenient, not just when it comes to spending, but also when it comes to loss and recovery. Otherwise they will just turn to what they know — a trusted custodian keeping it safe on their behalf. Staking needs to be accessible to even smaller amounts and on lighter devices, and fund rebalancing between staked tokens and accessible funds should be automatic. It needs to replicate savings — let the interest earning activities happen in the background. The user just needs to see that their money is getting a return by default without them having to work for it.
If we reduce our staking to 95% of our money staked 95% of the time, the difference in staking rewards will be imperceptible and individuals will be able to become truly self-sovereign, with total ownership over their financial lives. This will prevent fractional banking and keep our financial system healthy, trusted, and accountable.
Vitalik’s Devcon4 keynote on Serenity (Ethereum 2.0 / Shasper) and the move to proof of stake was inspiring, but we must remain vigilant to the risks of staking if we want the decentralized world of self-sovereign individuals we’re buidling to remain that way.