Chapter 20 of the book Reckless: The Story Of Cryptocurrency Interest Rates is published below. The full book is available on Amazon. The book was written before the bankruptcy of FTX and therefore does not include coverage of this event. However, the book does provide useful commentary in the run up to the failure of FTX, which provides context for the eventual calamity.
One of the most well-known potential weaknesses of Proof of Stake systems is the existence of staking derivatives. There is also the wider concept of simply outsourcing the staking process. This is when an Ethereum investor sends their Ethereum to a third party, who conducts staking on their behalf. This outsourced staking is potentially a serious problem with regards to the security and effectiveness of the staking consensus system. These third-party staking intermediaries take custody of the stake and the risk here is therefore potentially far greater than when Proof of Work miners use mining pools, as the usage of pools does not result in a change of control of the physical mining power.
The outsourcing of the staking process feels almost like a mainstream financial product, for both retail and institutional investors. Staking can economically be considered as a process which is purely financial in nature. Unlike Proof of Work mining, which can be thought of as an industrial process. As long as this industrial process continues, Bitcoin should continue to survive.
Most of the large cryptocurrency exchanges either offer or plan to offer custodial staking services. At the same time, staking seems quite suitable for an investment product. Why should anyone invest in a plain vanilla Ethereum fund or exchange traded product when they could invest in a version with staking and earn a higher return? Of course, many people actually need to use Ethereum to pay gas fees and balances needed for this cannot be staked, however most holders of Ethereum are still speculators and investors. For these investors they are likely to want staking investment products.
Core to an effective staking protocol is the slashing mechanism, a system whereby stakers are punished for bad behaviour, such as changing their vote and attempting to conduct double spend attacks. If staking is outsourced, the operators of the staking servers do not own the underlying stake and therefore they may not be sufficiently deterred by the slashing punishment system. Although, you could argue a similar problem occurs in Proof of Work, where ownership of the miners and operation of the miners could be separated, for example when a public company engages in Bitcoin mining.
Outsourcing the stake can also cause centralisation, if staking is concentrated in the hands of a small number of players. This could eventually result in the network being vulnerable to censorship and then the utility of Ethereum could quickly degrade. Financial products which pay a passive yield are often prone to the pressures of centralisation. The investment and financial services industry has a track record of consolidation and winners scooping up all the capital, often more so than in other industries. Regulation and economies of scale are a key driver for this. The centralisation here could be worse than in Proof of Work, where the natural geographic dispersion of appropriate energy assets, across multiple jurisdictions, could protect the system from centralisation to some extent.
This centralisation is already a significant problem in Ethereum. Based on data from beaconcha.in, the top five staking services already account for 60.7% of the network by stake. These services are often cryptocurrency exchanges, who do not own the underlying coins and are staking on behalf of their customers. These exchanges typically already have relationships with financial regulators and a service like staking, which pays a yield, could very well be seen as a regulated financial product. Therefore, the risk of regulation and censorship is very real, even in the medium term.
|Percent of stakers|
Due to some of the nuances in the protocol, the impact of this possible censorship is difficult to assess. Proof of Stake is a far more complex system than Proof of Work and therefore discussing how the network may be censored can be quite difficult. We will not go into the details here, but a possible outcome is that many of these services stop providing staking services or reduce the extent of their services and provide a degraded yield. The result of these staking services ending or coming under intense regulatory pressure could be the following:
- Limited actual effective censorship,
- A slower blockchain in periods of turmoil related to the censorship,
- Eventually, a more diverse staking landscape, with better censorship resistance characteristics,
- Fewer stakers,
- A lower Ethereum price, and
- Higher staking yields.
Tokenised Staking Derivatives
The entities performing the outsourced staking as a service business, could also issue tokens to their clients, representing shares in the staking pool. Staking rewards could then be issued to these token holders. These new tokens could be issued on top of the Ethereum blockchain. The coins would be just like Ethereum, except they have credit risk associated with the staking pools and you cannot pay gas fees with the coins.
There are several key advantages associated with these token products. They provide owners of the staking pool the ability to enter and exit more easily, by buying or selling the tokens, without any lags. The tokenised staking coins also mitigate another key potential problem associated with staking on Ethereum. The staking yield needs to compete with other yields inside the Ethereum system, for example yields you could earn by providing liquidity in DeFi. With this tokenised staking approach, stakers can now earn two yields at the same time, thereby partially negating this problem. For example, one could deploy the staking pool token into the DeFi ecosystem and earn even more yield. These staking tokens could even have basically all the key properties of Ethereum. You could use them to make payments, make markets and even use them as collateral to borrow other coins. This also can be said to solve the other problem with proof of stake systems. The staking tokens, in theory, could even be invested in productive projects or spent on consumer goods. Therefore, no funds are locked up and no useful investments are prevented due to Ethereum’s staking system. With these strong and clear advantages, it is even possible that almost all the staked coins end up in tokenised staking derivative pools. Therefore, pretty much all the economic problems with the staking protocol could be solved.
The above may sound too good to be true and it probably is. There must be a catch somewhere. We can’t have all these advantages and no real costs. This very much exposes an ideological difference that various commentators and analysts in the cryptocurrency space have when evaluating Proof of Stake systems. Some people believe that you can’t have something for nothing and look for weaknesses. They believe that if it appears as if you have something for nothing, this may persist for a while, but the system will be unsustainable and eventually fail, perhaps in a catastrophic crisis. Others, a more optimistic group, do believe a consensus system with no real costs is possible and are actively trying to construct one.
The flaw in their reasoning, that staking tokens solve all the economic problems, appears to be that many of the security assumptions on which the Proof of Stake consensus systems relies, may begin to break down. If everyone has staking tokens and uses them for a variety of functions, such as making payments, providing liquidity in DEXs or as collateral to borrow, the ultimate economic beneficiary of the tokens will not be the same entity as the entities which are staking. Therefore, the actual stakers may not be sufficiently compensated by the rewards in the staking system, or sufficiently threatened by the punishments in the system. This issue, of misaligned incentives is quite common in the investment industry. Another issue is that if everyone is staking, then perhaps there is no real staking yield at all. If the yield is paid to everyone, then it looks more like adding zeros on to the end of the currency than a genuine investment return. It would all be smoke and mirrors. This system could work for a while, perhaps many years, but eventually it could result in a catastrophic failure in consensus. The multi-layered staking system could then collapse.
Despite this potential weakness, staking derivative tokens have proved to be extremely successful so far. There are three main providers of these tokens. Lido has stETH, Binance has bETH and Rocket Pool has rETH. At the time of writing, Lido is in the lead and the stETH token has more Ethereum backing it than the two other tokens, which are small by comparison. A potential problem here is that this could be a winner takes it all type market. The economies of scale in tokens are extremely high. For example, the network effects when making and receiving payments are large and tokens with stronger liquidity on offer on exchanges can become dominant. Therefore, the centralisation risk is high and this is a considerable issue for Ethereum.
Around 28% of the staked Ethereum is currently allocated to the Lido pool and exists in the form of stETH. This is about US$7.2 billion worth of stETH floating around, at the time of writing. The most liquid venue for buying and selling stETH, is on the Curve DeFi exchange protocol.
In theory, the price of stETH should always be less than or equal to the price of Ethereum, because one can always subscribe for more stETH at par, by adding Ethereum to the Lido staking pool. One cannot yet redeem stETH for Ethereum, as stakers cannot yet withdraw. Therefore, for now, stETH should trade at a small discount. Once the withdrawal feature is implemented and activated, stETH should track the price of Ethereum more closely and its utility should therefore improve. This may result in the creation of even more stETH. Before the upgrade, stETH should trade at a discount, reflecting the uncertainty as to whether this upgrade occurs and when it occurs.
One key part of the June 2022 earn crisis left out of this book until now is stETH. Many of the earn platforms, like Celsius and trading counterparties, such as 3AC, had invested in stETH to earn the yield. However, their liabilities associated with this were typically in Ethereum, not stETH. Therefore, when the liquidity crisis occurred in June 2022, they had to pay back their clients in Ethereum, but they only had stETH. The trouble is of course that stETH is not redeemable. Therefore, the earn platforms had a significant duration mismatch.
Therefore, there was a rush to sell stETH on Curve. A significant stETH discount of around 3% first emerged in mid May 2022, as Luna failed. Then, during the peak of the crisis, on 18th June 2022, stETH traded as low at 0.925 Ethereum. Finally, by the end of September 2022, the price of stETH recovered, to a discount below 1%. 3AC received a significant haircut when it liquidated its stETH in the crisis. On 14th June 2022, 3AC sold 30,000 stETH. Celsius is believed to have held US$426 million of stETH, making it perhaps the largest holder. Again, Celsius is likely to have taken a considerable haircut.
Remember, Curve is not like a traditional exchange with an order book. Curve operates the Ethereum vs stETH market with two pools of liquidity, an stETH pool and an Ethereum pool. With all the pressure to sell stETH in May and June 2022, the pools became unbalanced. For example, in mid June, the pool had around five times as much stETH as it did Ethereum, 500,000 stETH and 100,000 Ethereum. One may think such an imbalance would cause more of a price dislocation than just 7.5%. However, Curve has a custom shaped curve with special parameters for each trading pair. Since the stETH vs Ethereum pair was designed when people expected the prices to be reasonably similar, the curve shape prevented the discount from reaching even larger levels. This benefited some of the distressed entities such as 3AC and Celsius. As we went into July 2022, some people wanted to buy stETH at a discount and eventually the pools became balanced again. At the time of writing the breakdown is 50.3% stETH and 49.7% Ethereum.
What the earn collapse showed, was that in a liquidity crisis, people preferred Ethereum to stETH. This was at least the case in this crisis. In the future, if the stETH ecosystem is more developed, people may be happy holding stETH as a form of liquidity in a crisis. In addition to this, an stETH crisis is unlikely to repeat itself in the same way, because next time there is a major cryptocurrency liquidity crunch Ethereum may have upgraded and stETH may be redeemable. On the other hand, even after the upgrade, there will be limits on the number of stakers who are allowed to withdraw in any given period. If all the stakers try to withdraw at once, the process could take over a year. Therefore, some kind of liquidity crisis causing a race to exit staking is possible, with the staking tokens trading at a discount.
Interest Rate Swaps
Another very different potential form of a staking yield derivative is an interest rate swap. With this type of product an investor could lock in the Ethereum yield for a period of time, converting it into a fixed income type product. For example, a staker could purchase a swap contract entitling them to receive fixed payments and pay variable payments, based on the actual Ethereum staking yield. This investor would then have locked in their staking yield at a fixed rate. They would no longer need to worry about more stakers joining or fewer miner tips causing the yield to fall. This swap type interest rate product is very popular in traditional finance. These fixed income type products can be attractive for certain investors, who for example, may have fixed liabilities they need to cover, for instance somebody who has borrowed Ethereum at a fixed rate. On the other hand, if someone has lent out Ethereum, they may be concerned the yield could increase and they could take the other side of this swap trade.
These swap products do not seem to exist yet. As the cryptocurrency economy becomes more accustomed to Ethereum’s inherent variable yield, it seems likely that many financial and derivative products may emerge that enable traders to speculate on, fix or hedge the important quasi-interest rate.