This crash is the reckoning of the Ethereum developers

Today's crypto crash is all about Ethereum.  Ethereum's price and the ETH/BTC ratio have been in a freefall, cratering far below the highs set in 2017.  Other networks and coins built on Ethereum have been taken with it, and people who need money to cover margin calls are using bitcoin to cover them, fueling even more losses.

The core issue is that people have finally realized that the Ethereum "merge" is not happening anytime soon.  They are also starting to figure out that even if the merge does occur soon, staked coins are not releasable until sharding, which is far more complex and has an uncertain timeline, debuts. 

The Ethereum developers have been, and continue to be, far too cautious with development, and the community is now losing faith in them.  Stablecoins that represent staked Ether, like stETH, have "decoupled" from the backing asset as depositors realize that Ethereum faces an uncertain future.  This decoupling has cascaded to the failure of cryptocurrency lender Celsius, which was earning interest, in part, through staked Ether.  The Celsius failure will be a big story in the coming years, being 20 times larger than the failure of Mt. Gox.

To understand what's happening, let's review the events of the past few days.  Then, I'll explain why ETH faces a choice: accelerate the merge, or cancel the transition to proof-of-stake and return the ether locked in the staking contract.


The "beacon chain"

ETH 2.0 was scheduled to release in three parts.  The first was the introduction of the "beacon chain," which has already been achieved and which is running successfully in parallel to the existing "ETH 1.0" chain.  The beacon chain runs on proof-of-stake, and the old chain continues to use proof-of-work.  To transfer and stake ETH between the two chains, users send money to a contract on ETH 1.0, after which time the money becomes locked and is no longer able to be withdrawln on ETH 1.0.  

The money that is then available on ETH 2.0 can be staked to earn interest, which was around 4% last week.  Celsius was able to comfortably pay interest in the 3% range by staking ETH - but that ETH cannot not be withdrawln, leading to problems servicing withdrawals.  There are various rumors of exactly how much money Celsius was staking, but I won't report them here because I can't find anywhere that the CEO himself confirmed them.


stETH

The problem of being unable to withdrawal ETH from staking was inconvenient to large staking "pools."  These pools exist because ETH can only be staked in blocks of 32, so people with less money need to join a pool to earn interest from staking.  Unlike at a mining pool, a customer wouldn't be able to withdrawal staked ETH, however, so once a pool is selected, they are basically committed to that pool until the ETH is withdrawlable.

The best solution to any problem nowadays in crypto, including this one, was some "stablecoin voodoo."  As I pointed out in the "crypto dictionary, part 1," any stablecoin that isn't backed by exactly the same amount of the backing asset is not a "stablecoin."  stETH is one of the largest pool's "stablecoins," with each stETH token representing one ETH locked in staking.  Ideally, one would be able to "sell" stETH locked in the pool for unlocked ETH at a 1:1 ratio.

Of course, money available now is not worth the same as money available later, so the stETH token depegged from ETH and traded at a ratio of 0.95 at one point on June 13.


The "Merge"

Getting back to the stages of ETH 2.0, the second step is supposed to be the "Merge."  The goal of the merge is largely to change the network to proof-of-stake and to eliminate its current proof-of-work mining.  Once the merge occurs, there will be one blockchain and ETH 1.0 will be deprecated (assuming that there is not a chain split, which often occurs during forks despite developers' best efforts using tools like "difficulty bombs.")

But somewhere in all the hype over the merge, it seems to have been lost in the fine print that there is a third step in ETH 2.0, which is the entire reason that the network is being upgraded in the first place.  "Sharding" will significantly increase the capacity of Ethereum by essentially building many separate Ethereum networks on top of Ethereum.

Most importantly, it won't be possible to withdrawal any of the staked ETH until sharding is released, and nobody currently has any idea when the release will occur.  If we're lucky, I would suggest that sharding is a 2025 feature.  Thus, the ETH developers' current path of cautiousness is what caused the carnage that has happened over the past weeks.


The options

Recently, the Ethereum developers announced yet another 2.5-month delay in the Merge, which was the proximate cause of everything that's occurred over the past few days.  Then, coincidentally with the timing of the Celsius failure, the developers suddenly reversed course, stating that the "difficulty bomb" that imposed a timeframe on the merge would not be delayed after all.  There is obviously a growing realization within the community that their overcautiousness - only 14% of nodes failed to merge in the test upgrade - is a significant risk to Ethereum's future.

Vitalik Buterin is the only person who can make a decision about how to move forward, but I see that he has two options at this point - to accept more risk in making the transition, or to cancel the transition to ETH 2.0 altogether.  I'm going to argue here that cancellation and refunding of staked balances is a legitimate path forward that would still achieve mass adoption of Ethereum.


Proof-of-stake is risky

The first misconception about the Merge is that proof-of-stake is a "proven" technology that has a lot of precedent in the cryptocurrency world.  On the contrary, the vast majority of coins that have attempted to transition from proof-of-work to proof-of-stake are no longer listed at any exchanges, which means that they have no value.  In fact, I can't recall a single coin that has successfully "transitioned" from proof-of-work to 100% proof-of-stake.

There have been coins that transitioned to a combination of the two, like Peercoin, which currently has both miners and stakers.  There are also coins that started out with 100% proof-of-stake, like NXT.  While NXT has never been attacked, it has declined in usage and community interest ever since its release.  If successful, Ethereum will be the first large network that has ever achieved this transition.


Proof-of-stake is not necessary

Besides being risky, proof-of-stake isn't actually necessary for the Ethereum developers to accomplish their goals of adding more capacity to the network.  There isn't anything preventing proof-of-work from functioning on the "shards."  In fact, the current coin ecosystem in general can be seen as a huge collection of shards that was created in response to the Bitcoin Core developers' decision to limit bitcoin's blocksize.  Bitcoin, litecoin, dogecoin, and a large number of other coins all have different miners mining their blocks.  In our case, we use exchanges to change value from some of the more volatile networks, like Digibyte, into the more stable networks, like litecoin.

In justifying why proof-of-stake was chosen for Ethereum sharding, Vitalik Buterin mentions that a specific shard could be "targeted" for a 51% attack.  While a shard could be attacked in a proof-of-work sharding arrangement, hashrate always follows value.  Shards are not exact copies of each other, the value of transactions on each shard is not the same, and therefore the reward for mining each shard need not be identical across shards.  We would expect higher-valued shards to be better secured.  Even if some low-value shards are attached, most of the value would be preserved.


Proof-of-stake is far more complex

In mining, what happens to miners who submit invalid blocks is simple - they don't get any money.  If they do submit valid blocks but those blocks are submitted all at once and have negative characteristics like double spends, then the majority of miners on the network will likely orphan the blocks.

On the other hand, in proof-of-stake, slashing is used to punish stakers who submit invalid blocks.  One of the problems with slashing is that the process is probably more likely to cost money to people who have configuration issues or network connectivity problems than it is to actually cost the money of dishonest people.


Sharding may be obsolete before it is released

Sharding may simply be obsolete before it is released, because other networks are currently being built on top of ETH that work in similar ways.  Polygon, for example, is a network that aims to create "commit chains," where many transactions are bundled into a single transaction and committed to the main ETH network.  Development of Polygon and other "Layer 2" networks continues in parallel with ETH 2.0, and third-party solutions may very well overtake ETH 2.0's official implementation at some point.

Additionally, sharding requires two changes - to allow decentralization of both "data" and "state," which may be released separately.  Therefore, there may actually be four releases required before staked ether is unlocked.


There may not be enough stakers willing to lock up money after the current crash

The "depegging" of stETH clearly indicates that there are a significant number of people who, if given the opportunity, would like to immediately withdrawal their staked ETH.  In fact, it shows that there are more people wanting to withdrawal than are willing to begin staking; otherwise, the price on the stETH/ETH market would increase to 1:1.  This dynamic is similar to how Tether works - when more people want to withdrawal Tether than deposit it, the value of Tether falls below $1.  Unlike Tether, however, there is no centralized authority that can destroy Tethers to regain the peg.

If the stETH/ETH ratio remains at 0.93, then we should expect network security to decline 7% immediately upon release as that many people pull coins out of staking.

What happens if, when sharding is eventually released, a huge amount of staked ETH is withdrawaln from staking at once?  The testnet merges can account for the all the technical details, but they can't account for human behavior.  ETH may suffer a significant decline in security compared to the existing proof-of-work security level.


Conclusion

Ethereum is in a precarious position right now and is taking down the entire market with it.  The core Ethereum developers need to take decisive action to restore confidence in the network.

Taking a slight bit of risk with less testing, or abandoning the proof-of-stake effort and pursuing an alternate path, would provide certainty to the network's future that is currently severely lacking.  As long as the current path is followed and the delays continue, applications dependent upon Ethereum will continue to suffer.  And since the entire industry is already dependent upon Ethereum in some way, including how Celsius is invested in just about every DeFi token imaginable, the future of cryptocurrency depends on one man - Vitalik Buterin - waking up and taking decisive action in either direction.