This is part two of a series on the Merge. In part one, we explored what the Merge is. In part three, we will learn how the Merge will work. And in part four, we will look at what you need to do to prepare for the Merge.

In part one, we started exploring the Merge and saw that it will be a massive -- and fairly high risk -- change to Ethereum. That raises a question: if the Merge is so complicated and risky, then why bother? After all, Ethereum seems to be working perfectly well today.

So, here, we will look at why Ethereum has decided to make the switch to Proof of Stake. And, not only will we look at the potential benefits, we'll evaluate the risks as well.

The Benefits of PoS

Ethereans believe that Proof of Stake (PoS) improves upon Proof of Work (PoW) in numerous ways. They'd argue that it brings sizeable environmental and economic benefits to the blockchain, while also fitting Ethereum's culture and ideology better than PoW.

Environmental Benefits

The environmental advantages are probably talked about the most.

PoW has earned itself an exceptionally bad reputation for having extreme energy use and high carbon emissions. And, while some of the claims against it are inaccurate or exaggerated, it's difficult to argue that reputation is unwarranted.

In part one, we saw that PoW incentivises miners to accumulate computing power. The more computing power they can direct towards PoW, the more likely they are to win. As a result, miners will keep adding computers to their setup until it stops being profitable to add any more. Over time, this has transformed mining from an accessible, home-based hobby into an expensive, highly professional business. Today, many miners use warehouses full of hyper-specialised computers called ASICs (Application-Specific Integrated Circuit). Those setups require masses of electricity to run.

Now, mining proponents will point out that high electricity usage doesn’t necessarily mean high emissions. Nor does it mean that mining is evil or wasteful.

Historically, miners have made use of overproduced or stranded energy that would otherwise be wasted. Increasingly, miners rely on renewable energy or use carbon offsets to minimise their environmental impact. And, there are even very reasonable arguments that mining can aid the transition to a predominantly renewable-powered world.

Even so, mining clearly does use a lot of electricity and has a sizeable carbon footprint. For many, that is unforgivable and entirely negates any potential benefits. It's also terrible PR. PoW is pretty esoteric, and arguments in its favour are technical and complex. As a result, PoW -- and cryptoassets in general -- are perceived to be a major waste of energy and contributor to global warming. That's not good for adoption.

By swapping to PoS, Ethereum can sidestep the entire environmental debate around PoW. While Bitcoiners try to explain complex counter-arguments to people who don't want to listen, Ethereans will be able to nod and say the Merge cut energy use by 99.95%. It's certainly an easier approach.

So, it's good from a marketing perspective. It will remove one of the barriers that slow adoption. It may also enable investment from funds that have certain ESG commitments. And, it will satisfy the many climate-conscious Ethereans that don't buy into the pro-PoW arguments, so it's a better cultural fit.

Economic Benefits

Some would argue that the environmental benefits alone would be enough to justify the Merge, but that is far from the only positive impact of PoS. The second major benefit is economic.

PoS is generally believed to provide more cost-efficient security than PoW. By this, I mean the cost of successfully attacking a blockchain will increase by a larger amount for every dollar paid to block producers under PoS than PoW.

Intuitively, that makes sense. With its specialised computers and insatiable appetite for electricity, PoW comes with high costs that must be more than offset by the rewards on offer. Meanwhile, staking requires no specialised equipment and has minimal ongoing costs. That means it remains profitable and attractive even if rewards are low.

As a result, Ethereum can safely slash ETH issuance after the Merge by paying stakers less than it pays miners.

Under PoW, Ethereum creates about 13,000 ETH per day to pay miners. That will fall to about 1600 ETH per day immediately after the Merge -- around 87% less. Issuance will increase if more stakers come online, but we'd still be talking about maybe 4000-5000 ETH per day in a reasonable worst-case scenario -- still a better than 50% reduction.

Now, it is worth noting that miners don't act as a constant source of sell pressure. The idea that they immediately sell most of their freshly printed ETH is a complete misconception. Instead, on-chain data suggests miners tend to exacerbate market conditions by hoarding in bull markets and selling in bear markets. Therefore, the reduction in issuance may not protect ETH holders from dilution in the way that many expect.

Still, significantly reducing ETH issuance should have a positive impact on price. At the end of the day, less ETH will be printed and put into the hands of people who must eventually sell to cover costs. And, if miners were prone to selling when the market was at its weakest, the shift to PoS may reduce ETH's downside volatility.

So, the reduction in issuance alone should be good for the price of Ether. But there's more -- a cherry on the cake. Last year, Ethereum received an upgrade called EIP-1559. This changed the way that transaction fees worked, burning the majority of the fee rather than handing it to block producers. This has the effect of further reducing ETH issuance, potentially taking it negative in the right conditions.

The higher transaction fees rise, the more ETH will be burned. The exact gas price required for deflationary Ether depends on the number of stakers. As things stand, any price above about 15 Gwei would see more ETH destroyed than created in each block. A few months ago, that would have looked like an absolute certainty. However, the bear market has brought less activity and lower transaction fees. Based on averages over the past month or so, it looks like the burn will just about match issuance after the Merge.

Whatever happens, the shift to PoS means that a greater portion of issuance will be burned compared to PoW, further reducing Ether's inflation rate. Essentially, it accentuates the core economic benefit of the Merge and makes ETH an even more attractive asset.

Security Benefits

We've already touched on PoS's cost-efficient security, but there's another advantage on top of this. It also enables a more useful and targeted response to attacks.

If someone accumulated enough computing power to attack a PoW chain, the only response would be changing the hash function used for mining. That would make the attacker's ASICs useless, but with massive collateral damage. Every honest miner's ASICs would also become useless after the change. Miners would need to migrate first to GPUs, and then to new ASICs once they were developed for the new hash function. That means the solution to the attack might be more damaging than the attack itself.

Meanwhile, under PoS, the response could be focused specifically on the attacker. Most of the time, the attacker's funds would be slashed as soon as they broke the blockchain's rules. If the attack was more difficult to detect or somehow avoided automatic slashing, a special upgrade could be released to have the same effect. Either way, the attacker would suffer the severe economic penalty alone.

State-Level Threat Protection

PoS doesn't just make blockchains more secure, it also gives stakers better protection from state-level threats compared to PoW.

The hardware and electricity requirements of mining mean it has a large and very visible physical footprint. Meanwhile, staking can be done discretely from almost any location.

If more countries acted like China and tried to ban mining, it wouldn't take much effort to find and shut down most mining facilities. Some smaller mining farms might survive, but the biggest and most important wouldn't stand a chance. The vast majority of mining power in a country could be wiped out with relatively little time and effort.

If the same country wanted to ban staking, they'd have a much harder time. Sure, some validators might be scared away. Others, especially staking pools run by major exchanges like Coinbase, would be easy to find and close. But it would be almost impossible to track down every at-home validator, even with a highly concerted effort.

In some worst-case, dystopian future, a PoS chain would likely be more effective and reliable than a PoW chain. That's a pretty minor benefit and hopefully, it won't be needed. But, good blockchains are built with these scenarios in mind, and if the worst does happen, it could be important.

Enhanced Decentralisation

PoS arguably promotes decentralisation better than PoW. However, this isn't as clear-cut as some suggest.

PoW suffers from a few centralisation vectors.

For example, very few firms are capable of manufacturing the ASICs used in mining. That gives those firms immense power in the industry.

PoW also advantages larger, industrial miners over hobbyists and enthusiasts. Professionals have more money to spend on bigger and more powerful setups. They can also establish connections with ASIC manufacturers and source deals for cheap electricity. All of this gives them a significant advantage over smaller miners and individuals. And that advantage compounds over time. The more they win, the more they earn. Eventually, mining becomes concentrated among a relatively small number of professionals.

PoS is fairer and more accessible -- at least in theory.

Staking doesn't require specialist equipment or low-cost electricity, so it can be done from most homes. Staking pools and derivatives, like those offered by Lido and especially Rocket Pool, make staking even more accessible. They allow people to stake without meeting the minimum capital requirements imposed by the blockchain. However, they can also become centralisation vectors of their own. Staking pools are likely to accumulate enough users and funds to become the dominant player in most PoS blockchains.

It's also worth noting that a blockchain won't be "decentralised" just because it uses PoS. It's very easy to build a PoS blockchain that makes "decentralised" at-home staking difficult, expensive, or practically impossible. The hardware requirements for validators on most high-throughput blockchains are astronomically high. Normal people using residential wifi simply wouldn't be able to keep up with the chain. So, as with PoW, block production would be limited to a small number of professionals.

To its credit, Ethereum has been carefully designed to ensure at-home staking is viable. So, it is at least possible for block production to become meaningfully more decentralised after the Merge. I'm extremely sceptical about it -- and we'll talk more about that later -- but I genuinely hope it happens.

Summary

So, those are the main benefits of the Merge.

It will radically reduce the energy use and carbon emissions associated with Ethereum. It will slash the cost of securing the network, enabling a major reduction in ETH issuance. It will give block producers better protection from state-level threats, while also improving Ethereum's defences against malicious block producers. And, it will (potentially) decentralise block production, giving more Ethereans than ever an opportunity to actively participate in the network's operation.

What Are The Downsides?

Of course, nothing can be all good. So, what are the main arguments against PoS?

Centralisation

A major concern around PoS is that it is in fact more centralising than PoW.

I know I just argued that it could be more decentralised, but there are good arguments on both sides and I want to cover them all as best I can.

The main risk is that stake accumulates almost entirely in staking pools and centralised exchanges. That would give those entities outsized influence over the network, and could even allow them to attack the chain if they wanted.

We've already seen that Ethereum has been designed to ensure individuals can stake from home, which arguably reduces this risk. But, I don't think it will. At the end of the day, however easy it might be to run a validator, it will always be easier, cheaper, and more convenient to let someone else do it. In my eyes, delegated staking through pools and exchanges will be the default option. Only a tiny number of the most enthusiastic Ethereans will care enough to run their own validators.

And, it's not just about the effort involved. Staking pools and exchanges may offer better rewards and higher returns than solo staking -- at least in the short term.

PoS doesn't suffer from the same economies of scale as PoW. However, it's still possible for the largest and best-resourced stakers to receive outsized rewards. This is because of something called MEV (Maximum Extractable Value), which is essentially "free" money that can be made by carefully selecting and ordering the transactions in a block.

MEV is a fascinating but fairly complicated topic that deserves its own deep dive, so I won't go into details today. What matters here is that MEV can provide an additional revenue stream to the stakers who know how to access it. That requires either specialist knowledge or access to others with that knowledge. And that benefits the largest stakers because they are in a position to hire or partner with MEV specialists to maximise their income.

Now, to be fair to Ethereum, efforts are already underway to democratise MEV at the protocol level. This would reduce the disparity between professional and at-home stakers. Those updates won't be available for some time, but a similar third-party solution by Flashbots will be ready in time for the Merge. Validators will need to know about and opt-in to using this tool, but it should still reduce the bias towards the biggest stakers. So, perhaps we can be optimistic that this issue will be eliminated in time.

Even if a PoS blockchain can avoid centralisation through staking pools and exchanges, it may still suffer in another area. Major cloud computing platforms like AWS could become crucial to the blockchain's operation.

Some validators will surely run their own hardware but, just like staking itself, it may be easier to outsource this effort. Many will perceive cloud computing platforms to be an easier, safer, and more reliable option than anything they could run themselves.

The danger of many validators relying on a handful of cloud computing platforms is that each platform becomes a major point of failure. If a standalone validator failed, nobody would notice and Ethereum would continue unaffected. But, if AWS went down and 40% of validators were using it, then it would be a major problem. And technical issues aren't the only thing to worry about. It isn't far-fetched to think that cloud providers could suddenly purge validators from their platforms -- after all, one has already banned anything blockchain related.

Ethereum's PoS has been designed to discourage the use of things like AWS. If a cloud provider took a large number of validators offline, Ethereum would interpret that as an attack and economically penalise those validators. Therefore, it's smarter and safer for validators to avoid anything that could coordinate failures. But it remains to be seen if that incentive is enough.

Less Censorship Resistant

The possibility of centralisation increases the risk of something else: censorship. This has become a big talking point in recent weeks, and it's another topic that really deserves its own deep dive.

The idea is that staking pools and exchanges will be forced to comply with government sanctions. That might mean ignoring transactions from sanctioned addresses. It might even mean ignoring transactions from addresses indirectly related to sanctioned addresses. In the worst case, it would mean ignoring or rejecting blocks that contained those transactions.

That would be ideologically disastrous for Ethereum. Blockchains are designed to be neutral, accessible to all, and censorship-resistant. They are supposed to guarantee people's rights and freedoms, acting as a payment solution of last resort for marginalised groups. A censorship-embracing version of Ethereum would be seen by many as a perversion of something great.

And, there are more than ideological arguments against this. Perhaps you don't care about potential censorship. Perhaps you think it's a good thing. Those are completely valid opinions. But, you should still view this as a major risk. That's because the most extreme forms of censorship would likely lead to Ethereum fracturing into an array of regional networks. Block producers would be accepting or rejecting different blocks based on their government's guidelines and sanction lists. That would result in a vast array of different blockchain histories, each considered canonical by small groups of regional block producers. The idea of a single, global Ethereum network would be dead.

All things considered, this is probably the greatest threat to PoS blockchains.

Potential solutions have been discussed.

For example, Ethereans could react to censorship with something called social slashing. This would entail making a special rule change to reject blocks proposed by the censoring parties. Ethereum would then perceive those block producers to be inactive and penalise them. The validators would be forced to unstake their ETH or watch as it was whittled away with economic penalties. However, this approach has some major drawbacks. Chief among them is the contradiction of censoring the censors. Surely social slashing would also undermine confidence in Ethereum's neutrality and censorship resistance.

Luckily there are better options available. However, these require changes to the way Ethereum works and they likely won't be available for some time.

The good news is that censorship is unlikely to be as much of an issue as some suggest.

The influence of staking pools and major exchanges is certainly dangerous. Exchanges are at particular risk of caving to government pressure, and there's no equivalent for that under PoW. But the threat of social slashing, combined with the fact that rigorous and widespread censorship just isn’t practical or easy to implement, make anything close to the worst-case censorship scenarios extremely unlikely.

We may see weaker forms of censorship in the short term -- the kind where certain transactions are simply ignored by certain block producers. But that threat isn't new or unique to PoS, it can happen under PoW as well.

So, while PoS may be a little less censorship resistant than PoW, it probably (hopefully) doesn’t matter a whole lot.

Complexity

While PoS is conceptually simpler than PoW, it's much more difficult to design and implement an effective PoS algorithm.

PoS tends to require a greater quantity of more complicated code than PoW. That means more opportunities for bugs to creep in, and a larger surface area to attack. For this reason, some view PoS as less trustworthy and reliable than PoW.

Encourages Hoarding

The counterpoint to PoS's economic benefits is that it prevents a healthy distribution of coins by encouraging hoarding. Stakers are essentially paid not to sell, and their low running costs give them little reason to either. This is in complete contrast to PoW, where the high cost of mining compels miners to divest their holdings. Miners can also sell coins to reinvest in new equipment to increase future returns, increasing that incentive to sell.

We've seen that this dynamic can be good for the price of a PoS asset, and that will keep existing holders happy. But it won't be so good for newcomers looking to acquire their first coins. Over time, it's likely to create an unhealthy and extremely uneven distribution of coins, where wealth concentrates among a small group of early supporters. If Ethereum's fee burning does kick up a gear and ETH becomes deflationary, the effect could be particularly pronounced.

It's difficult to say how serious this issue could be. For cryptoassets that resemble shares in a company more than money, it's arguably not so bad. People rarely complain about a handful of individuals owning most of a company, after all. But, Ether is often promoted as a money. And, a money that is overwhelmingly owned by a tiny group of people doesn't seem particularly attractive. Even if ETH's 'money-ness' was limited to being a simple store of value, an immense concentration of wealth still wouldn't look good.

Ultimately, PoS's economic benefits may well be undermined by the hoarding culture it could create. The incentive to hold, combined with the lack of forced selling and fee burning may be detrimental to Ether's desirability.

Weak Subjectivity

There are a few arguments made against PoS that are, in my opinion, quite weak. However, they're brought up frequently, so it makes sense to cover them. The first of these is that PoS requires something called weak subjectivity.

Weak subjectivity means that users require outside information from a trusted source to ensure they are following the correct version of a blockchain.

If a user was presented with multiple conflicting versions of a PoS chain, they would select the "official" version by counting the number of historic votes on each chain. But they can't know if those votes are entirely legitimate. There's a chance they were cast by malicious validators who are no longer staking on the real blockchain. As a result, the user could be tricked into following an incorrect version of the chain. The solution is to create weak subjectivity checkpoints, which are like waypoints that everyone in the network agrees are valid. Taking these checkpoints in addition to the blockchain data they downloaded, the user would be able to determine the canonical chain. However, they would need to trust their source for these checkpoints.

PoW doesn't suffer from weak subjectivity, so these checkpoints aren't necessary. Miners can't cheat in the PoW guessing game, so they can't create fake votes on a false chain. When presented with multiple conflicting PoW chains, it will always be safe to select the one with the greatest total PoW. Therefore, PoW requires fewer trust assumptions than PoS.

PoS's proponents argue that the additional trust assumptions are minimal. After all, it's not difficult to verify a checkpoint with a few independent sources. Additionally, all blockchains -- including PoW chains -- already require similar trust assumptions. For example, most users trust the developers behind the blockchain software to act honestly.

I can see why a tiny group of purists and trust minimalists would object to weak subjectivity. But, for the vast majority of normal people, it's probably nothing to worry about.

Nothing At Stake

The nothing-at-stake problem is the other major criticism of PoS that is often repeated but doesn't really hold up. It's fair to say that it was a fundamental problem with early, naive attempts at PoS, but it's been solved for a while. In fact, genuine issues like PoS's complexity are a side-effect of solving this very problem.

Nothing-at-stake stems from the fact that staking isn't grounded in the real world like PoW.

With PoW, miners must allocate real resources to their efforts; it has real-world costs. If a miner is presented with multiple, conflicting versions of the chain, the rational thing to do is to mine only on the version they believe is valid. Mining on the invalid chain would be pointless, given it would almost certainly yield no reward but still require real-world resources.

Conversely, staking is not associated with any real-world costs. Staked coins exist entirely within the blockchain they secure. Therefore, if that chain was replicated across multiple competing versions, the coins would exist on all of them. As a result, a rational validator would attempt to produce blocks on every version of the chain they came across. In that case, they increase their chance of winning while risking nothing. There is nothing at stake to discourage them. And, that's a problem. Validators working across every branch they find would make a PoS chain easier and cheaper to attack. It would undermine the blockchain's security.

The solution to the nothing-at-stake problem is to make it extremely costly to stake on multiple versions of the chain. This is where slashing comes in. If someone submits evidence of a validator doing this, some portion of that validator's funds will be destroyed. Now there is something at stake, and validators' incentives are re-aligned with the blockchain.

Summary

Despite the nothing-at-stake problem being resolved with slashing, it's not uncommon to hear PoW advocates mention it as a major flaw in PoS's design.

Perhaps that's because the lack of real-world costs makes PoS feel like it's making something from nothing. Perhaps it's because it feels like PoS relies on an empty, circular logic of a blockchain being secured by its own native asset. Or, perhaps it's something else entirely.

Whatever the reason, any arguments about PoS being fundamentally broken are almost certainly wrong. Modern PoS systems have solved the issues that plagued early attempts. These days, the fundamental problems have been entirely resolved or reduced to relatively minor tradeoffs. The issues that remain are the ones we've examined today.

The Merge will bring many benefits to Ethereum, but it also brings new risks.

It may increase centralisation and therefore censorship. It adds complexity to Ethereum's code, increasing the chance of bugs. It could skew Ethereum's economics too heavily in favour of existing holders. And, for those who want to reduce trust to an absolute minimum, it introduces weak subjectivity to the blockchain.

Didn’t You Forget Something?

A lot has been said about the Merge over the past few months, and not all of it has been accurate. Depending on what you've heard and who you've listened to, you might be thinking I've missed something important. So, let's review some of the popular misconceptions about the Merge.

Staking Will Produce High Yields

Many Ethereans have claimed that the Merge will transform ETH into an ultra-attractive yield-bearing asset.

The argument is that after the Merge, everyone will be able to stake their ETH to earn generous real yield while experiencing minimal risk. That means they will earn returns well above Ether's inflation rate. In the most optimistic projections, stakers might earn double-digit yields while Ether's supply shrinks. Recent estimates tend to be more tepid, but still estimate yields of 5% or 6% against a roughly 0.5% rate of inflation. That's still pretty good. But, it's also not very realistic. In my opinion, this aspect of the Merge has probably been overhyped.

If staking rewards really are that generous -- and they may well be in the immediate aftermath of the Merge -- then many more stakers will come online to try and earn them. Not only would that dilute yields, but it would also increase Ether's inflation rate as more ETH would be minted to pay validators. Therefore, nominal yields would fall, and real yields would fall by an even greater amount!

At the end of the day, eye-poppingly high yields just aren't sustainable. Eventually, yields should settle at whatever value that the market feels is sufficient compensation for the risk and effort of staking. If most people perceive staking to be almost risk-free, then that won't be a large number. I struggle to see staking rewards exceeding 1% or 2% in the long run.

And that's not all. Staking yields will come from three sources: ETH issuance; the portion of transaction fees paid to block producers; and whatever MEV the block producer can extract.

ETH issuance is given to every validator that does their job properly. That means it is by far the most predictable portion of staking rewards. It also means it's more like a penalty against non-stakers than a reward for stakers. You don't really gain anything from it, you just don't lose.

Transaction fees and MEV will vary wildly from block to block. And, I suspect many are overestimating how much they could earn from them. Simply averaging total fees and MEV over the number of stakers will give an over-optimistic yield projection that most stakers will never earn.

There are two main reasons for this. First, in any given time period, many stakers won't get an opportunity to propose a block. Therefore, they won't get an opportunity to earn these rewards.

Second, the stakers that do get to propose blocks aren't guaranteed to receive anything close to the 'average' fees and MEV from them. Looking historically, a large portion of total MEV and fees are captured in a small portion of blocks. And, that makes sense. Moments of extremely profitable on-chain activity don't happen very often. The chance of a given validator proposing a block on one of those occasions is pretty low. Therefore, most of the time, earnings from fees and MEV will be substantially less than the averages suggest. As a result, many stakers could be left disappointed with their earnings.

Overall, I think this is a case of the truth being stretched a little too thin. Ethereum stakers will earn yield, and sometimes it might be quite generous. But high yields won't be long-lasting or widespread. This isn't a game changer that will turn heads in the world of traditional finance. It's just another, relatively minor benefit of the Merge.

The Merge Will Reduce Fees

The worst case of misinformation around the Merge is probably the idea that it will reduce Ethereum's transaction fees. It won't do that.

This myth may come from the fact that Ethereum will produce blocks slightly faster after the Merge. Under PoW, Ethereum produces a block about once every 13 seconds. Under PoS, it can produce a block every 12 seconds. So, technically, Ethereum's capacity will increase. But, it's a negligible change that will have no perceptible effect on fees.

For those that are disappointed by this, it's worth noting that Ethereum has an entirely separate set of plans for scaling and making transactions more affordable. Most of those plans revolve around layer 2 platforms like Optimism, Arbitrum, and zkSync -- which are all useable today. There will also be a number of future updates focused on scalability. Unfortunately, the Merge is not one of them.

So, Is It Worth It?

For me, the Merge does seem like a worthwhile and beneficial upgrade. However, I'm not as excited about it as others seem to be.

The environmental benefits are certainly nice to have. Even accounting for the potential benefits, it is difficult to justify PoW's energy consumption when PoS does effectively the same thing more efficiently.

However, I'm sceptical that the Merge will do much to change opinions and end criticism of Ethereum.

I suspect many of the loudest critics of crypto's energy use don't actually care about energy use. Instead, they just dislike cryptoassets and are looking for a stick to bash them with. After the Merge, they'll either find something new to complain about or will just keep talking about energy use anyway. After all, facts don't matter when you hate something that much.

Wider audiences probably don't care about energy use either. Maybe I'm overly cynical, but I think most people either don't care at all or don't care enough to avoid Ethereum if they had a compelling reason to use it. Much of the talk online is just people saying what they feel social pressure to say.

For that reason, I don't consider the environmental benefits of the Merge to be a game-changer in the way that some suggest. I don't think it will unleash a flood of new, climate-conscious users because those users don't really exist. Don't get me wrong, this isn't a bad thing. It can't hurt Ethereum to respond to and remove one of the most frequently cited criticisms against it. But, I struggle to see it making the difference that many are hoping for.

The economic benefits are more interesting.

The significant reduction in ETH issuance should help it compete against Bitcoin as a gold-like store of value. That's especially true once you account for the possibility of deflationary days thanks to burned transaction fees. With this tightening of supply, it's easy to see why so many people are bullish on ETH over the medium- and long-term.

Still, I think criticisms of this model are valid. It won't be a good look if Ethereum's wealth is concentrated among a tiny number of stakers. It's also worth questioning if Bitcoin's 'hard money' narrative actually makes sense for Ethereum. Are people more likely to build on or use Ethereum because of it, or could it even put them off?

I'm interested to see how this one plays out. In all likelihood, I'm overthinking things and it won't make much difference either way. But, I can't help but worry that post-Merge economics favour existing ETH holders a bit too much.

The risk of centralisation is my biggest concern by far.

While I can see how staking could further decentralise control of the network, I don't see that as a likely outcome. In a few years, I fear a handful of exchanges and staking pools will utterly dominate the staking landscape. After all, they already control a significant portion of staked ETH today -- when you'd expect solo staking to be near its peak. The question then becomes, what will they do with their immense influence?

Probably not that much, all things considered. I've already said I don't expect to see worst-case censorship. Weaker censorship is certainly possible, but won't be drastically worse than what could happen under PoW. Plus, Ethereum should make changes to eliminate that risk in the longer term. There's also no risk of stakers corrupting governance and forcing through unwanted changes because Ethereum doesn't use on-chain governance.

The bigger risk is that centralisation makes Ethereum less reliable and resilient. If 95% of validators were run by 3 entities, the entire network would feel the effects of one of them going offline. That's a problem.

Additionally, it would just be sad to see Ethereum captured in that way. Outside of staking, Ethereum has been painstakingly designed to enable maximum decentralisation. A big part of Ethereum's culture is about promoting decentralisation. It has the feeling of a "by the people, for the people" project. Anything less than decentralised block production would feel like a failure. And yet, it seems inevitable.

I really hope I'm wrong on this. I hope something happens to promote solo staking. But, I struggle to see a world in which a large number of people choose to run their own validators instead of leaving that to someone else. That may not compromise Ethereum or leave it at risk of censorship. But, it would feel like a massive and rather disappointing underachievement.

Altogether, there are certainly question marks and minor concerns around Ethereum's shift to PoS, but nothing about it appears debilitating. At the same time, it should bring a swath of small benefits. Ethereum's security, economics, and public image should all improve after the Merge. And, if nothing else, the Merge clears the way for a number of other, more exciting upgrades down the line.

To me, the Merge is predominantly an incredible technical achievement. It just happens to come with these minor benefits and the potential to pump ETH's price. I understand the excitement of others, I just don't think it's entirely warranted.

So, that’s what the Merge will do. To find out how the Merge will work, check out part three.