You misunderstood, mining centralization will not break Bitcoin #2.

CoinDesk columnist Hasu sparked meaningful discussions on Twitter after releasing the first column. Hasu collected some interesting questions and wrote the second installment to further clarify misconceptions about miners and mining mechanisms.

CoinDesk columnist Hasu released the first column article and sparked meaningful discussions on Twitter. Hasu collected some interesting questions and wrote the second episode to clarify misunderstandings about miners and mining mechanisms.

The following is the translation of the original article and commentary:

Yesterday, I wrote a column article for CoinDesk, outlining that the concentration of mining power will not break the Bitcoin system. The reason is that miners with significant computational power would incur more losses if they were to attack the network or protocol in some way, based on opportunity cost considerations. (The same argument applies to validators in PoS).

The article sparked meaningful discussions on Twitter, and I spent the whole day engaging with different people. Thank you very much for reading the article; this is indeed what a writer hopes for. In this follow-up article, I want to take the time to address some excellent questions and countermeasures.

Isn't it better if mining power is more decentralized?

Absolutely. I regret not explaining this more clearly in the previous article. Widely distributed mining power is definitely superior to concentrated mining power, but even if the mining power is concentrated, it will not harm the system.

You can think of Bitcoin as having different layers of security mechanisms. When you remove one layer, the security of the system decreases but does not collapse immediately. Imagine having ten miners, each controlling 10% of the computational power and never colluding with each other. In this scenario, most attacks are practically impossible. An attacker would need to overpower the computational power of the other nine miners individually, which is the best solution for network security.

However, even if miners control more computational power than this, Bitcoin still has a good attribute where the risk value of miners is linearly proportional to their computational ability in the network. Therefore, the incentive to be honest still exists because misconduct would result in severe losses, while honesty would yield block rewards.

Nic Carter 【Note: Co-founder of the on-chain data website CoinMetrics】 provided a great summary:

"Bitcoin consensus is effectively protected by two layers of defense.

#1 Controlling computational power is difficult to reach the level that affects network security

#2 Even if controlling the network, disrupting the network is not cost-effective"

Many people are not aware of the existence of the second point.

The general understanding is that as long as a single entity controls more than 50% of the computational power, the network becomes insecure. Of course, we hope for more decentralized mining power, but our ability to influence this is limited.

If centralization of mining power cannot harm Bitcoin, why do systems with lower overall computational power get attacked?

Ethereum Classic (ETC) and Bitcoin Gold (BTG) are the two most famous examples of cryptocurrencies in the top 50 assets that have been attacked.

The intuitive response might be to attribute this to the difference in overall computational power (Bitcoin is 175 times larger than ETC and 985 times larger than BTG). I believe that this is not the main reason these systems are insecure; it is due to miners lacking a mandatory long-term commitment. Both ETC and BTG systems endorse the fallacy of ASIC resistance, and this goal of resisting ASICs aims to ensure that hobbyist miners can compete to some extent by using cheaper hardware like GPUs.

However, the benefits of fairness brought by ASIC resistance significantly impact the security of the system. Bitcoin ASIC miners are specialized devices that have no utility other than mining Bitcoin. If Bitcoin were to disappear, their market value would drop to zero. Miners in the Bitcoin ecosystem need to have a significant number of Bitcoin ASIC miners, so their profits are inherently tied to the health of the Bitcoin network.

On the other hand, GPU miners do not have this specificity and are accessible to amateur enthusiasts. Most people have GPUs in their gaming PCs, which can be used for gaming or mining other GPU-supported cryptocurrencies (or sold to other gamers or GPU miners). Their value is not tied to the value of a specific token.

Furthermore, as GPUs have many uses (such as machine learning or video processing), there exists a computer equipment market where such general-purpose hardware can be rented. Attacks on ETC and BTG were executed through renting hardware, allowing miners to acquire temporary computational power over the network without having to commit long-term to a specific network.

The reason why certain tokens with lower computational power are at risk is that attacking the system does not incur economic losses for miners, largely due to the fallacy of ASIC resistance.

Thus, Bitcoin miners are financially bound to the network, and attacking the network is akin to hurting oneself more than the enemy.

Can the Chinese government not force miners to attack Bitcoin at zero cost?

In fact, I believe this is incorrect because just like private entities, nations incur opportunity costs by attacking Bitcoin. Easily disrupting profitable industries like mining would sacrifice their future tax revenues.

Moreover, Bitcoin mining is well-suited for countries with abundant cheap energy but experiencing overcapacity domestically. In the past, the concept of refining aluminum was essentially exporting electricity from low-cost power countries to high-cost power countries.

【Note: Refining aluminum requires a significant amount of electricity, so importing aluminum from low-cost power countries is akin to importing cheap electricity.】

Suppressing the mining industry would also undermine local trust in property rights and the rule of law, which are crucial drivers of economic prosperity and foreign investment.

Additionally, it is essential to emphasize that for attackers willing to bear unlimited opportunity costs, no system is secure. If the US or China were determined to destroy Bitcoin at any cost, we would be powerless. However, we can deter them as much as possible by increasing the cost to prevent them from harming Bitcoin, which is our most potent deterrent.

Unless we start scrutinizing miners (moving the system to permissioned), we cannot prevent the centralization of computational power. However, this raises the question of how Bitcoin with a single miner differs from centralized companies like Paypal.

Doesn't the argument of "the more power, the greater the loss" also apply to centralized systems?

Before we begin, let me address the first answer. We all do not want computational power to be centralized, but we cannot refute the fact of centralization. Do not assume what cannot be proven!

In the way mining works, any miner could broadcast their colluded secret chain at any time. Of course, this also applies to a brand-new miner. Models that do not consider miner collusion are unrealistic and significantly lower security.

While you may not agree with what I am about to say, in this assumption, expending so much electricity on PoW is justified. If I were to attempt to provide a different explanation, I would tell you that even if all computational power is controlled by a single miner currently, Bitcoin differs significantly from systems like Paypal or commercial banks. Therefore, the worst-case scenario is not as dire as you might imagine.

1 Bitcoin is entirely auditable and trustless. Users can verify whether central miners comply with network rules, regardless of whether there is one miner or a thousand miners, these rules still apply. Users can also transparently assess the miner's work, as they can detect when a miner initiates a double-spend attack or censors transactions.

2 Users can easily exit. In the Bitcoin system, if users are dissatisfied with the current miner's work, they can collectively fork to another PoW algorithm to discard it, which is much easier than switching collectively to a Paypal competitor (or one that may not even exist). All users have read access to the ledger (UTXO set), making it easy to access that state and leave it when necessary, whereas if Paypal fails, even if someone launches a new Paypal, you cannot retrieve your balance from Paypal.

3 Governments have less influence over miners. You may now stand from a capitalist perspective, thinking that those harming users are unlikely the companies themselves—this is usually the result of government intervention, whether through direct regulation or indirect pressure.I agree that nations are the biggest threat to Bitcoin because they view it as a threat to their currency and financial sovereignty. However, the impact of nations on Paypal or commercial banks is fundamentally different from their impact on miners. Let's analyze the reasons:

a. Influence is inversely proportional to liquidity. Those who can easily leave are less susceptible to coercion, which applies to miners! If miners feel that local policies are unfavorable to them, they can effortlessly move to another country. Therefore, the impact of local policies on miners is not as significant as perceived. Moreover, we have seen many countries establishing special economic zones for the mining industry.

b. Miners can be replaced. If local governments cause the central miner to "disappear," it's a time when entering the mining industry with zero barriers is possible, as other countries' miners can easily continue mining through the network. Thus, governments must realize that engaging in malicious acts like confiscating mining equipment and using it for attacks can only be done once. Ultimately, a miner can always be organically disrupted by another miner who can earn more money. For example, if the old miner does not process certain transactions (censorship), they would incentivize new miners to process these transactions to earn money.

(Thanks to Raphael Auer, figo, and latetot for their questions)

Further Reading

• You misunderstand: Concentration of miners will not break Bitcoin
• Unknown mining pool's computational power nears 50%, Bitcoin Cash faces threats

Join Telegram now for the most comprehensive information on fintech, blockchain insights, and industry examples!