What would happen if a mining pool admins having majority of hash power turns evil?

Question

This question interest me mostly in the context of the Ethereum, but it might be interesting to see how this problem is solved by the other cryptos as well.

I understand that the consensus about what is included in the ledger of the next block is established by the game-theoretic disadvantage of the miners who mine a block that includes ledger entries which are likely to be disputed by majority of the other miners.

But as a mining pools get bigger and sometimes advocate using their own software to mine the coins, once their hash rate exceeds 50% they will have a power to form their own consensus, based on the whim of the admins of the mining software.

With that power they will be able to block certain transactions to get into the chain, effectively freezing assets on any accounts they don't like and disturbing the execution of the smart contracts.

This situation can happen even more easily, when the ETH would switch to the proof-of-stake, since it is easier to accumulate a capital rather than to accumulate the hardware.

What are (if any) provisions in ETH and other crypto currencies against such scenario?

---

I find it is strange and somewhat disturbing, that such a fundamental question did not got answered. AFAIK the situation I described is very real concern for the bitcoin: https://www.cs.cornell.edu/%7Eie53/publications/btcProcFC.pdf . How does the Ethereum fare?

Answer 1

After reading two papers: https://blog.ethereum.org/2015/01/28/p-epsilon-attack/ and https://www.cs.cornell.edu/%7Eie53/publications/btcProcFC.pdf I conclude, that the neither ETH nor BTC (and most if not all any other crypto currencies) are secure against such attacks. It is also not secure against a so-called Schelling scheme, i.e. attack made not by the party that controls a big pool of miners, but simply a party that has enough coins to bribe miners into mining his own double-spending block. If the attacker is able to bribe majority of the miners, such an attack can be zero-cost for the attacker.

Specifically, "The P + epsilon Attack" paper concludes:

Of course, in practice, proof of work has survived despite this flaw, and indeed it may continue to survive for a long time still; it may just be the case that there’s a high enough degree of altruism that attackers are not actually 100% convinced that they will succeed – but then, if we are allowed to rely on altruism, naive proof of stake works fine too. Hence, Schelling schemes too may well simply end up working in practice, even if they are not perfectly sound in theory.

I hate to be a bearer of bad news, but the world with affordable, decentralized and deregulated finance services Ethereum and other cryptcurrencies bring might not be as secure, as we might want think.

---

Another sad update in this article: Decentralization in Bitcoin and Ethereum Networks, 11 Jan 2018