Impact of Mining Pools on De-centralization

Figure 7.1 depicts the distribution of computing power among mining pools in the Bitcoin network between October 10 and October 14, 2015. Our findings show that more than 75% computing power in Bitcoin was controlled in the investigated period by five major centralized mining pools. If these pools were to collude in order to acquire more than 50% of computing power share in the network, they could effectively control the confirmation of all transactions occurring in the system. This includes preventing transactions from being executed, approving a specific set of transactions, and double-spending transactions [39]. Moreover, given the results of [40-42], F2Pool and AntPool, which control around 38% of the computing power in the network, can further considerably increase their advantage in the network by performing selfish mining; in the worst case, these two mining pools can determine together the fate of all transactions in the Bitcoin system.

Note that conscious miners can play an important goal in preventing a mining pool from controlling the computing power in the network. For example, miners of the Ghash.io pool actively abandoned the pool in 2014 in the fear of allowing the pool operator to acquire more than 51% of the computing power in the network [43]. This has forced Ghash.io’s owner to issue a public statement reassuring the community that the mining pool will make sure to take “all necessary precautions” in order to avoid controlling 51% of the computing power in the network.

We however point out that it suffices that a given mining pool controls the computing power in the network for a short amount of time (i.e., before raising the suspicion of miners) in order to cause considerable damage in the network. Namely, a malicious pool could, for example, double-spend all transactions occurring within that period of time or selectively reverse payments—a misbehavior that might be economically justified when reversing/ill-spending large amounts.

While most existing mining pool protocols assume the existence of a logically centralized operator that orchestrates the block generation process, a number of fully decentralized mining pools, such as P2Pool, have been proposed. Such pools share the benefits of centralized pools since all the participating users get regular payouts that reflect their contribution toward generating a block. However, these pools do not require the existence of any centralized coordinator and operate in a completely decentralized fashion. Currently, P2Pool only holds a marginal share of the computing power in the network; Bitcoin users can only hope that such decentralized pools can be transformed into profitable businesses in the near future in order to attract most miners. However, the Bitcoins reward mechanism provides no particular incentive for users to use such decentralized alternatives.

Currently, members of mining pools need to frequently submit cryptographic proofs to the pool operator in order to demonstrate that they are indeed contributing their computing power to the benefit of the pool. This mechanism is used to enhance trust among different untrusted pool members. In [44], Miller et al. observed that if the proof-of-work would effectively enable the miner to steal the reward without leaving any evidence, then miners do not have any incentive to join pools. Namely, any pool operator wishing to outsource the proof-of-work risks losing the mining reward. Miller et al. then proposed two constructs of such nonoutsoureable puzzles, which ensure that even if there are legal contracts between the pool operator and the miners, then miners can effectively steal the awards without leaving any evidence of misbehavior. Implementation results show that these puzzles add additional but tolerable overhead to the cost of Bitcoin blockchain validation.

Although few mining pools clearly control the computing power in the network (see Figure 7.1), we argue that there is still hope for reducing the impact of mining pools on the Bitcoin system. Given that Bitcoin relies on the notion of controlled supply that effectively limits the total number of generated BTCs (i.e., the amount of BTCs that are generated for each block is halved every 4 years), the ultimate dominance of mining pools is expected to decrease with time, since their profits would depend less and less on self-awarded BTCs and more on transaction fees. This, in turn, also increases the contribution of individual users to the Bitcoin economy. Indeed, mining pools operators would then have less incentives to accept client versions that are adopted by the minority of the clients (see Section 7.6). Users would then contribute more to the decision making process in Bitcoin by deciding to adopt a client version that suits their preferences. Recall that since the Bitcoin source code is open source, there are already a considerable number of different Bitcoin implementations [39].