Bitcoin's Achilles Heel

Introduction

More than a decade after the revolutionary whitepaper by Satoshi Nakamoto[1], bitcoin and the blockchain are still shaking the core foundations of how humans perceive money. Bitcoin’s disruptions in major traditional industries cannot be understated. Despite the groundbreaking feats of blockchain technology, it has failed to achieve its most important promise of one-CPU-one-vote, the core premise upon which its consensus protocol is built. Bitcoin has evolved into a store of value, rather than a medium for undertaking regular everyday transactions. On average, bitcoin is able to process about 4.6 transactions per second[2] in contrast to Visa’s 1,700 transactions per second[3]. This means that it may take minutes or even hours depending on network congestion, to make a bitcoin transaction. While there have been a number of spinoff cryptosystems intent on fixing the blockchain’s failings, the ultimate dream of a truly decentralized peer-to-peer cryptosystem is far from reality. It is worth noting that the blockchain's inherent weakness is not a bug but a feature of how its proof-of-work (PoW) consensus protocol is implemented. This essay seeks to explain bitcoin’s PoW and how it is responsible for bitcoin’s scalability issues, and to a greater extent waste electric and computing power.

Proof-of-Work (PoW)

The idea of PoW was invented in 1993 by Cynthia Dwork and Moni Naor in a conference article titled “Pricing via Processing or Combating Junkmail”[4]. The article proposed a mechanism for dealing with the frivolousness of junk mail by requiring a participant to compute a selectable function in order to gain access to a resource. Dwork and Naor noted that this mechanism could be implemented for general use in controlling access to a shared resource. For bitcoin, the shared resource is a distributed peer-to-peer PoW chain (blockchain), a chain of the hash of a block (ledger) of transactions. When nodes receive broadcast transactions, they work to create (mine) blocks by expending CPU power to compute the PoW function based on a predetermined difficulty. Using CPU power as votes, other nodes may express acceptance of a valid block by working on extending it and refusing an invalid block by refusing to work on it. The majority (51%) of nodes determine the course of the chain. This consensus protocol prevents double-spending even in the absence of a trusted entity. By convention, incentives are baked into the network in the form of coins. The first transaction in a block is a special transaction that starts a new coin credited to the owner of the block.  While incentives are the best reward for nodes supporting the network, the competitive nature of how blocks are mined, has become the network’s own achilles heel.

The blockchain operates a traditional competitive winner-takes-all type of cryptosystem. Two main problems arise from this type of structure. The first problem is continuous centralization. This arises when only nodes with large CPU powers among the lot are able to mine blocks. Coin rewards from PoW lead to a scramble for incentives as with any system designed around merit. The scramble inherently increases the mining difficulty at the behest of nodes with relatively small computing power. As it stands now, about 1% of nodes on the blockchain control about 99% of the computing power expended on the network. Running a blockchain node on a PC with a regular CPU can be described as crazy at best. The convention in recent times is to acquire special-purpose GPU miners which are way faster and more efficient than regular CPUs. Putting many individual nodes together in a pool to form a single node is the other ‘profitable’ alternative. On the blockchain, motivation for incentives supersedes that of pure enthusiasm and enthusiasm is essential for innovation. How can nodes, programmed to be divided against each other work to support the network? As our Lord Jesus said in Matthew 12:25, “Every kingdom divided against itself will be ruined and every city or household divided against itself will not stand”[5].

The second problem is energy inefficiency. When nodes work to mine the next block in the chain, only one node wins. The resources expended by the loser nodes in trying to compete are wasted. It is estimated that the blockchain wastes about 78 terawatt-hours of electricity annually. To put this into context, it is about four times Ghana’s electricity consumption (~0.093 TWh) in 2019[6]. Climate scientists warn that bitcoins emissions alone could have a significant effect on the environment[7].

Conclusion

We have explained proof-of-work and how such a consensus mechanism is implemented in the blockchain network. It has been established that blockchain PoW works in a competing fashion and that incentives are awarded in a winner-takes-all manner. These features are responsible for bitcoin’s energy inefficiency and continuous centralization.

References

1. Nakamoto, S., 2019. Bitcoin: A peer-to-peer electronic cash system. Manubot.

2. Kenny, L., 2019. The Blockchain Scalability Problem & the Race for Visa-Like Transaction Speed. https://towardsdatascience.com/the-blockchain-scalability-problem-the-race-for-visa-like-transaction-speed-5cce48f9d44

3. Visa. https://usa.visa.com/run-your-business/small-business-tools/retail.html

4. Dwork, C. and Naor, M., 1992, August. Pricing via processing or combatting junk mail. In Annual international cryptology conference (pp. 139-147). Springer, Berlin, Heidelberg.

5. Matthew 12:25, Holy Bible, New International Version.

6. Energy Commission of Ghana. 2020. National Energy Statistics. http://energycom.gov.gh/files/2020%20ENERGY%20STATISTICS-revised.pdf

7. Mora, C., Rollins, R.L., Taladay, K., Kantar, M.B., Chock, M.K., Shimada, M. and Franklin, E.C., 2018. Bitcoin emissions alone could push global warming above 2 C. Nature Climate Change, 8(11), pp.931-933.