19/5/18: The Scary Inefficiency & Environmental Costs of Bitcoin

Bitcoin is just one of the cryptocurrencies, albeit the dominant one by market capitalisation and mining assets deployment. The cryptocurrency is best known for volatility of its exchange rate to key fiat currencies and other commodities, but the more interesting aspect of the Bitcoin (and other cryptos) is their hunger for energy. Cryptos are based on blockchain technologies that promise a range of benefits (majority unverified or untested or both), amongst which the high degree of security and peer-to-peer data registry, both of which are supported by the mining processes that effectively require deployment of  a vast amount of hash/algorithmic calculations in order to create data storage units, or blocks. In a sense, energy (electricity) is the main input into creation of blockchain records of transactions.

As the result, it is important to understand Bitcoin (and other cryptos) energy efficiency and utilisation, from three perspectives:
1) Direct efficiency - value added by the use of energy in mining Bitcoin per unit of BTC and unit of information recorded on a blockchain;
2) Economic efficiency or opportunity cost of using the energy expended on mining; and
3) Environmental efficiency - the environmental impact of energy used.

To-date, estimating the total demand for electricity arising from Bitcoin mining (let alone from mining of other cryptos) has been a huge challenge, primarily because Bitcoin miners are too often located in secretive jurisdiction, do not report any data about their operations and, quite often, can be highly atomistic. Although Bitcoin mining is a concentrated activity - with a small number of mega-miners and mining pools dominating the market - there is still a cottage industry of amateur and smaller scale miners sprinkled around the globe.

Thus, to-date, we have only very scant understanding of just how much of the scarce resource (energy) does the new industry of cryptos mining consume.

A new paper, published in a peer-reviewed journal, Joule, which is a reputable academic journal, titled "Bitcoin's Growing Energy Problem" and authored by Alex de Viries (Experience Center of PwC, Amsterdam, the Netherlands) attempts exactly this. The paper is the first in the literature to be peer-reviewed and uses a new methodology to discern trends in Bitcoin's electric energy consumption. The paper does not cover other cryptos, so its conclusions need to be scaled to estimate the entire impact of cryptocurrencies energy use.

The findings of de Viries are striking. He estimates the current Bitcoin usage of energy at 2.55 gigawatts, close to that of Ireland (3.1GW), approaching 7.67GW that "could already be reached in 2018", comparable to Austria (8.2GW). When reached, this will amount to 0.5% of the total world electricity consumption.

Per 'efficiency of blockchain', a single transaction on Bitcoin network uses as much electricity as an average household in the Netherlands uses in a month. Which is, put frankly, mad, wasteful and utterly unrealistic as far as transactions costs go for the network.

Per de Viries: "As per mid-March 2018, about 26 quintillion hashing operations are performed every second and non-stop by the Bitcoin network (Figure 1). At the same time, the Bitcoin network is only processing 2–3 transactions per second (around 200,000 transactions per day). This means that the ratio of hash calculations to processed transactions is 8.7 quintillion to 1 at best. The primary fuel for each of these calculations is electricity."

The key to the above numbers is that they vastly underestimate the true costs of Bitcoin and other cryptos to the global economy. The paper focuses solely on energy used on mining. However, other activities that sustain Bitcoin and blockchains are also energy-intensive, including trading in coins/tokens, storage of information blocks, etc. Worse, mining and processing / servicing of the networks required use of constant electricity supply, which means that the energy mix that goes to sustain cryptocurrencies operations is the worst from environmental quality perspective and must rely on heavy use of fossil fuels in the top up range of electricity demand spectrum. The environmental costs of Bitcoin and cryptos is staggering.

Scaling up Bitcoin figures from de Viries; paper to include other major cryptocurrencies would require factoring in the BTC's share of the total crypto markets by energy use. A proxy (an imperfect one) for this is BTC's total share of the cryptocurrencies publicly traded markets which stood at around 37.3% as of May 16, 2018. Assuming this proxy holds for mining and servicing costs, total demand for electricity from the cryptocurrencies and blockchain use around the world is more than 2.55GW/0.37 or more than 6.9GW, with de Viries' model implying that by year end, the system of cryptocurrencies can be burning through a staggering 1.35% of total electricity supply around the world.

The problem with the key cryptocurrencies proposition is that the system of blockchain-based public networks can deliver lower cost, higher efficiency alternatives to current records creation and storage. This proposition simply does not hold in the current energy demand environment.



The full paper can be read here: de Vries: "Bitcoin's Growing Energy Problem" http://www.cell.com/joule/fulltext/S2542-4351(18)30177-6.

17/1/17: Economics of Blockchain

One of the first systemic papers on economic of blockchain, via NBER (http://www.nber.org/papers/w22952) by Christian Catalini and Joshua S. Gans, NBER Working Paper No. 22952 (December 2016).

In basic terms, the authors see blockchain technology and cryptocurrencies influencing the rate and direction of innovation through two channels:

1. Reducing the cost of verification; and 
2. Reducing the cost of networking.

Per authors, for any "exchange to be executed, key attributes of a transaction need to be verified by the parties involved at multiple points in time. Blockchain technology, by allowing market participants to perform costless verification, lowers the costs of auditing transaction information, and allows new marketplaces to emerge. Furthermore, when a distributed ledger is combined with a native cryptographic token (as in Bitcoin), marketplaces can be bootstrapped without the need of
traditional trusted intermediaries, lowering the cost of networking. This challenges existing
revenue models and incumbents's market power, and opens opportunities for novel approaches to
regulation, auctions and the provision of public goods, software, identity and reputation systems."

A bit more granularly, per authors,

• "Because of how it provides incentives for maintaining a ledger in a fully decentralized way, Bitcoin is also the first example of how an open protocol can be used to implement a marketplace without the need of a central actor." In other words, key feature of cryptocurrencies and blockchain is that it removes the need to create a central verification authority / intermediary / regulator or repository of data. The result is more than the cost reduction (focus of the Catalini and Gans paper), but the redistribution of market power away from intermediaries to the agents of supply and demand. In other words, a direct streamlining of the market away from third parties power toward the direct power for economic agents.
• "Furthermore, as the core protocol is extended (e.g. by adding the ability to store documents through a distributed ledger-storage system), as we will see the market enabled by a cryptocurrency becomes a  flexible, permission-less development platform for novel applications." Agin, while one might focus on reductions in the direct costs of innovation in that context, one cannot ignore the simple fact that blockchain is resulting in reduced non-cost barriers to innovation, further reducing monopolistic market power (especially of intermediaries and regulators) and diffusing that power to innovators.

So what are the implications of this view of economics of blockchain? "Whereas the utopian view has argued that blockchain technology will affect every market by reducing the need for intermediation, we argue that it is more likely to change the scope of intermediation both on the intensive margin of transactions (e.g., by reducing costs and possibly influencing market structure) as well as on the extensive one (e.g., by allowing for new types of marketplaces)." So far, reasonable. Intermediation will not disappear as such - there will always be need for some analytics, pricing, management etc of data, contracts and so on, even with blockchain ledgers in place. However, the authors are missing a major point: blockchain ledgers are opening possibility to fully automated direct data analytics and AI deployment on the transactions ledgers. In other words, traditional forms of intermediation (for example in the context of insurance contract transactions, those involving data collection, data preparation, risk underwriting, contract pricing, contract enforcement, contract payments across premia and payouts, etc) all can be automated and supported by live data-based analytics engine(s) operating on blockchain ledgers. If so, the argument that the utopian view won't materialise is questionable.

The paper is worth reading, for it is one of the early attempts to create some theoretical framework around blockchain systems. Alas, my gut feeling is that the authors are failing to fully understand the depth of the blockchain technology.