Bitcoin for skeptics

It is hard to avoid hearing about Bitcoin and other so-called cryptocurrencies these days. Almost every issue of every newspaper has an article about some aspect of Bitcoin. As a researcher in cryptography I have been interested in Bitcoin since around 2013 (though I confess to have never "mined" a block or invested in Bitcoin). The purpose of this article is not to explain the technology of Bitcoin or to give a detailed overview of it. And I am definitely not giving investment advice! Instead I want to highlight a few aspects of the Bitcoin story that I think are of particular interest to skeptics.

History

The inventor of Bitcoin is Satoshi Nakamoto, which is a pseudonym for some individual or group of people whose identities are not publicly known. A paper "Bitcoin: A Peer-to-Peer Electronic Cash System" appeared online in late 2008 and the Bitcoin software was available and the currency launched in early 2009. The main purpose of Bitcoin is to be a decentralized money that is not managed by a central bank or any other trusted authority. Since 2009 there have been a large number of other cryptocurrencies (I prefer the name cybercurrencies) created with varying success.

There are a number of speculations about the identity of Satoshi Nakamoto, some based on quite sophisticated textual analysis and others on the most flimsy of coincidences. I refer to Wikipedia for an overview: https://en.wikipedia.org/wiki/Satoshi_Nakamoto

What is clear is that Satoshi Nakamoto is very smart and very rich. I find it quite remarkable that it has been possible for such a person or people to keep their identity totally secret (unlike say Banksy, whose identity is presumably well-known to both a large number of people in the art world and also various government authorities).

The first main usage of Bitcoin came from trading illegal goods (e.g., on the Silk Road website). Over the last few years Bitcoin has been used less as an instrument of trade and primarily as an instrument of speculation.

The most important thing to understand about Bitcoin is that it is not a physical object. Essentially a Bitcoin is an ability to spend. Technically, to "own" a Bitcoin means to own a private key. This secret gives a person the ability to generate a digital signature with respect to a certain public key, and hence transfer ownership to another person. (It is my interest in public key cryptography and digital signatures that got me interested in Bitcoin). If you want to learn more about the technical background of Bitcoin then there are many online resources and tutorials.

A fundamental concept in cybercurrencies is the blockchain, which is a fancy name for a public ledger of transactions. To understand this consider other financial instruments. With cash there is no public ledger of transactions. I can withdraw a note from an ATM and spend it in shop A. Then later, customer X can be given that note in change and spend it in shop B. Then that shop can deposit it at a bank. There is no public list of transactions that allows anyone to see that I spent some money in shop A and that person X spent money in shop B. Further, even though my bank keeps a record of serial numbers of notes issued by ATMS, my bank cannot determine who is X or which shop was A.

With credit cards, the bank knows exactly when and where I have spent money, and provides me with a detailed monthly ledger of my transactions. But I trust the bank to not provide this information to anyone else.

With Bitcoin every transaction is public. If a payment is not included in the blockchain then, by definition, it did not happen. A confusing property of Bitcoin is that it is often called anonymous. This is to an extent true, as people's names are not listed in the blockchain. What is listed are identifiers that are associated with public keys. A payment shows that a certain value of Bitcoins has been transferred from one identifier to another. The transaction is public and one can determine which identifiers are rich people, but the identity of the actual people is unclear. This is why we still don't know who Satoshi Nakamoto is, even though we know he/she/they must be rich. There are other cybercurrencies, such as Zcash, that achieve much stronger notions of anonymity.

Blockchain is currently a huge buzzword in business. An extraordinary number of companies internationally and in NZ are exploring how they can use blockchains to streamline their operations. I'm personally a bit uncertain about how ubiquitous blockchains are going to become, but I have colleagues who believe they are an amazingly promising technology. As with any cryptography solution, once they are working well they will be totally invisible to average users.

The other major buzzword in Bitcoin is "mining". This is nothing like digging for gold. Mining means maintaining the public ledger. A better name would be "book-keeping". The process involves checking, for each transaction in a block, that the digital signature is correct and that the spender has the available funds. Just like any book-keeping, this tedious job is rewarded by payment. One of the genius ideas in the design of Bitcoin is that the reward for book-keeping is the creation of new Bitcoins, and that's why the word "mining" was chosen to describe the whole process.

Because book-keeping is not so difficult there is a risk that too many people would try to do it simultaneously and all of them would demand the reward of a Bitcoin. This would have several bad consequences such as forks in the blockchain and devaluation of the currency. Hence another feature of blockchain is to make mining hard by including a "proof-of-work". It is an absolute necessity of Bitcoin that managing the ledger be hard work and take time. Three major consequences are that transactions are not instantaneous, that only a small number of transactions can take place in any given time period and that Bitcoin uses a lot of energy. More on the energy use later. Due to the restrictions on transactions, Bitcoin is not scalable; it cannot handle anywhere near the number of transactions currently handled by Visa or Mastercard, for example.

One final remark: There is a finite supply of Bitcoins. At some future time (estimates vary, anywhere between 2030 and 2140) the mining reward model will probably be replaced with some form of transaction fee model.

Value of Bitcoin

For the first few years of its existence, the price of a Bitcoin was not highly variable. During this period it was used primarily for commerce (especially internationally and for dodgy items). However, in 2017 the price of Bitcoin exploded. This is entirely due to speculation by investors and is nothing to do with its intrinsic value or use as a trading instrument. The price of Bitcoin is widely, and correctly in my opinion, described as a bubble and a collective delusion.

If one seeks a rational explanation for the price of Bitcoin then one must study the history of speculation and financial bubbles. As far as I am concerned, Bitcoin is just the most recent example in a long line of bubbles, driven I suppose by human greed, hype and gullibility. It is obvious one should be skeptical about the value of Bitcoin, as it has no intrinsic value (unlike, say, gold or the NZ dollar, which is managed and underwritten by the NZ Reserve bank and is directly linked to the value of the NZ economy by mechanisms such as the requirement that taxes be paid in NZ dollars).

Electricity use

Recent newspaper headlines have highlighted the massive energy consumption costs of Bitcoin, such as using more electricity per year than Ireland or consuming all the world's electricity by 2020.

"Bitcoin mining consumes more electricity a year than Ireland" (Guardian, 27 November 2017) https://www.theguardian.com/technology/2017/nov/27/Bitcoin-mining-consumes-electricity-ireland

and

"Bitcoin Mining on Track to Consume All of the World's Energy by 2020" (Newsweek November 12, 2017) http://www.newsweek.com/Bitcoin-mining-track-consume-worlds-energy-2020-744036

These articles are based on a report by Alex de Vries, who runs the blog Digiconomist. His estimate of the total energy consumption of Bitcoin mining software reached 35 terawatt-hours per year, of the same order as the annual energy consumption of countries such as Ireland, Bulgaria and Denmark.

Headlines about consuming all the world's energy are sure to raise skeptical antennae, and one can find several responses to these claims, but first I want to clarify what the energy is being spent on.

I already mentioned the proof-of-work that is part of the mining algorithm. In Bitcoin this requires computing a cryptographic hash algorithm a very large number of times until a certain "lucky" output appears. All around the world are racks of computing devices (usually special-purpose ASIC hardware designed specifically for this task) that are spending all their time executing cryptographic hash programs. The workload gradually increases over time, as the Bitcoin system is designed to re-parameterize itself so that the elapsed time for mining stays roughly constant even when computing power increases (and the book-keeping reward per block also decreases over time). A persistent criticism of Bitcoin is that this work is very energy-intensive and is completely useless.

There are alternative cybercurrencies that have found a more useful mechanisms for proof-of-work: CureCoin is a cybercurrency where the proof-of-work performs computations that contribute to research on protein folding; Primecoin uses a proof-of-work algorithm that searches for prime number chains (I'm a number theorist, and even I don't find it that interesting, but still...); the Ethereum proof-of-work provides a distributed computing platform that is designed for managing smart contracts, and this is being used by a large number of organisations; Sia mining provides a distributed cloud storage platform.

The Bitcoin hype-merchants go berserk whenever it is claimed that Bitcoin is wasting electricity. A typical response is that the energy is not wasted but is creating value. Such sentiments are all over the internet. I find this unconvincing. The value of Bitcoin is not providing food, shelter, health or education. It is not reducing inequalities, enabling trade, or driving innovation. This response is not consistent with the fact that Bitcoin is a speculative bubble.

Bitcoin price between 2010 and 2016, peaking at around US $1000

(Data from : https://www.buyBitcoinworldwide.com/price/)

Bitcoin price during 2017, peaking at around US $18,000

(Data from : https://www.buyBitcoinworldwide.com/price/)

A slightly more subtle, but equally disingenuous argument is found in a recent Bloomberg article "No, Bitcoin Won't Boil the Oceans" (December 8 2017) https://www.bloomberg.com/view/articles/2017-12-07/Bitcoin-is-greener-than-its-critics-think

The article quotes a figure of 8.27 terawatt-hours per year for Bitcoin and then notes that this

"might sound like a lot, but it's actually less than an eighth of what U.S. data centers use."

The article continues: "Bitcoin's consumption won't necessarily keep rising as it has. Data centers, for example, have gotten a lot better. Not long ago, the Department of Energy was predicting that their electricity use would double every five years, and Google was getting slammed for consuming enough to power 200,0000 homes. In recent years, though, the centers' total electricity use has flattened even as their number has kept growing. As it turned out, better cooling and power management technology improved efficiency. Bitcoin miners are no less motivated by profit, so it stands to reason that they will seek to become more efficient and employ the cheapest energy available".

This argument might be convincing to a non-expert. But there is a world of difference between a business like Google, that has an interest in improving their systems to make them more efficient, and a system like Bitcoin that has a specifically engineered proof-of-work that rescales itself in response to improved computing systems. The entire purpose of this is to ensure that miners do work. And that means using energy.

"No, Bitcoin isn't likely to consume all the world's electricity in 2020" (CNBC, December 12, 2017) https://www.cnbc.com/2017/12/21/no-Bitcoin-is-likely-not-going-to-consume-all-the-worlds-energy-in-2020.html

That says "Digiconomist's estimate of Bitcoin's energy consumption is the basis for a lofty projection that reminds some experts of debunked forecasts that led businesses to over-invest in internet infrastructure."

"Bitcoin's sky-rocketing energy use is a viral story. We checked the math" (PRI, December 12, 2017) https://www.pri.org/stories/2017-12-20/Bitcoins-sky-rocketing-energy-use-viral-story-we-checked-math

This article gives several other estimates of the energy cost of Bitcoin. It concludes: "The most current and widely cited estimates put Bitcoin's yearly energy use at somewhere between 4 and 35 terawatt-hours. To put that in perspective, all of Google used about 5.7 terawatt-hours in 2015."

It may be surprising that estimates vary so widely from 4 to 35 terawatt-hours. The simple fact is that no-one knows exactly how much energy is being spent on Bitcoin mining. Mining is being done by anonymous groups all over the world and no-one can keep track of who is mining Bitcoins. However, one thing for sure is that Bitcoin is using a very substantial amount of energy and that all this energy is completely wasted in the sense that it provides no long-term value to humanity except to prop up a temporary speculative bubble. Worse, as I have explained, the Bitcoin system is itself designed to require increased energy demand over time. Hence the long-term prospects of the Bitcoin bubble are alarming from an environmental point of view.

Luckily, many of the more recent cybercurrencies, in particular Ethereum, have been designed to avoid these issues. There is also a different model for managing a Blockchain called proof-of-stake that does not require large amounts of work to be performed. So hopefully Bitcoin will crash sooner rather than later, and Ethereum and other cybercurrencies will become the market leaders.

Lost Bitcoins

Recall that a Bitcoin is not a physical object. To own a Bitcoin is to know the private key corresponding to your public key. If you lose your private key then there is no way to get your Bitcoin back. There is no central authority that can reimburse a hapless user.

In order to mitigate this, there are companies that provide a wallet service. Essentially they manage private keys on behalf of users, for a fee. Users need to have a password to access their Bitcoin account on the wallet service. So a user needs to remember their password.

Over the history of Bitcoin there are many stories of Bitcoins being stolen or lost. Theft of Bitcoins has usually been done by hacking a wallet company and stealing the private keys, and hence transferring those Bitcoins to new identities. Loss of Bitcoins usually arises from someone throwing away or losing a hard-disk or other computing equipment on which their private key was stored. There is no recourse for either of these events. No-one can help you if you lose your private key.

Or can they? Unsurprisingly, hypnotists and psychics have been quick to claim that they can provide solutions to people's Bitcoin crises.

For example, these articles:

"Would-Be Bitcoin Billionaires Resort to Hypnosis to Recover Their Passwords" (Dec 20, 2017) http://www.complex.com/life/2017/12/Bitcoin-hypnosis-recover-passwords

"Hypnotist Offers Bitcoin Password Recovery Sessions" https://cryptovest.com/news/hypnotist-offers-Bitcoin-password-recovery-sessions/

"Forgot the Key to Your Bitcoin Wallet? You Need a Crypto Hypnotist" https://news.Bitcoin.com/forgot-the-key-to-your-Bitcoin-wallet-you-need-a-crypto-hypnotist/

"Russian crypto-hypnotists help recover lost Bitcoin wallets" https://www.rt.com/business/412559-hypnosis-russia-Bitcoin-passwords/

One of these articles contains the astonishing sentence "The success of hypnotherapy in recovering forgotten passwords is dependent upon numerous factors, including the skill of the therapist and - most crucially - the extent to which the subject is capable of recalling the information". Who would have thought that the ability to recall the information would be relevant to help someone remember a forgotten password?

The Russian crypto-hypnotist "could not explain the difference between her crypto-practice and ordinary hypnosis but said almost half of her 20 clients managed to recall their e-wallet passwords."

Note that no-one is expected to be able to "remember" their Bitcoin private key (which is a random-looking 256-bit binary string or 64 hex characters). The hypnotists are helping people to remember forgotten passwords, which is not implausible, but there seems no reason for a hypnotist to brand themself as a "crypto-hypnotist" to do this.

It seems that psychics are more interested in predicting trends: "Past psychic predictions about Bitcoins have turned out to be true, a lot of mediums said that the value of these coins will sky rocket and it did!" (http://www.psychicreaders.mobi/psychic-predictions-Bitcoins-currency-modern-world/)

Well, that proves it then. Don't listen to me, listen to the psychics. To be honest, I'd agree. Psychics know exactly as much about the future value of Bitcoin as anyone else.