Blockchain, a way of implementing a distributed ledger (distributed record-keeping), is a novel technology with little real-world practicality. The original Bitcoin white paper published back in October 31, 2008 spurred little interest in distributed ledgers. The distributed ledger was ignored for years until Bitcoin started receiving mainstream attention and a few years had passed.
I simply couldn’t find much evidence that distributed ledgers are useful for any real-world applications (other than speculative asset bubbles). Once you understand that blockchains are bad at solving real-world problems, then you will understand why Bitcoin will fail. The blockchain imposes limitations that makes Bitcoin a bad version of something that has been tried in the past: e-gold (description here and Wired profile here).
A company’s stance on blockchain can also serve as a test of a company’s management. In my view, companies pushing blockchain technology (e.g. IBM, Microsoft, Intel, Oracle) are disconnected from customers’ actual needs and have mediocre management. Companies that don’t talk about blockchain (e.g. Facebook, Amazon, Google, Apple) are more likely to produce sensible technology that will work in the real world.
Blockchain and its “benefits”
The only truly innovative part of Bitcoin is its implementation of a distributed ledger. Instead of having a single trusted party maintain a single ledger (that an oppressive government can coerce), anybody can join the network and receive their own copy of the ledger. A set of fixed rules governs how the copies of the ledger should be maintained. In theory, this does not allow for a single party to exert control over the network. The supposed benefits of decentralization are:
- The ledger can’t be changed without the Bitcoin owner’s permission. An oppressive government can’t seize a citizen’s assets since a transfer of funds would require the owner’s private key (which is like a password that grants control over someone’s Bitcoins).
- The rules governing Bitcoin can’t be changed. Central banks can’t devalue the currency by printing money. (In reality, Bitcoins devalue due to inflation built into the system.)
In practical use, the upsides will never materialize. For Bitcoin to see mainstream adoption for legitimate purposes, the surrounding ecosystem of Bitcoin exchanges will have to follow government regulations and interface with traditional banking. Otherwise, governments concerned about illegal activities (and therefore money laundering) will force banks to cut off ties with the Bitcoin ecosystem, preventing ordinary citizens from buying and selling Bitcoin on exchanges. Governments can attempt to seize Bitcoins by arresting individuals and threatening them with jail time. In that sense, the government can exert a lot of control and influence over Bitcoin. This is currently the case as banks are afraid of the fines they would have to pay if they were to breach anti-money laundering laws. Many of them limit access to the Bitcoin ecosystem.
And while the ledger and rules governing Bitcoin “can’t” be changed, users actually want changes to the ledger and changes to the rules of the game.
- Somebody exploited a bug to create 92 billion Bitcoins in a system designed to reach only 21 million Bitcoins (see here and here). Bitcoin’s extreme hyperinflation needed to be fixed… by swapping the ledger (with the ledger from an earlier point in time before the bug was exploited) so that the ‘money printing’ exploit could be reversed.
- If Bitcoin is to see wider adoption, it will require fundamental changes to increase its transaction capacity. Current capacity limits have caused transaction fees to skyrocket to a peak of around US$55 (chart) and for confirmation times to increase to several hours. However, the Bitcoin community has been divided as to the best technical path to take to solve the issues.
The users of a system might as well rely on a trusted party to make changes and to choose a direction in driving the software forward. That trusted party could be a democratically-elected party or a benevolent dictatorship. If there is dissatisfaction with the project’s leadership, open-source projects can be ‘forked’- those wishing to take things in a different direction can make their own version of the software as the software’s source code (the secret recipe for making the software) is open to everybody. A centralized open-source project (or a number of them) would do the same thing. The distributed ledger isn’t necessary. Unfortunately, distributed ledgers have technical downsides. In practical usage, distributed ledgers impose technical limitations onto a system with no benefit.
Bitcoin’s strength is in speculative asset bubbles
It is unlikely that Bitcoin will survive as a platform with legitimate uses. Bitcoin is a bad payment network tied at the hip to an intrinsically worthless currency.
As a payment network, Bitcoin is slow, expensive, has terrible security, and is not user-friendly. Something similar to Bitcoin has been tried before with e-gold (description here and Wired profile here). The major advantages of e-gold over the current Bitcoin system are:
- Fast confirmations. Payment confirmations did not range from minutes to 16+ hours.
- Lower fees in some cases- e-gold’s 1% storage fee is comparable to Bitcoin’s current inflation rate of 4.3% (chart) that pays for overhead. E-gold transactions were 50 cents or 1% of the transaction (whichever is lower) versus Bitcoin’s average ranging from $2 to $55 last year.
- Far superior security. Bitcoin’s security is abysmal.
- More user-friendly.
Bitcoin’s supposed advantage over e-gold (decentralization) will likely be irrelevant in practice. It offers no advantages over current payment systems.
The other major difference between Bitcoin and e-gold is that Bitcoin is backed by ones and zeroes (inherently worthless numbers generated by a computer) whereas e-gold was backed by gold and other precious metals. Currently, financial markets are ascribing significant value to Bitcoin’s 1s and 0s. Here are some misconceptions fueling the speculative bubble in Bitcoin:
- Decentralization will provide freedom from the government, money printing, central banking, and (risky) fractional reserve banking. E-gold was designed to do all of those things but was forced by the US government into changing its business practices. This ultimately forced e-gold to shut down.
- Bitcoin or some other cryptocurrency will be the dominant currency in the future. E-gold’s failure strongly suggests that cryptocurrencies will fail because it is unlikely that cryptocurrencies will be more effective than e-gold.
- Bitcoins are safe because they are protected by encryption and cryptography. In reality, Bitcoin’s security is abysmal. Gold and fiat currencies have never experienced incredibly high losses due to theft.
- Bitcoin will have lower fees than price-gouging banks. In reality, Bitcoin’s current inflation rate of 4.3% goes towards overhead while transaction fees have spiked as high as US$55 / transaction.
- Blockchain is a disruptive technology. In reality, nobody cared about distributed ledgers for years.
What allows the misconceptions to exist is blockchain. There is the mistaken belief that Bitcoin’s key innovation will actually allow a decentralized system to work. The reason why the bubble exists is because speculators don’t understand why blockchain is useless. Mostly, the situation exists because people don’t understand Bitcoin and simply assume that it works because the technobabble sounds so complicated.
Why Bitcoin is a bad payment network (feel free to skip this section)
Many many ideas have been tried in payments and in digital money, especially during the Dot-Com era when there were many well-funded startups (some of which had the backing of big banks). Paypal saw something like 30 or so of its competitors go out of business. There isn’t much that hasn’t been tried in payments and there is a graveyard full of failed ideas.
Bitcoin bears similarities to Paypal and the defunct e-gold (description here and Wired profile here). However, Bitcoin lacks any useful innovation over previous systems. The key innovation, the distributed ledger, solves a non-problem (centralization). Other than that, Bitcoin doesn’t attempt to solve any of the difficult problems in payments. It is also inherently flawed at dealing with some of the existing problems in payments.
- Account security: As with any financial account, scammers will try to trick consumers into giving away their passwords via phishing scams. A good payment processing system will look for footprints of such attacks, e.g. the IP address used to log into an account when it is emptied. Because IP addresses reveal the user’s location, a good payment processing system will flag logins where the user has traveled faster than commercial aviation (since that is highly improbable). This type of surveillance goes against the ethos of Bitcoin. But it’s worse than that. Historically, a significant fraction of the Bitcoins in circulation have been stolen. The second Mt Gox breach was estimated to have resulted in a loss of 650,000 Bitcoins, which was roughly 5% of all Bitcoins in existence at the time. Bad actors (some of whom are likely exchange employees) have been very good at stealing Bitcoins. Bitcoins are attractive targets because today’s complicated computer systems have a huge attack surface (this includes crooked employees and social engineering), because Bitcoins don’t have weight like cash or gold (a single person will take a long time to physically move millions of dollars), because Bitcoin is designed to be irreversible, and because the banking industry has yet to develop strong money laundering controls to stop thieves from converting stolen Bitcoins into cash.
- “Fast” settlement: The reason why many financial transactions take a long time to clear (days or months) is to allow for fraudulent transactions to be reversed. However, many financial institutions will allow low-risk transactions to clear immediately and will bear the risk of losing money if it turns out that the transaction was fraudulent. This makes the system far less inconvenient without losing the fraud-fighting properties of long settlement times.
- Fraud: Fraud is the reason why all of Paypal’s competitors failed in the eBay auction payments market. It’s a very difficult problem. In its early days, Paypal was ruthless in freezing accounts (especially those of legitimate customers- just Google “Paypal sucks”), saying no to risky business, and in providing no customer service.
- Convenience: Customers like payment systems that are easy to sign up for, clear quickly, and have good user experiences. Bitcoin’s design is not very user friendly. A consumer needs to get a bank account. Then they need to fund an account at a Bitcoin exchange, which will require a credit card or wire transfer… requiring them to participate in the traditional financial system. This is inconvenient if the customer tries to do this at a financial institution that closes accounts associated with Bitcoin (the anti-money laundering burden is not worth it for most banks). If the consumer is worried about their Bitcoins being stolen (roughly a third of all Bitcoin exchanges have been compromised), they can transfer the Bitcoins to their own Bitcoin wallet (many of which have been compromised). Then they should make a backup of their private key on paper because they will lose their Bitcoins if their hard drive crashes or if they forget the password protecting their private key. Compare all this to credit cards- consumers don’t worry much about losing their credit card, rarely have their money stolen, and don’t have to learn about computers.
- Money laundering: Most countries have laws that require financial companies to police their customers and to implement safeguards against money laundering. This includes knowing the identity of the customer. On the compliance front, the Bitcoin system does not allow for accounts to be frozen and would have to rely on (centralized) exchanges to comply with money laundering safeguards. The Bitcoin ecosystem would also need to develop safeguards against the current tumbling services (which masks a user’s public paper trail on the Bitcoin ledger) and should implement some way of blacklisting stolen Bitcoins currently in circulation. Such blacklisting would require centralization.
Bitcoin also fails in areas that aren’t difficult:
- Fast payment confirmation: Due to Bitcoin’s distributed ledger design, the fastest a transaction can be confirmed is around 10 minutes. Paypal and Visa can confirm transactions in seconds, so somebody purchasing a good can receive it in seconds rather than waiting for minutes.
- Overhead that does not cost billions of dollars a year: The Bitcoin system is supported by Bitcoin ‘mining’. The idea is that users would use their CPU to perform a lot of math calculations to support the integrity of the blockchain. Record keepers in the network will trust the side that has poured more computing power into its calculations; cryptography allows each node in the network to easily verify the amount of math performed. If the honest users do more math than bad actors trying to disrupt the system, then the Bitcoin network will function as intended. If the good guys don’t dedicate a lot of computing power to Bitcoin mining, then a bad actor can simply round up more computing power than the good guys and gain the trust of other nodes on the network. To try to prevent this, there’s an incentive system for the good guys to build up a large standing army of CPUs. Inflation is designed into Bitcoin to encourage the good guys to maintain a large army. This is done by giving out Bitcoins to Bitcoin miners depending on how much work they put in (the number of Bitcoins given out in a year is more or less pre-determined). Because the price of Bitcoin has skyrocketed, billions of dollars in Bitcoins will likely be given away this year to miners fighting over their share of Bitcoins. The ‘mining’ analogy is used because Bitcoin miners put in work and are rewarded with digital ‘gold’. Unfortunately, the current system results in an overhead of billions of dollars for a very low volume of transactions.
- Network capacity: Bitcoin is currently limited to around 3 transactions a second. As the Bitcoin network approaches this limit, fees skyrocket and payment confirmations can take several hours. This has caused some retailers like Steam to drop Bitcoin as a payment option. While this technical obstacle can be overcome, Bitcoin has dysfunctional governance that has prevented work on the issue. In the Bitcoin community, there was a civil war over whether the block size should be increased. (Doing so would solve impending capacity issues of the Bitcoin network in a simple, elegant fashion. The motivations for opposing it are unclear.) Opponents of a larger block size used denial of service attacks, propaganda, and censorship to fight this change. Ironically, a system that is supposed to be free from oppressive governments is being influenced by bad actors launching cyberattacks on their opponents.
Problems unique to Bitcoin:
- Security vulnerabilities: The distributed ledger has some outstanding security vulnerabilities that have yet to be resolved (e.g. malleability attacks, denial of service attacks on the peer-to-peer network, Sybil attacks, etc.).
- Centralization of Bitcoin miners: In practice, Bitcoin mining has been concentrated in undesirable ways… making it much easier for one party to disrupt the Bitcoin network with a 51% attack (see the earlier discussion about “overhead that does not cost billions of dollars a year”). Bitcoin mining will concentrate in countries with lots of cheap electricity. It will be concentrated among big companies that build large, efficient data centers. And it will continue to be dominated by ASICs, computer chips designed to do one thing and one thing only (mining Bitcoins in this case). Whichever company designs the best ASIC will see their ASIC dominate Bitcoin mining. Ideally, there should not be any backdoors programmed into that ASIC… such as the one programmed into Bitmain’s Antminer S9. If most of the mining power is concentrated in a single country (currently this is China), among a handful of companies, or in a single ASIC design (Bitmain, a Chinese company, currently has the best ASICs)… there is a risk of a 51% attack.
- Unknown economics: When the Bitcoin ‘money printing’ stops after all 21 million Bitcoins are mined, it is unclear if the economic incentives will still lead to a well-functioning system.
Bitcoin is a terrible store of value
- It is far, far easier to lose Bitcoins to theft than it is to lose gold.
- Various technical and economic obstacles need to be overcome for Bitcoin to be a usable payments network for mainstream use.
- Why would you trust a system where a software bug was exploited to create 92 billion Bitcoins?!
Blockchain as a litmus test
Companies that promote distributed ledgers are more focused on appearing to be innovative rather than solving their clients’ problems. To spot many of these companies, you can simply look at those donating to the Hyperledger project (which is developing distributed ledgers and giving away the source code).
- The publicly-traded Premier members are ACN, AXP, EADSF (Airbus), BIDU, CSCO, DDAIY (Daimler), FJTSY (Fujitsu), HTHIY (Hitachi), IBM, INTC, JPM, NIPNF (NEC Corp), and SAP.
- Some publicly-traded general members are: AMS:ABN (ABN AMRO), AET, BBVA, BNP Paribas, CME, ETR:DB1 (Deutsche Boerse), LLY, GM (GM Financial), TYO:9613 (NTT Data), ORCL, RHT, KRX:018260 (Samsung SDS), SBRCY (Sberbank), LON:SDR (Schroders), STT, EPA:HO (Thales), TRI, TSE:X (TMX Group), VMW, WFC, and WIT.
I raise my eyebrows at the technology and IT services providers on that list such as Microsoft, IBM, and SAP- these guys should know better because it’s their core business.
I would point out that it would make sense for companies to sell sensible solutions and to slap the word “blockchain” onto them for marketing reasons. However, I’m not aware of any publicly-traded companies that do this. Seeing the words “blockchain” and “distributed ledger” should be a reasonably reliable shortcut in figuring out whether or not a company is crazy. Some companies like Microsoft and IBM even offer buzzwords Blockchain as a Service while Intel and Oracle have teamed up for a Blockchain Cloud Service… perhaps working a little too hard in trying to appear cutting-edge.
Granted, it likely won’t end in tears. Some tech giants will continue to make money despite wasting resources on useless technologies. A few of the companies are blessed with scale advantages (Intel, Visa, Mastercard), network effects that lead to a near monopoly situation (Microsoft Windows), or a big competitive lead that will take a long time to catch up to (Microsoft, SAP, Oracle). These companies will simply make less money and their competitive advantages will erode over time. Others may not be so lucky. Cisco currently faces very strong competition so its political rot will likely cause Cisco to fall even more behind its competitors.
Banks and financial companies will not be hurt that much by their incompetence when it comes to technology. The reason why many financial companies are so bad at technology is because technology rarely determines the winners and losers in their field. As long as their core business is strong, their profits will greatly exceed the waste from bad technology. The real issue with banks is that they constantly find new ways of losing money as they get into trouble from non-core businesses, e.g. competing against Paypal in the early 2000s, the subprime housing bubble, subprime auto lending, etc. I wouldn’t focus too much on their skill in technology.
Conversely, I’m interested in technology, database, and IT services providers that aren’t talking about blockchain (e.g. the FANG stocks + Apple). These companies have more sensible corporate cultures. Google/Alphabet shareholders should be pleased to see that its only serious competitor in search (Microsoft) is wasting resources on blockchain.
AMD’s stock promotion
In this CNBC interview, AMD’s CEO Lisa Su claims that blockchain is a “positive foundational technology” and that cryptocurrencies only accounted for a small fraction of the company’s revenues. That’s crazy talk. Cryptocurrencies are the only blockchain implementation that has seen any mainstream adoption. When the crypto bubble bursts, AMD will be hurt as video card demand will drop dramatically (see a previous blog post for more detail). Yet AMD’s CEO would like to mislead investors regarding AMD’s sensitivity to cryptocurrencies, suggesting that AMD’s exposure is low. She is likely trying to mislead because the growth in cryptocurrencies has been masking poor performance in some of AMD’s business lines.
AMD’s technology is pretty good but the company has a problem with generating profits.
Closing thoughts
Unlike other manias, blockchain lacks a kernel of value creation. Asset bubbles in Dot-Com stocks and the Nifty Fifty were based on ideas that have merit as the Internet did change the world and quality businesses did perform better over very long time periods. Current excitement over cloud-based delivery of software (a technology that has been around for decades), fintech (making software for the financial industry), and venture capital (investing in startups) does have merit if you ignore valuations. Distributed ledgers on the other hand are useless.
I just don’t know when the bubble will pop.
*Disclosure: I am short a company (not mentioned in this article) that hypes its ‘blockchain’ solutions. Short AMD. Long FB and GOOGL.
Links
Pro-blockchain
Avoiding the pointless blockchain project
Blockchains are useless
Ten years in, nobody has come up with a use for blockchain
Blockchain is an awesome technology, but what, exactly, is it good for?
“When it came out I was of the mind that, ‘Hey, that’s kind of new and cool,’ said Rob Fleischman, a principal software architect at Akamai Technologies in Manchester. “But after talking with more and more tech people, I’m now almost convinced that blockchain is stupid and useless for almost everything.”
Don’t expect AWS to launch a blockchain service anytime soon
Letter to the editor of NIST reports – An deep dive on why \blockchains are useless’. Explains how blockchains aren’t actually immutable.
Bitcoin
The resolution of the Bitcoin experiment – An engineer who quit Google to work on Bitcoin-related apps explains his views on why Bitcoin is a failed experiment.
would you care about a payments network that:
* Couldn’t move your existing money
* Had wildly unpredictable fees that were high and rising fast
* Allowed buyers to take back payments they’d made after walking out of shops, by simply pressing a button (if you aren’t aware of this “feature” that’s because Bitcoin was only just changed to allow it)
* Is suffering large backlogs and flaky payments
* … which is controlled by China
* … and in which the companies and people building it were in open civil war?
Companies sinking money into blockchain
Hyperledger, an organization developing open-source blockchain software, lists its donor members on its website.
Blockchain consortium Hyperledger loses members – Some of Hyperledger’s members have lost some enthusiasm.
Enterprise Ethereum Alliance members – Ethereum is another distributed ledger that can be built on top of.
Bitcoin bubble
Data points on Bitcoin (on my Twitter account)