Every nerd must love the technology behind bitcoin. It is a surprising and creative solution to the problem of designing an electronic currency that is pseudonymous (participants do not need to reveal their real names) and is not controlled by a central authority (no single entity is able to revert transactions or freeze funds).
Satoshi's proof-of-work public blockchain protocol, published in 2008, solves this problem. It is extremely inefficient however, as it uses duplication to achieve decentralization. Many ledgers keep track of all bitcoin transactions in parallel, and these ledgers must be kept in sync. This is done by the so-called miners who verify past transactions and compete for the permission to add transactions to the growing blockchain ledgers. This permission is granted to whomever is first to solve a very work-intensive mathematical riddle, who is then also rewarded with newly minted bitcoins. This leads to a hardware arms race among the miners, and today a tremendous amount of electrical power and hardware resources are sacrificed on the altar of decentralization.
In the early days of bitcoin, everyone could be a miner, simply by downloading free code, running it on a personal computer and thereby becoming a part of the bitcoin peer-to-peer network. The resulting bitcoins were all but worthless however, since they couldn't be exchanged for anything.
This changed in 2011 with the establishment of darknet marketplaces that used bitcoin as payment systems. Darknet marketplaces are akin to e-commerce websites, but they specialize in illegal goods, such as drugs, and run on the Tor network, an inefficient layer on top of the internet designed to disguise the participants' IP-addresses. Satoshi's solution had found its problem: both pseudonymity and decentralization are strict requirements for payment systems on such illegal markets. Users eager to purchase drugs-by-mail demanded dollar-to-bitcoins exchange mechanisms, and darknet merchants created the opposing demand for bitcoin-to-dollar exchanges. To this day, cryptocurrency-powered darknet marketplaces are going strong.
A second problem solved by bitcoin was found in 2013: the efficient transfer of ransom money. Hackers would infiltrate the computers of businesses, hospitals, local governments and private individuals, lock up the data and demand bitcoin ransom payments to unlock. Again, pseudonymity and decentralization are strict requirements for the electronic transfer of ransom money. To this day, cryptocurrency-powered ransomware attacks are a major problem for computer systems around the world.[source]
Law enforcement has an obvious interest in piercing the pseudonymity behind bitcoin. This requires detective work, patience and some luck. The blockchain is public, and everyone can follow the flow of funds from bitcoin address to bitcoin address (these addresses serve as the "pseudonyms"). The problem, then, is to learn the true identity of the person controlling a given address. This can be done, for instance, if the person exchanges their bitcoins for dollars at a regulated cryptocurrency exchange which follows know-your-customer requirements, or if they use their bitcoins to have drugs delivered to a mailing address.
Criminals are aware of these efforts and use a variety of laundering schemes to obscure the origin of their bitcoin. Among these are so-called "mixing services" and "coinjoin privacy wallets"[source] which allow users to untraceably exchange their bitcoins for other peoples' bitcoins, the exchange of bitcoins for anonymizing cryptocurrencies such as Monero which have mixing built in, and the purchase of cash at darknet marketplaces that don't follow know-your-customer laws.[source] The fact that bitcoin remains the currency of choice for darknet markets and ransom payments shows that criminals believe these schemes to be sufficiently effective.
The defining characteristics of bitcoin and most other cryptocurrencies (pseudonymity, decentralization and the consequent inefficiency) prevent these currencies from finding legitimate uses.
Customers naturally shy away from an electronic payment scheme which does not offer the possibility of reverting transactions, even if those transactions were done under duress or as a result of fraud or mistakes. In the age of simple credit card chargebacks, such a payment system is simply unacceptable. Merchants reject bitcoin for different reasons: the inefficiency of the system causes transactions to be slow and expensive; furthermore, widespread bitcoin speculation causes its dollar value to be highly volatile, much more so than that of other currencies, making it all but impossible to quote prices in bitcoin.
These problems have not escaped supporters, who proceeded to invent a new use for bitcoin: "store of value". The idea is to exchange dollars for bitcoins, so that the value will still be there in times of crisis, war, hyperinflation and societal breakdown, when fiat currencies might collapse. The huge volatility of bitcoin stands in direct contradiction to its use as a store of value. There is, furthermore, no reason to believe that cryptocurrencies fare any better than fiat currencies in times of crisis. Will crypto exchanges still operate and still exchange bitcoin for something of value? How will the price of bitcoin fare under these circumstances? Will enough miners maintain the bitcoin peer-to-peer network so as to prevent double-spend and other 51% attacks? Will transaction fees remain affordable? Possibly yes, possibly no, nobody knows: but that is not the answer you want to hear if your goal is to purchase bullet-proof insurance against catastrophe. Historical data show that during various kinds of crises, gold has retained its value and has remained easily exchangeable. It can thus serve as a far more secure store of value than bitcoin.
That leaves one use of bitcoin, by far the largest today: speculation. People buy bitcoin in the hope that its dollar-price will rise. This rise has indeed been impressive, presumably because of the rising demand for bitcoin as a speculation vehicle (although there is solid evidence that the price explosions in 2013 and 2017 were due to price manipulations engineered by major crypto exchanges [source] [source]). Is it rational to expect this rise to continue?
In the short term, there is no predicting the appetites of the market. But is there value in bitcoin in the long run? One might argue that investment in bitcoin is an investment in the concept of a crypto future in general. By now bitcoin is an ancient protocol hampered by outdated technology; it is the largest cryptocurrency by market valuation, has the most miners and is therefore the slowest to change. (A large group of miners, who are invested in the status quo, would have to approve any major changes in the code base.) There are now many modern, lean and almost elegant cryptocurrencies which employ less inefficient schemes (such as proof-of-stake) and attempt to address bitcoin's shortcomings. So if there is a future in which a cryptocurrency is widely adopted for commerce, that currency wouldn't be bitcoin; it would be one of these newer competitors, either one that already exists or one yet to be invented. If that happens, speculators will sell their bitcoin in order to obtain this modern replacement, and the bitcoin price will collapse. In the other scenario, where the cryptocurrency concept continues to find no uses and withers away, bitcoin also dies. Speculators must also weigh the risk of governments choosing to outlaw cryptocurrencies to prevent money laundering, which would collapse the market.
There is thus no realistic scenario in which bitcoin becomes truly valuable in the long term.
Several newer blockchain protocols, Ethereum first among them, incorporate a concept of "smart contract". Smart contracts are neither. They are best described as "programmable wallets" that can send and receive messages and cryptocurrency tokens to and from other such wallets and individual users. All this behavior is automatic, under the control of a program stored on the blockchain. This program, which determines the behavior of the programmable wallet, cannot be changed once put on the blockchain and will live there forever. The computations performed by such a program are very costly since they are performed over and over again, identically by all miners, highlighting again the by now familiar inefficiency of decentralized blockchain solutions.
These programmable wallets should not be called "contracts" since they cannot model the real-world use case of contracts. In an ordinary contract, two or more parties promise to perform certain actions and sign with their real names. If a party fails to perform as promised, the other parties can sue, with various negative consequences on the delinquent party. In the blockchain world, with its defining property of pseudonymity, recourse to the courts is almost impossible: you don't know whom to sue.
Still, we're told that a wonderful world of "decentralized finance" can be built on top of this concept of smart contract. In fact, not even the most basic financial agreements can be implemented in this way: unsecured loans. In an unsecured loan contract, the creditor agrees to pay a sum of money to the debtor right now, while the debtor promises to pay a (typically larger) sum to the creditor at a later time. Examples are consumer credits, credit cards, CDs issued by banks and bonds issued by governments and businesses. A bank, for instance, can react to a delinquent consumer loan by ruining the debtor's credit rating and thereby preventing future loans or by obtaining a court order garnishing the debtor's wages or bank accounts. Smart contracts cannot enforce repayment, and none of these reactions are available in the decentralized pseudonymous blockchain finance world. Consequently, consumer credit in this world will be much harder to come by and will carry higher interest rates.
But the problems of the smart contract concept go deeper. These contracts cannot refer to real-world events without having to trust third parties. A simple insurance contract which would pay out if your house burns down illustrates the problem: how would the "smart contract" learn about the fire? There would have to be some trusted third party that informs the blockchain about fires. And this party better not be controlled by the insurance company. But how to verify that this pseudonymous third party is trustworthy? And what to do if it turns out not to be?
Smart contracts also suffer from the general problem of all decentralized solutions, already mentioned above: if a smart contract interaction was initiated by mistake, under duress or as a result of fraud, no authority is empowered to reverse the interaction. This is a defect customers would not accept.
The blockchain represents a publicly readable and permanent record of all financial transactions, creating a privacy nightmare. While criminals will work to hide their tracks and protect their pseudonyms, most consumers will not. Their identities can thus be discovered, revealing their financial blockchain transactions, forever.[source]
Another fundamental deficiency of smart contracts is their intransparency. These contracts are typically complicated programs written in any of a number of programming languages and compiled down to some bytecode that's stored on the blockchain, for anyone to inspect. Before interacting with such a contract, you should read, understand and verify this program, to make sure there are no bugs, unintended consequences or malicious backdoors. However, the program's source code is not available on the blockchain. If the author chose not to publish verified source code, you are reduced to studying a decompiled version of the program, which is extremely difficult. In any event, only experts (including hackers) will be able to detect bugs, unintended consequences or malicious backdoors in a smart contract. For ordinary users, the notion of informed consent is therefore all but non-existent; "reading the fine print" is all but impossible. You must trust.
To deal with these shortcomings, the decentralized finance world must invent complications on top of contraptions, on and on, increasing inefficiency and intransparency and creating new opportunities for fraud. Problems are created, not solved.
In some proposed decentralized finance projects, so-called "reputation tokens" are given out to participants who perform work useful to the project. These tokens can then be used, for instance, to influence the future of the project, to secure loans, or to gain lucrative investment opportunities.
Reputation tokens provide tangible benefits to their owners, which could be expressed in monetary terms. We may think of this value as the wages for the work performed. Since blockchain technologies are accessible worldwide on equal footing, these wages will be the same for workers all over the world. It is to be expected that, by standard market forces, the wages will turn out to be very low and most of the work will be performed by people in poor countries where opportunities for high wage work are rare. This is similar to the miserable compensation paid out by Amazon's Mechanical Turk or for video game farming.
Since reputation tokens have a monetary value and workers will have a need to cash out, there will arise a market for these tokens, overt or covert. Presumably, the project will try to suppress this market, but that is futile in a pseudonymous world. Even if workers were identified by their real-world identities and thus tied to the reputation tokens they earned, a covert market would spring up, where investors would pay workers to deploy their reputation tokens in the investor's interest.
Reputation tokens, therefore, do not represent actual reputation. They are a different form of money. It would be much simpler and more efficient to pay the workers directly in some cryptocurrency.
All money matters require trust. You have to trust that the money you receive today will still be valuable tomorrow. You have to trust that government interventions will suffice should your bank collapse. You have to trust that the SEC filings of a company you want to invest in are accurate. You have to trust that the judge deciding your contract dispute has not been bribed. Etc.
Proponents sometimes claim that decentralized finance schemes are "trustless". This is false – indeed, they require much more trust than their real-world counterparts, if only because they lack government regulation and there is hardly any recourse to the courts. And indeed, you need to trust people who have repeatedly proven to be untrustworthy.
Most speculators keep their cryptocurrency holdings at cryptocurrency exchanges. These are not regulated as tightly as banks, and there has been a long string of collapses of large exchanges, with depositors losing their investments. Often, crypto exchanges are hacked and cryptocurrencies are stolen; sometimes they simply go bankrupt; sometimes the owners run away with the deposits; sometimes all of the above. You must trust that your chosen exchange is different.
Storing cryptocurrency in a wallet on a personal computer is much more secure, but in that case you must trust that your hardware won't fail and that you won't forget your wallet password: in both cases your holdings are irretrievably lost.
If you buy and sell cryptocurrencies such as bitcoin or ether, you must trust that the quoted prices accurately reflect market forces. However, these cryptocurrencies are not regulated as securities by the SEC. Price manipulation schemes that would be illegal if applied to securities such as stocks or bonds are legal when applied to these cryptocurrencies, and are widespread.
As mentioned earlier, if you engage with any smart contract, you must trust that it does what it claims to do and that it does not contain any bugs. In most cases, you won't have recourse to the courts.
Stablecoins are widely used centralized cryptocurrencies whose value is pegged to a fiat currency such as the dollar. The issuing company typically claims that the stable coins are 100% backed, i.e. that the company holds one unit of the fiat currency in cash for each token of the stable coin. You must trust that this is true and that the stable coins' value will indeed remain stable. Contrary to experiences. Some stable coins are not backed by cash holdings but use some algorithmic price manipulation scheme, whose proper functioning you must trust. Contrary to experiences. [source]
If you deposit your cryptocurrency in some scheme that claims to pay interest, you must trust that your deposits will eventually be returned and that you are not dealing with a garden-variety Ponzi scheme. Contrary to experiences. Unlike in real-world Ponzi schemes, there won't be any identifiable person who can be held responsible.
Given that decentralization is one of the main motives behind the creation of blockchain protocols and also the main cause of their inefficiency, it might come as a surprise that centralization is rampant in the blockchain world and poses direct threats.
In 2020 it was reported that an estimated 60% of the bitcoin mining was carried out in China. This raised concerns: if the Chinese government controls these miners, it in effect control the bitcoin protocol. Miners ultimately decide which version of the mining software they run, and therefore determine the rules governing the bitcoin world. Furthermore, various overt and covert so-called 51% attacks become possible with this amount of mining power. The attacking miners would then receive a larger share of the mining profits than they deserve.
These concerns are in the past, since China outlawed mining and trading of cryptocurrencies in 2021. The reasons are obscure, but one can speculate. Miners use tremendous amounts of electrical power, and they pay for this power with fiat money they obtain from crypto exchanges in exchange for bitcoins they received as mining reward. The crypto exchanges in turn receive this fiat money from crypto investors around the world. In effect, therefore, a country that mines much crypto exports electricity to the world. Presumably, China prefers to use its electricity for domestic purposes.
But miner cartels can also arise organically. It is in each miner's genuine interest to join one or more such cartels: once a cartel controls more than half of the mining power in the network, it can increase its profits beyond its fair share by manipulating the protocol or run some other 51%-attack. All participating miners benefit. In economics, this is called "rent-seeking behavior" and is outlawed in many situations, though not in the context of crypto mining.
The formation of miner cartels is the Achilles' heel, a natural outcome of the incentive structure underlying all blockchain protocols.
Another tendency to centralization can be found in the operation of crypto exchanges. The large ones all run their own centralized stable coins, completely abandoning blockchain principles. (The reason is clear: issuing a stable token of nominal value $1 without backing it with cash holdings of $1 amounts to risk-free immediate profit.) Furthermore, painfully aware of the inefficiencies inherent in it, most of the operations of crypto exchanges avoid the blockchain altogether. If you transfer dollars to an exchange, buy bitcoins, hold them for a while, exchange them for ether, and then turn everything back into dollars, none of this will have been recorded on any blockchain. The exchange has an internal account book that keeps track of all their customers' holdings, just like any bank does: beautifully centralized and efficient. If you want your holdings to be sent to some blockchain address, they will do it, for a fee. The vast majority of bitcoin transactions never touch the blockchain.
Decentralized finance projects are typically created as distributed applications, or dapps. These are web applications or smartphone apps which allow end users to easily interact with a smart contract on a blockchain. Dapp and smart contract are written by the same company. This company will typically retain backdoor access to the smart contract, providing it with privileges others do not enjoy. The company therefore exerts complete centralized control over the project.
The vast majority of end users who interact with the blockchain will do so through dapps or using wallet software. In both scenarios, the user does not directly participate on the blockchain network as a full node because that would be too resource intensive. Instead, all interactions are routed through one of a handful of gatekeeping servers which run nodes on the blockchain and will perform blockchain actions on behalf of users. The two or three companies controlling these servers represent a single-point-of-failure and can in principle inspect, record, analyze and manipulate all traffic that passes through, since this traffic is not encrypted or signed.[source] These companies know about most blockchain transactions before anyone else, a huge and unfair advantage in finance. They further know the IP address of the user controlling a given wallet, exploitable information that is not available to others.
Almost every new decentralized finance project starts out by creating a new token which is to be used to pay for their services. These new tokens either live on their own blockchain or are created using smart contracts on an existing blockchain. But why? Couldn't I pay for the services with an existing cryptocurrency? Since all cryptocurrencies are essentially interchangeable (for a fee), why invent new ones constantly, one for each new service? Clearly, these complications create further inefficiencies.
The reason is obvious: these projects are not interested in using some existing decentralized cryptocurrency because they want to be in control of their own new token; they want to hold the first batch, sitting at the top of the pyramid. Whoever holds the first batch profits dramatically should the token rise in value. This is every crypto business plan in a nutshell.
In summary: even crypto businesses avoid existing decentralized blockchain technologies whenever they can - because those technologies don't solve any problems that people actually have.
We have seen that, at present, no legitimate problem has been solved by blockchain technology. But maybe this will change in the future? Should we continue the blockchain endeavor based on this hope?
I argue: no. Blockchain technology as it exists today is worse than useless. While it's possible that eventually some problem will be found that admits a blockchain solution superior to existing approaches, this would not be sufficient. Since it came into existence in 2009, blockchain technology has already done tremendous harm: to the environment, in the form of unfathomable amounts of wasted energy and resources,[source] and to the public, in the form of ubiquitous loss and fraud.[source] Every day blockchain technology continues to exist, more harm is done, the bleeding continues. A few useful applications in the future cannot make up for this accumulated harm. There is no credible evidence showing that the positive of potential future applications would ever balance out all this harm.
We should stop the use of blockchain technologies right now. Outlaw it as a form of money laundering and be done with it.