The Adventures of Bitcoin How a new currency’s advantages are still outweighed by a huge disadvantage.

Nowadays, creating a new currency is astonishingly easy.

Simply download the source code for an existing cryptocurrency—say bitcoin—change the name, and modify a few variables like the size of the block reward. Then announce it to the world. Voilà. An experienced computer programmer should be able to do it in an hour or two. Better yet, there are any number of automated websites that, after prompting for desired features like the currency name or block size, can generate the necessary code in a few seconds. For free.

Whether these cookie-cutter currencies are destined to reach the same level of success as traditional currencies like the dollar or the yen is a different matter. To succeed, a newly-introduced currency must be accepted by most residents of a nation. Broad acceptance has typically required careful planning, time, and energy on the part of the issuer. Take the euro, one of the world’s newest traditional currencies. Although Europeans began using euro banknotes in 2002, the idea for the euro is said to have been sparked as early as 1929 when Gustav Stresemann, in an attempt to heal the rift between France and Germany after the divisive Treaty of Versailles, asked the League of Nations: “Where are the European currency and the European stamp that we need?”

After several half-hearted attempts in the 1960s and ’70s, the political will to create the euro finally congealed in the late 1980s. The Maastricht Treaty of 1992, which set out the path for a single currency, barely passed muster with European voters, and was followed by a period of wrangling over the currency’s name, ecu or euro. Through the rest of the 1990s, members were obliged to bring their currencies within a narrow trading band relative to each other, until January 1, 1999 when the exchange rates were all frozen. The euro was born.

Banks now offered both local and euro-denominated deposits, but physical euros would not be introduced until January 1, 2002. To prepare Europeans for this cash rollout, shopkeepers began to affix euro prices to their wares along with legacy currency prices. A terrific quantity of notes and coins also had to be manufactured, a process that occupied European printers and mints for over four years.

How different from the 2014 debut of ‘Stalwart bucks,’ the brainchild of (then) Business Insider editor Joe Weisenthal. A copy of the popular cryptocurrency dogecoin (which features a dog as its mascot), Stalwart bucks took just a day or two to create. Weisenthal is by no means exceptional. New cryptocurrencies like Stalwart bucks are popping up every day or so. The website Map of Coins, for instance, tracks 667 existing and defunct cryptocoins.

Among them is Bitcoin, the poster child of the cryptocurrency movement. Whereas the euro is the handiwork of tens of thousands of bureaucrats spanning many decades, bitcoin was conceived in 2009 by a programmer or group of programmers named Satoshi Nakamoto—whereabouts and true identity unknown—with just 12,000 or so lines of C++ code. All other cryptocurrencies are variations on the theme. Not one of them is even close to surpassing the original in popularity or value which, at today’s price of $600 per coin is worth around $9.5 billion.

So far, bitcoin has not reached the point at which it is giving central bank money like the euro a serious run for its money. There is $9.5 billion worth of bitcoins in existence, a tiny fraction of the $1.2 trillion in U.S. dollar banknotes or € 1.1 trillion in Euro banknotes in circulation. Far from having achieved general acceptance, bitcoin usage is confined to a small band of tech geeks, hobbyists, and anarchists. Widely-used currencies do not simply come to life with a flip of the switch. Several key features determine whether a currency can achieve true lift-off.

Bitcoin as a hybrid

The best way to understand Bitcoin is by comparing it to two familiar types of money; government-issued cash (banknotes and coins) and bank deposits. As “digital bearer instruments,” bitcoins share traits with both money types. Let us first consider the property of being digital; then we will address what it means to be a bearer instrument.

Banknotes and coins are physical objects and, as such, they can only be passed hand-to-hand. Deposits, on the other hand, are immaterial book entries residing on a central database maintained by a bank manager. When deposits were first introduced hundreds of years ago, accounts were represented in a big paper ledger. The parties to a transaction had to walk to the bank and tell their banker which accounts in the ledger to debit and credit. The invention of the signed check made it possible to issue those instructions via a paper notice. Sometime in the latter 20th century this entire process was digitized. The banker’s ledger was transferred into digital format while electronic communications now allowed for the immediate transfer of instructions about how to update that ledger.

Like deposits, bitcoins are represented on computers by strings of 1s and 0s. Information on how to update bitcoin balances is broadcast digitally over the Internet. Unlike physical money like banknotes, bitcoins and deposits can be transferred very rapidly over large distances, making both of them a good medium for Internet payments.

Bitcoin is also like cash. Both are bearer instruments. To prove ownership of a banknote, the merchant need not verify that the note is registered in the buyer’s name with some third party, as is the case with a debit or credit card. From a merchant’s perspective, the buyer’s possession of a $20 bill, or their status as that note’s “bearer,” provides sufficient proof to establish that the $20 belongs to the buyer. Likewise with bitcoins. As long as an individual possesses the correct 256-bit private key in their “wallet” (e.g., on their phone, hard drive or even a piece of paper) then that constitutes sufficient proof of bitcoin ownership.

Because identification is not necessary to prove ownership, bearer instruments preserve the anonymity of buyers. This is a useful feature. When people want to purchase things that society frowns upon, anonymity is a way to escape censure. Alternatively, when buyers are worried that personal information will be skimmed by sellers and later used to “attack” the buyer, say by stealing their identity, paying with banknotes provides a good defense.

Another nice feature of a bearer instrument like cash is that once it passes from one hand to another, an irrevocable transfer of ownership has occurred. Compare this to a check payment: if a $20 check is used to buy a book, the buyer is free to walk away with the book but the seller still has to bring the check to a bank to cash or deposit it. If the check bounces due to insufficient funds, the book seller is short-changed. Because a banknote can never bounce, cash payments are riskless for all parties to a transaction.  Likewise for bitcoins; once the Bitcoin protocol has ensured that the coins are not counterfeits, then the transaction is final.

One last feature of a bearer instrument is “censorship resistance.” The institution that issues banknotes, a central bank like the Bank of England or Federal Reserve, cannot take any direct action to prevent people from spending or accepting banknotes. Contrast this to a bank deposit. A bank manager can always refuse to allow certain individuals to open an account, or, if the account is already open, they can prevent that account from receiving or disbursing funds.  In Bitcoin’s case, there is no way for the software code to prevent anyone from dealing in the stuff.

Bitcoin’s predecessors

So bitcoin can be thought of as a hybrid. It provides the digital capabilities of a deposit while replicating the anonymity, censorship resistance, and finality offered by cash. Bitcoin, however, is not the first hybrid. In the early 1980s, long before Bitcoin’s 2009 debut, privacy advocates were already devising digital bearer instruments. One of the brightest, a cryptographer named David Chaum, came up with a scheme called eCash. Chaum viewed it as his duty to provide individuals with a means of payment that protected their personal data from financial institutions and merchants.

As Internet usage grew, eCash looked sure to break into the mainstream. Several banks would sign up to trial the product in the 1990s, including Deutsche Bank. Netscape and Microsoft were rumored to be interested in commercializing eCash. In the end it failed to gain a foothold, perhaps because credit cards succeeded in becoming a sufficiently trustworthy form of payment over the Internet. Chaum’s company, Digicash, went bankrupt in 1999, ten years before Satoshi Nakamoto debuted bitcoin.

Because Nakamoto designed bitcoin with its precedents in mind, it is worth exploring how eCash worked. First, a bank customer would download some software which they could use to generate, say, a $1 digital coin with a unique serial number. The bank would in turn affix their sign to the coin and debit the customer’s bank account by $1. To buy something, customers simply transferred the coin to the merchant who accepted eCash. To protect the customer’s anonymity, Chaum introduced a technology called “blind signatures.” Blinding ensured that the bank that had signed the coins would not be able to access those coins’ serial numbers, and thus could not match them to a specific person.

One problem with digital cash, however, is that it is very simple for users to replicate. Simply copy a $1 digital coin and you have got $2 of them. This is called the double-spending problem. Banknotes are not susceptible to double-spending. Thanks to constantly evolving security features like watermarks and holograms, it is difficult to make cost-effective replicas. As for deposits, because bankers closely guard the ledger that tabulates account entries, it is difficult to hack in and double or triple-up one’s balances.

Chaum solved the double spending problem by having the issuing bank maintain a database of the serial numbers of all already-spent coins. When the customer spent a $1 coin signed by a certain bank, the merchant would call up that bank and provide the coin’s serial number. The bank would check its database to ensure that the coin had not been spent. If the number was there, the coin was not spendable, otherwise the transaction was free to proceed.

It is worth pointing out that Chaum’s eCash was not the only digital money to emerge in the 1990s. In 1996, Florida oncologist and self-taught coder Douglas Jackson opened E-Gold, a service that offered customers fully-backed gold deposits transferable over the Internet. These were not strictly bearer money since the customer had to open an account in order to be able to own e-gold. In practice, however, e-gold offered a degree of anonymity since it was easy for users to set up accounts using pseudonyms. In 2005 the FBI, concerned about the use of e-gold by “carders”—criminals who trafficked in stolen credit and debit card numbers—raided E-gold premises. E-gold would limp on until 2007, when Jackson was indicted on charges of money laundering, conspiracy, and operating an unlicensed money transmitting business.

Nakamoto’s Design

Satoshi Nakamoto was privy to these early failures and wanted to design a system that avoided their weaknesses while preserving the anonymity, finality, and censorship resistance of bearer money. In his famous November 2008 white paper, published online in the midst of the global credit crisis, Nakamoto says that the problem with previous attempts at digital money “is that the fate of the entire money system depends on the company running the mint, with every transaction having to go through them, just like a bank.” In a subsequent forum post, Nakamoto addresses what he calls the “old Chaumian central mint stuff,” noting that:

” … a lot of people automatically dismiss e-currency as a lost cause because of all the companies that failed since the 1990s. I hope it’s obvious that it was only the centrally controlled nature of those systems that doomed them.”

Nakamoto’s solution was to introduce a feature that previous attempts at digital money had never dared attempt: decentralization. The Bitcoin protocol is designed so that it has no central points of control. There is no third-party database to record serial numbers so as to ward off double-spending attacks. Instead, the task of validating and recording transactions is outsourced to a distributed network of anonymous computers running Bitcoin software, or “nodes.” Anyone with a computer can become a node. As for the money supply, there is neither a “Chaumian central mint” nor an e-gold server that issues and redeems tokens; rather, the evolution of bitcoin supply is set ahead of time by the code that makes up the Bitcoin protocol.

Earlier attempts at digital money did not attempt decentralization because the technology had not yet been proven. It took the broad success of peer-to-peer music file sharing networks like Shawn Parker and Shawn Fanning’s Napster to pave the way for Bitcoin’s distributed design. Rather than hosting music on a central server, Napster provided a directory that allowed users to connect to a network of peers who had songs available for download on their personal computers. Because there were millions of nodes it was difficult, though not impossible, for authorities to censor the network. While lawsuits eventually forced Napster to shut down, subsequent file sharing services like Bit Torrent sprung up in its place.

Like Napster, Nakamoto’s novel design provides a coordinating mechanism for a large population of distributed and anonymous nodes. Take the task of determining whether transactions are legitimate. With centralized money, it is the bank’s job to ensure that the money that is being spent is in the user’s account. The bitcoin network takes a different approach. Every bitcoin transaction that has ever been made is recorded. Because this history is stored in ‘blocks,’ with each block being a collection of transactions, it is referred to as the blockchain.

When two people exchange bitcoins, they announce the details of this transaction to the network of nodes. These nodes, which are located all over the world, independently cross reference the bitcoins used in the transaction against their version of the blockchain to ensure that they are legitimate. If the coins are counterfeit, or double-spent, they will not have any linkage to the historical record and the transaction will be ignored. The first node to reach clarity about the validity of a transaction broadcasts their answer to the entire network, thus allowing each node to update its version of the blockchain in preparation for verifying the next batch of incoming transactions.

Ensuring that the multitude of nodes is storing identical versions of the blockchain is a tricky affair. Because the system is distributed, not all nodes are directly connected to each other. Frictions like network lag impede the fast flow of communications across the network. Consider what might happen if one set of validated transactions is broadcast by a node at the exact same time that another node declares a second set of transactions. As this information gets transmitted through the network, a node in one part of the world—say Canada—might receive the first set while another node—say in China—receives the second set. If so, these nodes will effectively be running different versions of the blockchain. Nakamoto devised a simple rule to ensure that these conflicts are resolved: the longest chain is the right one. In this scenario, the next set of transactions to be broadcast will be appended to either the Chinese or Canadian version of the chain, making one longer than the other. All nodes using the short version immediately abandon it and adopt the longer chain, network consensus being reestablished.

How can bitcoin users be sure that history is not rewritten and balances maliciously re-allocated to a thief? In the case of deposits, customers trust the bank to be honest about the ledger because it is in the bank’s long-term interest to serve its customers. Government regulation and legal systems add a secondary layer of protection. Preventing malicious rewrites is a tricky problem with bitcoin because everyone who participates in the bitcoin network is allowed to retain their anonymity. At no cost, a thief can quickly deploy a team of dummy nodes that can, if their output is large enough, fool the remaining honest nodes into propagating a false longest chain that redistributes coins to the thief.

To protect against dummy nodes, the process of updating the blockchain has been designed to require the expense of energy. This feature sets a high bar for thieves. Attempts to fool the network into accepting a false record are far less likely if any alteration to history involves significant costs. Burning energy so as to update the blockchain is called mining. Prior to broadcasting a set of valid transactions, or block, to the network, a node begins to mine, or work on a difficult mathematical problem. Mining requires processer time, and this in turn requires electricity, which is costly. Only when the mathematical problem has been solved can the node broadcast the block of transactions to the network along with its “proof of work.” An incorrect answer will cause the network to ignore the block because it lacks proof of work.

When a node successfully adds a freshly mined set of transactions to the blockchain, it earns a reward of new bitcoins. The current bitcoin reward is 25 coins per transaction block, around US $13,000. The chance of winning this reward, which goes to the fastest node, attracts a large population of honest nodes who compete to mine the next block. To create a false longest chain, a thief must acquire enough computer power to outwork all other honest nodes. Fifty-one percent of all mining power is sufficient to do the trick, but this constitutes a huge block of computing power. Because this mining power can be put to better use by engaging in honest mining so as to harvest rewards, bitcoin users generally trust that the blockchain is a truthful description of the allocation of bitcoins.

This combination of a shared history and proof of work is a brilliant way to create a decentralized anonymous bearer instrument. It works amazingly well. Apart from a handful of minor glitches, the Bitcoin network’s resilience has been proven for more than seven years now. However, despite Bitcoin’s technical success it has not become a mainstream medium of exchange. Even among the narrow band of bitcoin users, the preference is to hoard them as a way to profit from sharp price jumps rather than for making payments. By contrast, ride-hailing firm Uber, created the same year as Bitcoin, has already achieved global success.

Unanchored

While many bitcoin advocates believe that mass adoption is just around the corner, a major drawback stands in its way. In sacrificing all central points of control, Nakamoto achieved his design goal of decentralization but simultaneously condemned the price of bitcoins to be permanently unhinged. Unlike eCash, e-Gold, or banknotes, all of which are stable because the issuing bank promises to manage the value of these instruments relative to some external price or index, Bitcoin has no central issuer. A central bank, for instance, can always reduce the supply of banknotes when consumer price inflation begins to exceed its target. It simply sells off the assets it holds in reserve, buys the unwanted cash, and destroys it. With Chaum’s eCash, the bank maintains a strict peg to the national currency unit by promising to redeem each dollar’s worth of eCash with a $1 or £1.

Bitcoin, on the other hand, has no backing. No counter has been set up where users can redeem bitcoins with cash, gold, or some other asset, nor are bitcoins subject to an inflation target; the purchasing power of bitcoin is left entirely to the whims of market demand. Should market demand suddenly rise, bitcoin can double in price. Should it collapse, bitcoin will be worth $0.

And volatile bitcoin has proven to be! On average, the day-to-day price changes experienced by bitcoin owners are about four times more volatile than the S&P 500, the U.S. benchmark measure of equities. It is hardly comforting that stocks, which involve a significant degree of risk, are so much less volatile than bitcoin, an instrument which purports to be money. In April 2013, the price of bitcoin sickeningly plunged from $260 to $50 in around 24 hours, an 81 percent decline. There was no obvious reason for the crash. More recently, bitcoin endured a 21 percent crash between June 20-22, 2016 and a 29 percent crash in early August 2016 after it was revealed that a Hong Kong-based Bitcoin exchange had halted all trading due to a security breach

Bitcoin’s lack of a price anchor mitigates its potential for mass adoption. Restaurants, grocery stores, and other retailers tend to keep their sticker and menu prices fixed for long periods of time. In a 2004 study, economists Bils and Klenow found that the average price spell of American consumer goods and services is 4.3 months, with coin operated laundry having the longest price spell (79.9 months) and gas the shortest (0.6 months). This provides owners of U.S. dollars with an incredible amount of clarity about the future. With $100 in his or her wallet or bank account, a dollar owner can calculate exactly how much goods and services he or she can afford to buy over the next few days or weeks.

The clarity that sticky prices provide to U.S. dollar owners is not extended to bitcoin owners. The dollar, not bitcoin, is the unit of account in the U.S.  Retailers set prices in dollars, or, put differently, communicate with customers using the $ symbol. Anyone who wants to pay in bitcoin must specify their payment preference at the checkout counter, either online or in a brick-and-mortar store. The clerk takes a good’s U.S. dollar sticker price and multiplies it by the current bitcoin-to-dollar exchange rate to get the final bitcoin price of the good.

To complicate things, the bitcoin exchange rate comes from an online bitcoin exchange where trades occur every millisecond. Because the exchange rate is always fluctuating, the bitcoin price of a store’s goods is updated every second. This means that someone who owns, say, 10 bitcoins, has no idea how much they will be able to afford the next second, let alone the next day.

In other words, Bitcoin faces an uphill trek because it can not match the conveniences that sticky prices afford to U.S. dollar owners. It is neither the unit of account, nor is it stable. This deficiency could be rectified by establishing a central point of control over bitcoin that allows a powerful actor like a central bank to peg bitcoin to the national currency, say at 1 bitcoin to $1 or ¥1. Gone would be the crazy price volatility, and Bitcoin users would suddenly gain the same clarity about the future as other money users.  However, this sort of measure would conflict with Nakamoto’s design goal of a completely decentralized system.

Many members of the bitcoin community firmly believe that as usage increases, price fluctuations will steadily flatten out of their own accord until bitcoins are about as stable as the yens and dollars of the world. This seems unlikely; but even if bitcoin were to succeed in stabilizing itself, it faces one last hurdle. Money serves the population as a universal standard, much like a language or system of weights and measures. Since all members of a society are familiar with the domestic unit of account, switching is incredibly difficult, in the same way that it would be nearly impossible for all Americans to suddenly adopt French as their common language. Without the support of a strong body like the government, a bitcoin unit of account seems unlikely.

It is unfortunate that all the useful features of bitcoin—decentralization, resiliency, anonymity, finality, censor resistance, and its ability to be used online—are twinned with price volatility. However, to write Bitcoin off this early in time would be premature.  The technology is still young and amorphous. Predicting how it will evolve is almost as difficult as pinning down Bitcoin’s ever fluctuating exchange rate.