What is Bitcoin?

The bitcoin phenomenon has attracted quite a lot of attention lately. It has been discussed in many blogs and media websites. But what are they? Are they really money? Can they replace actual monetary institutions? Who issues and manages the bitcoins? Does the regression theorem hold, and if so how? What the bitcoins are, and how do they work, it is not very clear, I think, for two main reasons. It is a very new and innovative system, we are not used to think in bitcoins terms. Second, due to its technical aspects, the economic information comes mixed with technical details and is not easy to isolate the economic aspects of the bitcoin from the technical issues (which, I may add, are quite interesting as well). While it may be too early to foretell what may happen with this experiment, it is not too early to try to understand how they work. The bitcoin idea was inspired in a commodity money (i.e. gold), and even though there are important differences, this analogy can help to understand how bitcoins work and how it differs from other monetary institutions.

How do they work and where do they come from?

A bitcoin exchange is not a transfer of dollars, or euros, or any other currency through a “bitcoin server.” It is, in itself, a transfer of bitcoins. bitcoins are a unique computer code (we can think of a file if it’s easier, like mybitcoin.btc, that cannot be copied) that can be transferred from one person’s virtual wallet to another person’s virtual wallet (a peer-to-peer transfer). Let’s say A wants to buy a good from B that costs 100USD. If A buys this good from B with a bitcoin is as if instead of putting euros in his cashier machine he puts bitcoins. Then B can exchange the bitcoins to dollars just as he can exchange euros to dollars. Bitcoins, then, work as a different currency. This is important, a bitcoin transfer is not from and to a bank account, but from wallet to wallet. Simply put, there are no banks (for this reason there is no problem of fractional reserves versus 100-percent requirement). But where does A gets his bitcoins from? Easy, he buys them from a currency exchange office after downloading a free and open-source software. Just as A can buy euros by selling dollars, he can buy bitcoins by selling dollars. Where, then, does the currency exchange office gets the bitcoins from? Where do they come from in the first place. Bitcoins, like gold or oil, are buried. But, instead of being buried underground they are buried under complex algorithms that have to be solved to extract them. Gold and oil have to be mined. Bitcoins, also, have to be “mined.” Of course, bitcoins are not underground, so what does mining mean in this case? It means that a computer, especially servers, need to solve a complex algorithm, once the algorithm has been solved the miner gets a bunch of bitcoins in return that he can use to buy goods and services or exchange them for dollars or any other currency. Now, in the case of a commodity like gold or oil, sites easier to access are exploited first, leaving only the hardest one for later extraction. This decreasing marginal return is simulated by the system by progressively increasing the difficulty of the algorithms that have to be solved to get bitcoins in return. These algorithms, for example, may require an increasing precision when calculating the decimal numbers of an irrational number, like pi. The following graph from Wikipedia shows the bitcoin supply (for a 10 minute issuance frequency) from 2009 to 2017.

Furthermore, just as there is a fixed amount of gold or oil given by nature, there is fixed amount of bitcoins buried under increasing complex algorithms. To extract bitcoins is in itself an investment of server time, and the more bitcoins have already been extracted, the more expensive is to extract new ones. The cost of extracting bitcoins is not trivial, farms of servers may be needed with special refrigeration so they don’t overheat. The fact that a pool of miners can be put together hints that the cost is not trivial for individuals or small firms. Just as people do not wonder around looking for gold or oil, individuals may not find profitable to invest to mine bitcoins. Any currency exchange, for example, can have his own farm server mining out bitcoins to sell in the market or can buy them from a miner. There are, then two groups: (1) bitcoin users who buy the units so they can acquire goods and services, and (2) miners that can invest on a farm of serves to “mine” 24×7. But all this does not answer the question of why would anyone want to use bitcoins. What’s the benefit? Because there are no intermediaries (banks) transfers are from virtual wallet to virtual wallet, so there’s no cost of transaction (commission or fee). In our example A wanted to buy a good from B that costs 100USD. Let’s say A and B are very far away and a cash transaction is out of the question, A has to wire the money through his bank account to B’s bank account. The bank charges 10USD fee to A. Then, A’s cost is 110USD, 100USD for B’s good plus 10USD to the bank as commission. There is a transaction cost that can be saved if the bank where not in the middle. The following deal can be offered by A to a currency exchange office: A asks a currency exchange office to mine the equivalent to 100USD in bitcoins, sell the bitcoins to him and then buy them back from B at 100USD. This way B gets the 100USD for his good. But the currency exchange has to allocate server time and energy to this process. Let’s say that to mine the 100USD in bitcoins he needs to resign to 5USD services to his clients. Then the situation is that A won’t be willing to pay more than 110USD to the currency exchange/miner and the currency exchange/miner will not be able to accept less than 105USD. Both sides, A and the currency exchange/miner, can bargain a price between 105USD and 110USD. A similar benefit can be on B’s side if B’s bank also charges a commission, but for simplicity we don’t need to repeat the example. I don’t know how the bitcoins actually started, but to reduce transaction costs is certainly a benefit that helps to understand the rationale of using bitcoins.

A note on the supply side

Bitcoins have to be “mined,” but the total amount of bitcoins is fixed in the system. So, is supply fixed or variable? Supply is variable in a similar sense the supply of gold or oil is variable despite how much is available in the world is fixed by nature. In the case of bitcoins the total endowment that can be extracted is determined by the system. Only if the price at which the gold or oil can be sold justifies extracting new quantities supply will increase. Similarly, only if the price at which a bitcoin can be sold justifies the investment on “mining” the supply of bitcoins will increase. Bitcoins cannot be used until they have been mined out. It is easy to know how many bitcoins are out there to be mined, and where to find them is not a problem. For the case of commodities, on the other hand, we neither know exactly how much are left to extract and where to find them. Also, we can calculated how long it may take to mine a bitcoin, but we don’t know how long it will take us to find the next source of gold or oil. Because the bitcoins are buried under heavy algorithms and anyone who wants can become a miner, there is no central monetary authority, in other words there is no “Central Bank of Bitcoins.” The Economist suggests that the bitcoin is a primitive monetary system because there is no central bank to act as a lender of last resort. I agree that bitcoin is a “primitive” monetary system, after all there are no banks. But this isn’t a problem in itself, any new product is “primitive” at the beginning. The fact that there are no banks accepting bitcoins is what makes it “primitive” in a monetary sense, not the absence of a central bank. Central Banks, it may be well to recall, are not a market phenomenon. Who was the central issued of gold during the gold standard?

Who accepts bitcoins?

The bitcoin is still a new phenomenon. It is not easy to foretell if it will keep growing or eventually will become out of use. Nonetheless this link provides a list of sites that accept bitcoins as medium of payment. The link also shows the currency exchanges that buy and sell bitcoins. This link does not only show prices, but also volume of operation. For the last 30 days, for example, 1,500,000USD where exchanged against bitcoins. Almost all of the volume was carried through a single currency exchange: mtgox.

A note on the absence of banks

There are no banks in bitcoin world. This means there is no market for loanable funds of bitcoins. Two parties may use bitcoins to perform a particular exchange and then turn to a more common currency like dollars or euros. If, instead, saving in bitcoins takes place, this would be a case of hoarding, not saving in a bank account (a supply of credit in the loanable funds market). As long as users do not want to save in bitcoins, and bitcoin banks get into the picture the scope of their use may be limited. But there is a barrier to the presence of bank not present in the case of a commodity money. In the case of gold, silver, etcetera, there is an advantage in using bank notes. It is much more easier and comfortable to carry banknotes than metallic minerals. But there’s no clear advantage in making an electronic transfer of a “electronic banknote from Bank X” with respect to transfer an “electronic bitcoin.” If banks eventually get to work with bitcoins, to offer an easier transfer of funds will probably not be the reason. Some other benefit than ease of use will have to be offered.

What about the regression theorem?

Some sites and posts have shown some concerns on how would the regression theorem (RT) apply in the case of the bitcoins. Even though the TR may not be central to understand how bitcoin works, it still seems to have brought some attention in different blogs. After all, what’s the value of use of a set of bits (a very very very long sequence of digital zeros and ones ordered in a particular way)? The regression theorem, however, is not intended to answer how any given currency originates in the market, but how the phenomenon of money originates in the market. Let’s say we want to apply the RT to the fiat dollar, which has no value of use. We regress in time until the first time a fiat dollar is used as money. Where does a value of reference come from so that the parties can perform the first exchange? Well, in this case of how much a dollar backed in gold could buy the day before. The purchasing power of the gold in its last day of use is transferred, as in inertia, to the fiat dollar in its first day of use. A similar to the case if we track back the euro to its first day of use. So, it still holds that at the first moment the fiat dollar is used as money a “non fiat dollar reference value” comes from somewhere else. The value of reference, however, is still in monetary terms. But this means that we can keep regressing in time and, eventually, we will hit a commodity used for the first time as medium of exchange and the marginal value of use will be the reference. A similar kind of exercise can be thought for the bitcoins (without the compulsory change as the fiat money, though). Just as we jumped from fiat dollars to gold, we add a jump from bitcoins to fiat dollars to gold. If we understand the TR in a narrow sense, that any currency should have an exogenous value reference on its first use, this still holds as long as we accept other currency as an initial exogenous reference. The TR, however, was first presented in 1912, fiat currencies were not as common as today, and because Mises was dealing with Menger’s theory on the origin of money as a salable good is natural that the first exogenous valuation would already be the marginal value of use. He didn’t have to deal with a “discontinuity” of currencies. On the other hand, if we understand the TR more broadly, not of how a given currency comes to be in the market, but how money as a market phenomenon can take place, then the regression shouldn’t stop when bitcoins start because there are still other currencies in the market, we need to find the link at the beginning of the new currency and keep going back until the first monetary exchange takes place. This doesn’t mean that the RT question is not relevant for any given currency, it certainly points out to an important problem. See, for example, Selgin’s (1994) “On ensuring the acceptability of a new fiat money.” If the bitcoin project were to had aimed for a wide scale start, it should probably had to anchor its price against another currency. But if the start is in a more narrow group, then agreement between two parties and a cheap extraction of early bitcoin can play the trick if it allows to save enough transaction costs to bare the investment and price risk involved (bitcoin price volatility). Early bitcoins were easier to be mined by an individual or small company without heavy server work. Returning to our example between A and B. If the currency exchange alertness spots this opportunity, and he is the first step into bitcoin world, he can capture a good amount of bitcoins and offer deals to specific parties offering a lower transaction cost. I should say that I can’t tell if this was or not that case with early bitcoins, but it’s a conjecture of how a currency like a bitcoin may start to be used in a narrow group despite not having value of use and not being fixed/anchored to another currency.

Is the bitcoin money?

This question can be tricky if one is not careful. Is the Argentinean or Chilean Peso money? Well, probably not if we are thinking worldwide or for other geographical regions than Latin America. But certainly is in Argentina and Chile respectively. Is the bitcoin money? Well, not if we think worldwide. But yes for the group of people and firms that actually used them as a medium of exchange, even though in this case they may not be located in a defined geographical area. It is difficult to foretell what will happen with the bitcoin in the future. How far can make it, if it will collapse or just smoothly go out of use. But the bitcoin case is an interesting example to analyze where a currency used in exchanges do not require a central bank. Nicolas Cachanosky is a doctoral student in economics at Suffolk University, as well as a previous Sound Money Essay Contest winner. This is an expansion of the bitcoin post at Punto de Vista Económico.

Related Articles – Sound Money, Sound Money Project