Blockchain news

ico crypto schedule

The digital agenda of virtual currencies: Can BitCoin become a global currency, SpringerLink

The digital agenda of virtual currencies: Can BitCoin become a global currency?

  • Pavel Ciaian Email author
  • Miroslava Rajcaniova
  • d’Artis Kancs

Abstract

This paper identifies and analyzes BitCoin features which may facilitate BitCoin to become a global currency, as well as characteristics which may impede the use of BitCoin as a medium of exchange, a unit of account and a store of value, and compares BitCoin with standard currencies with respect to the main functions of money. Among all analyzed BitCoin features, the extreme price volatility stands out most clearly compared to standard currencies. In order to understand the reasons for such extreme price volatility, we attempt to identify drivers of BitCoin price formation and estimate their importance econometrically. We apply time-series analytical mechanisms to daily data for the 2009–2014 period. Our estimation results suggest that BitCoin appeal indicators are the strongest drivers of BitCoin price followed by market compels. In contrast, macro-financial developments do not determine BitCoin price in the long-run. Our findings suggest that as long as BitCoin price will be mainly driven by speculative investments, BitCoin will not be able to contest with standard currencies.

Keywords

1 Introduction

A broad range of virtual currencies have emerged during the last decade, such as BitCoin, LiteCoin, PeerCoin, AuroraCoin, DogeCoin and Ripple. The most successful among them is BitCoin, both in terms of its extraordinaire growth in the number of currency users and popularity by retailers. Since its introduction in 2009, BitCoin has been characterized also by a phenomenal increase in the number of transactions and market capitalization, which surpassed five billion US dollar in two thousand fifteen and since then has recorded further growth.

In the public media and in scientific community there is ongoing a upbeat debate on wheather BitCoin can actually function as a substitute for standard currencies such as US dollar, Euro or Yen. There are two contesting views in the literature on whether BitCoin fulfils the three criteria of a currency (a medium of exchange, a unit of account, and a store of value) (Mankiw two thousand seven ). One part of the literature argues that BitCoin does not behave as a real currency, but rather resembles speculative investments (e.g. Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ; Williams two thousand fourteen ). Other line of research, stresses the positive aspects of BitCoin and perceives it as a global virtual currency with strong future potential (e.g. Plassaras two thousand thirteen ; Satran two thousand thirteen ; Luther and White two thousand fourteen ; Folkinshteyn et al. Two thousand fifteen ). Also from a legal point of view, there is no global agreement on the status of BitCoin, as there are no international laws regulating BitCoin. Each country regards BitCoin differently and regulations are permanently evolving. For example, the Internal Revenue Service in the US has assigned BitCoin as barter on the grounds of the market-orientated treatment. Finland considers BitCoin as priced commodity. Germany has recognized it as private currency. Yet other EU Member States by and large are in the wait-and-see stage. On the other mitt, institutions such as European Central Bank (ECB) and European Banking Authority (EBA) define BitCoin as virtual currency.

In light of these and other open questions about BitCoin, in April two thousand fifteen the European Commission organized a Virtual Currencies Conference, one where among others the potential and challenges of BitCoin to become a global currency were discussed. The Virtual Currencies Conference acknowledged the need for further research to better understand the socio-economic and monetary aspects of the virtual currency ecosystem.

In order to shed light on these and related questions, the present paper attempts to identify and analyze BitCoin features which may facilitate BitCoin to become a global currency, as well as characteristics which may impede the use of BitCoin as a medium of exchange, a unit of account, or a store of value. Among the BitCoin features, which may facilitate its use as a currency, we have identified low transaction costs, high anonymity and privacy, learning spillover effects, infinite divisibility and no inflationary pressures. Among the BitCoin features, which may impede its use as a currency include the absence of a legal tender attribute, difficulty to procure BitCoins, relatively high stationary costs of adoption, dependence on network externalities, absence of an institution enforcing dispute resolution, absence of BitCoin denominated credits, deflationary pressure, enormously high price volatility, and issues with cyber security.

Our 2nd contribution to the literature is to undertake a comparative analysis of the identified BitCoin features with traditional currencies and their possible influence on BitCoin functions as a currency. Our analysis builds on previous studies, which have identified several advantages of BitCoin over traditional currencies. Very first, as noted by EPRS ( two thousand fourteen ), EBA ( two thousand fourteen ), and Folkinshteyn et al. ( two thousand fifteen ), BitCoin has no physical representation, such as paper bills or metal coins, which saves costs related to the production, transportation, and treating of physical currency. These costs can be substantial for standard currencies. Similarly, an advantage of BitCoin is that it is more convenient than standard payment mechanisms for small-value purchases, as it permits for money transfers at low costs and relatively prompt (Hayes et al. One thousand nine hundred ninety six ). As noted by Folkinshteyn et al. ( two thousand fifteen ), the advantages of low transaction fees and brief execution time make BitCoin an ideal medium of exchange. In contrast, Grinberg ( two thousand eleven ) opposes this view and argues that the response of the traditional e-commerce sector to the competitive pressure from BitCoin may induce them to reduce their transaction costs and thus offset the advantage of BitCoin, while providing also other benefits such as higher security. 2nd, an extra advantage of BitCoin, compared to standard currencies, is that it may reduce opportunities for theft, such as bank robbery. BitCoin could help curtail vandalism of vending machines, public phones, etc., because there would be no cash to steal. Similarly, businesses who treat cash, such as taxi drivers and petite shops, could be much less vulnerable to robbery, if they would use BitCoin. Eventually, BitCoin is a global currency, implying that there are no transaction costs related to currency exchange.

On the other forearm, BitCoin faces several challenges compared to standard currencies. One of the main challenges of BitCoin is its security (Moore and Christin two thousand thirteen ; Böhme et al. Two thousand fifteen ; Yermack two thousand fourteen ). Given that BitCoin transactions take place exclusively over the internet, cyber-security is its main threat. Particularly vulnerable to cyber-attacks are large holdings of BitCoins as well as BitCoin exchanges. The security problem is largely attributed to the lack of an oversight institution that would ensure security of BitCoin transactions and BitCoin system (ECB two thousand twelve ; Plassaras two thousand thirteen ; Moore and Christin two thousand thirteen ). 2nd, the use of a fresh currency, such as BitCoin, requires an initial investment for both businesses and consumers (Velde two thousand thirteen ). This includes costs linked to getting acquainted with BitCoin system in general and to the adoption of the payment technology in particular. For example, BitCoin requires the use of electronic devices including specific software. Similarly, BitCoin may also suffer from information asymmetry, as its system is relatively sophisticated and therefore may not be lightly understood by all potential users (ECB two thousand twelve ). Third, Yermack ( two thousand fourteen ) argues that BitCoin may fail to become a global medium of exchange, as it is used in too few exchanges of goods and services; it has a negligible market presence globally. Presently, there are around 0.11 million BitCoin transactions executed per day (BlockChain two thousand fifteen ) and most of these BitCoin transactions involve transfers inbetween speculative investors. BitCoin use for purchases of goods and services is only minor—20 % of all BitCoin transactions (Yermack two thousand fourteen ). For example, if compared to two hundred ninety five million conventional payments and terminal transactions per day done in Europe alone, the market share of BitCoin emerges to be minuscule (EBA two thousand fourteen ). Fourth, Yermack ( two thousand fourteen ) argues that BitCoin is ineffective as a implement of risk management of price volatility as it is uncorrelated with major world currencies (i.e. US dollar, Euro, Yen, British Pound) or gold price, which is a common use of currencies by businesses for hedging the risks associated with currency volatility. Macroeconomic switches that cause adjustment in exchange rates of currencies are not reflected in any way in BitCoin price movements implying that its price volatility cannot be lightly hedged. Eventually, BitCoin faces the problem of network externalities in its adoption (Gowrisankaran and Stavins one thousand nine hundred ninety nine ), as its benefit in exchange is positively correlated with the number of users (Plassaras two thousand thirteen ; ECB two thousand twelve ).

BitCoin and Euro price development in USD, 2009–2011. Source: BlockChain and Oanda

In order to better understand the reasons for such utterly high price volatility, we attempt to identify drivers of BitCoin price and estimate their importance econometrically. This is our third contribution to the literature. Previous studies (e.g. Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ) suggest three types of drivers determining BitCoin price development: (1) market compels of BitCoin supply and request (Two) BitCoin appeal, and (Three) global macroeconomic and financial developments. We apply time-series analytical mechanisms to these drivers using daily data for the 2009–2014 period. Our estimation results suggest that BitCoin appeal variables are most significant drivers of BitCoin price followed by market coerces. Our estimates do not support previous findings that macro-financial developments are driving BitCoin price in the long-run.

Generally, our findings suggest that as long as BitCoin price will be mainly driven by speculative investments, BitCoin will not be able to rival with standard currencies.

The rest of the paper is structured as goes after. Section two provides background information about BitCoin, which is a relatively fresh virtual currency. In the context of the three currency criteria, the key features are discussed and contrasted with standard currencies to analyze whether BitCoin fulfills the main functions of money. Section three identifies the main factors affecting BitCoin price based on previous studies; outlines the econometric approach—a Vector Auto Regressive (VAR) model—and specifies the estimated model. It also details the data sources used in the empirical analysis, the construction of the estimable model’s variables, and discusses the estimation results. The final section concludes.

Two Is BitCoin a currency?

Two.1 Background of BitCoin

Type of currencies

Money (currency) format

Certain types of local currencies

Banknotes and coins

Commercial bank money (deposits)

The most popular virtual currency is BitCoin, created by the Japanese programmer Satoshi Nakamoto in two thousand nine (Nakamoto two thousand nine ). It was the very first open source virtual currency, as BitCoin is managed by an open source software algorithm that uses the global internet network both to create BitCoins as well as to record and verify its transactions. Being a cryptocurrency, BitCoin uses the principles of cryptography to control the creation and exchange of BitCoins. BitCoins can be stored in local wallets (e.g. private computer, smartphone) using an open-source software or in an online wallet (Brito and Castillo two thousand thirteen ; Murphy two thousand thirteen ; CoinDesk 2015b ).

Compared to standard fiat currencies, such as US dollars or Euro, a distinguishing feature of BitCoin is that the quantity of units in circulation is not managed by a person, group, company, central authority, or government, but by a software algorithm. BitCoins are created in a ‘mining’ process, in which computer network participants, i.e. users who provide their computing power, verify and record payments into a public ledger called BlockChain. In comeback for this service they receive transaction fees and freshly minted BitCoins. A stationary amount of BitCoins is issued at a constant a priori defined and publicly known rate, according to which the stock of BitCoins increases at a decreasing rate. In two thousand one hundred forty the growth rate of BitCoin will converge to zero, when the maximum amount of BitCoins in circulation will reach twenty one million units; according to the current algorithm it will not switch after 2140.

BitCoin does not have physical representation. Instead, it is stored either on electronic devices (e.g. private computer, mobile, tablet) or entrusted to an online service and is transferred via the internet. BitCoins can be spent on both goods and services, if accepted by the retailer. Users interact with each other directly and anonymously, without third-party intervention. There is neither a central clearing house nor any other intermediary institution involved in the transactions such as central bank or government agency (Table one ). Presently, BitCoins can be acquired either (1) by exchanging them for standard money (e.g. US dollar, Euro) on a BitCoin exchange or from a BitCoin dealer, (Two) by obtaining them from sales of goods or services denominated in BitCoin, or (Three) through a mining process (Plassaras two thousand thirteen ; CoinDesk 2015c ).

BitCoins can be used to buy goods or services worldwide, provided that the transaction fucking partner accepts BitCoin as a mean of payment. A transaction implies that the present possessor of BitCoins transfers the ownership of a certain amount of BitCoins to a different market participant in exchange for other currencies, goods or services. A continuously growing number of companies accept BitCoins as payments for their goods and services, at the beginning of two thousand fifteen there were more than 100,000 venues accepting BitCoins (CoinDesk 2015a ; Cuthbertson two thousand fifteen ).

There are two rivaling views in the literature on whether BitCoin fulfils the three criteria of a currency (a medium of exchange, a unit of account, and a store of value). One part of the literature argues that BitCoin does not behave largely like a real currency, as it does not fulfil the main functions of a currency, but rather serves as a vehicle for speculative investments (Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ; Williams two thousand fourteen ). Other line of research stresses the potential of BitCoin and perceives it as a global currency with strong future potential (e.g. Plassaras two thousand thirteen ; Satran two thousand thirteen ; Luther and White two thousand fourteen ; Folkinshteyn et al. Two thousand fifteen ). In the following we attempt to identify and analyze the most significant advantages and disadvantages of BitCoin in the context of the three currency functions.

Two.Two Medium of exchange

Currency characteristics of BitCoin with respect to standard currencies

Advantage of BitCoin

Disadvantage of BitCoin

Medium of exchange

Anonymity and privacy

Not legal tender and difficulty to procure BitCoins

Learning spillover effect

Immobilized costs of adoption

Dispute resolution not available

Absence of BitCoin denominated credit

Unit of account

Relative price comparability problem

Source Authors’ presentation

Two.Two.1 Transaction costs

An appealing feature of BitCoin when used as a medium of exchange is its comparative advantage in transaction costs relative to standard currencies, as BitCoin transaction fees are considerably lower than comparable costs of traditional means of payment (e.g. payment cards or bank transfers). BitCoin transaction fees cover only the costs of maintaining the system (i.e. clearing system) paid to miners; there are no costs linked to third-party intermediaries, such as a regulatory authority, that would perform validation, storage or security functions; the BitCoin system (maintained by miners) is financed through transaction fees and freshly issued BitCoins. Three Transaction costs of standard currencies are considerably higher as they need to cover all costs of intermediaries. Besides the costs of clearing system, the fees charged for standard money transfers need to also offset the cost of storage, authentication, transport, security, etc. (Šurda two thousand twelve ; EBA two thousand fourteen ). Average transaction fees per transaction for BitCoin transfers are inbetween zero and one %, while traditional online payment systems charge fees inbetween two and five % or even more (EPRS two thousand fourteen ; EBA two thousand fourteen ; Folkinshteyn, Lennon and Reilly two thousand fifteen ). Estimates for the US dollar showcase that these costs (costs of processing and accounting of money, storage, transport, and security) account to as much as $60 billion annually (ca. Four.Four % of the US dollar value). Four Further, BitCoin offers swifter transaction execution than traditional online payment systems. The total processing time of BitCoin transfers is inbetween ten and sixty min (EBA two thousand fourteen ).

We may conclude that comparably low transaction and maintenance costs together with quick execution of transactions may facilitate the acceptance and use of BitCoin as a currency (Table two ).

Two.Two.Two Anonymity and transparency

An area, where BitCoin has a relatively high appeal, involves exchanges linked to illegal and criminal activities (e.g. money laundering; narcotic trade, tax evasion). While this was not the intent of BitCoin creators, it is rather the way individuals choose to use it, as certain BitCoin characteristics give its predisposition to be advantageous in such activities. For example, BitCoin payment transactions are anonymous and do not require the provision of individual identity information. Further, BitCoin permits for effortless international transfers (e.g. money laundering) without supervision, as BitCoin transactions are peer-to-peer and require just internet access. With its infrastructure being spread across the globe, it is difficult to intercept individual transactions (Bryans two thousand fourteen ; EBA two thousand fourteen ). These features grant BitCoin a rather high predisposition to be used in illegal activities. For example, the internet portal Silk Road, created in January 2011, provided online marketplace for the sale of illegal narcotics and weapons using BitCoins for payment, accounting for as much as half of all BitCoin transactions. Silk Road was shut down following an FBI investigation in 2013. Often it is argued that this event generated benefit to BitCoin by boosting its popularity (EBA two thousand fourteen ; Yermack two thousand fourteen ).

Gambling is other area where BitCoin emerges to be growing as a medium of exchange, as it protects costumer privacy and permits receiving funds from consumers not being able to use other payment methods. The most popular BitCoin denominated online gambling site is Satoshi Dice operating since two thousand twelve (Böhme et al. Two thousand fifteen ). Generally, there is a growing number of online gambling sites, the BitCoin Wiki ( two thousand fifteen ) lists around one hundred BitCoin based casinos, poker sites, bingo games, betting services and lotteries.

The BitCoin platform is semi-transparent and public, meaning that anyone is able to go after the chain of transaction. All BitCoin payments have a traceable history that can be viewed by anyone. A single anonymity breach can uncover a user’s entire BitCoin transaction history. For example, the skill of the identity of any user from any transaction (e.g. obtained from the mailing address used for delivery of purchased goods with BitCoin, or from the bank account used to purchase BitCoin) permit to track that user’s transactions backward and forward through the BlockChain history. Hence, BitCoin system is often referred as pseudonymous in the sense that total history of all transactions and every BitCoin is preserved on the publicly ledger. However, several options are available to overcome this anonymity problem and make transactions non-traceable, for example, by using fresh addresses for each payment received, using switch addresses when sending payments, or using BitCoin mixer services to break the link inbetween an user and its BitCoins (Bitcoinhelp two thousand fourteen ; Crawford two thousand fourteen ; BitCoin two thousand fifteen ; Böhme et al. Two thousand fifteen ).

We may conclude that relatively high anonymity of BitCoin users and transparency of BitCoin transactions may facilitate the popularity and use of BitCoin as a currency (Table two ).

Two.Two.Three Legal tender

One distinguishing feature of BitCoin is that it is not a legal tender (as opposed to standard currencies which would imply its mandatory acceptance in exchanges). Private or public businesses are not legally obliged to accept BitCoin as a payment form for goods and services they trade. The use of BitCoin as a medium of exchange is thus fully dependent on its voluntary adoption by market participants (EBA two thousand fourteen ).

According to Yermack ( two thousand fourteen ), an extra obstacle for BitCoin to become a widely used medium of exchange, arises from the difficulty of procuring fresh BitCoins. BitCoins can be obtained only from online exchanges or dealers (except for successful BitCoin miners).

We may conclude that, generally, the absence of legal tender may impede BitCoin use as a currency, because any business can determine individually on the acceptance/not acceptance of BitCoin (Table two ).

Two.Two.Four Immobile costs

The use of a fresh currency, such as BitCoin, requires an initial investment for both businesses and consumers. This includes costs linked to getting acquainted with the BitCoin system in general and to the adoption of the payment technology in particular. For example, BitCoin requires the use of electronic devices including specific software. Five

In addition, a relatively high level of computer skill is required for understanding and using BitCoins, which may represent a barrier to many potential users and may constraint its broader adoption (i.e. information asymmetry problem). BitCoin is based on a complicated code, which is understood by only a few persons and is managed by even fewer individuals (programmers), who oversee the entire system without accountability, arbitration or recourse (Velde two thousand thirteen ; Yermack two thousand fourteen ).

On the other mitt, BitCoin use may increase users’ interaction with virtual currencies and thus may have learning spillover effects by helping to improve their abilities and skill in virtual transactions and online financial applications (Hayes et al. One thousand nine hundred ninety six ; Berentsten one thousand nine hundred ninety eight ; Plassaras two thousand thirteen ).

We may conclude that in presence of high stationary costs, the popularity and use of BitCoin as a currency may be gravely impeded at least in the short-run, as presently the technical skill about BitCoin is still rather low in society. However, BitCoin may generate some extra benefits to its users through learning spillover effects (Table two ).

Two.Two.Five Network externalities

The incentives for market participants to use BitCoin depend on the number of existing users. That is, the benefit of making the investment and using it in exchanges depends on the number of other suppliers and consumers of goods and services using BitCoin. If only a few businesses accept BitCoins, there are little incentives for consumers to acquire them. On the other palm, if only a few consumers use BitCoins, businesses have little incentives to invest into the equipment for processing BitCoin payments for their goods and services. This is a well-known problem of network externalities (Gowrisankaran and Stavins one thousand nine hundred ninety nine ). Hence, one of the main BitCoin challenges in becoming a global currency is to coax users to use it in their purchases and businesses to accept it as payment form for their goods and services (Berentsten one thousand nine hundred ninety eight ; ECB two thousand twelve ; Plassaras two thousand thirteen ).

Yermack ( two thousand fourteen ) argues that BitCoin largely fails to be a global medium of exchange as it is used to a limited extent to intermediate the exchange of goods and services. Presently there are around 0.11 million BitCoin transactions executed per day (BlockChain two thousand fifteen ) which is insignificant compared, for example, to two hundred ninety five million conventional payments and terminal transactions per day done in Europe alone (EBA two thousand fourteen ). According to BlockChain ( two thousand fifteen ), the use of BitCoin as a medium of exchange is continuously enhancing over time as more and more businesses tend to accept it as a form of payment. Hence, despite the rather low BitCoin invasion in the global exchanges presently may not be an obstacle for BitCoin to become a global medium of exchange in future. Very first, given that BitCoin is a relatively fresh currency, it may take some time till it gets adopted in a broader context. 2nd, there exist examples of puny countries with own currencies (e.g. Fiji) which intermediate the exchange of goods and services within the national boundaries but have only a minor share globally.

We may conclude that the issue of network externalities may impede growth of BitCoin use as a currency in the short-run. However, these issues may become less relevant in future, as BitCoin use continuously increases (Table two ).

Two.Two.6 Dispute resolution

The use of BitCoin in market exchanges bears a certain risk, because of the absence of any protection against disputes inbetween parties involved in the exchange. Once a BitCoin transaction is realized, it is irreversible and cannot be disputed. There is no centralized mechanism available to revert an erroneous transaction or to treat the disputes with the aim to provide protection against human errors of fraud that may occur in exchanges (e.g. protection against disputes over non-fulfillment of contract). Presently, the correction of an erroneous transaction is possible only through a voluntary agreement of the parties involved in the exchange (EBA two thousand fourteen ; Böhme et al. Two thousand fifteen ).

We may conclude that the absence of an institution regulating and enforcing BitCoin related disputes, the popularity and use of BitCoin as a currency may be impeded, particularly for risk-averse market participants (Table two ).

Two.Two.7 Credit market

Another obstacle for BitCoin to become a widely used medium of exchange may be that BitCoin cannot be used to take loans because, under the current system, every loan would need to be made in BitCoin. Whereas the basis of standard banking system is fractional-reserve, which determine how much fresh money can be created through loans, BitCoins are unique and cannot be duplicated, they only exist as an electronic analog a kind of physical coin. This boundaries the expansion of BitCoin (Hanley two thousand fourteen ). For example, the absence of BitCoin denominated credit cards and consumer loans do not permit purchases on credit—a method widely used in most developed retail markets, which thresholds the expansion of BitCoin as a medium of exchange (Yermack two thousand fourteen ).

We may conclude that in the absence of BitCoin denominated credit, BitCoin growth is gravely constrained, as the share of credit transactions is rather high (and growing) in modern societies (Table two ).

Two.Trio Unit of account

To serve the function of a unit of account, as any other currency, BitCoin should be able to measure the relative value of goods and services and other transactions (e.g. debts) in the economy. We identify two key characteristics of BitCoin as a unit of account, which differ substantially from traditional currencies: divisibility and price volatility (see Table two for summary).

Two.Three.1 Divisibility

An significant distinguishing feature of BitCoin is its almost infinite divisibility, implying that prices may be quoted in four or more decimal places. Indeed, divisibility is a necessary characteristic of a currency to accommodate the valuation for all types and sizes of transactions.

On the other forearm, the price differences in the magnitude of several decimal places (e.g. four or more) may be confusing to consumers and may pose problems to them to comprehend and compare relative prices of goods and services. For comparison most world currencies use no more than two decimal points for price quotations (Yermack two thousand fourteen ).

We may conclude that, generally, due to almost infinite divisibility possibilities of BitCoin its use and popularity may grow compared to standard currencies, albeit petite price denominations may reduce capability of consumers to accurately distinguish relative prices (Table two ).

Two.Three.Two Price volatility

BitCoin prices demonstrate utterly high short-run volatility, which diminishes its capability in indicating an effective unit of account (Yermack two thousand fourteen ). The frequent BitCoin price switches cause direct and indirect costs to businesses and consumers. Businesses that use BitCoin have to adjust prices frequently, otherwise they may realize decrease in comes back (because of underpriced goods and services) or loss of competitiveness (because of overpriced goods and services). This is particularly problematic for businesses trading outputs in BitCoins, while paying for production factors and intermediate inputs in local standard currency (e.g. US dollar, Euro, Yen, British Pound), causing discrepancy in relative prices inbetween outputs and inputs in presence of high BitCoin price volatility. Frequent price switches in turn become confusing to consumers, as it becomes more difficult to spot the true relative prices of goods and services. Six

Albeit, high price volatility of BitCoin may inflict risk to its holders, entrepreneurial innovations provide alternative solutions which diminish the price volatility risk. For example, market exchange pricing may facilitate price setting to businesses (retailers) and widen spending options to consumers. The market exchange pricing enable retailers to set prices in one currency (e.g. US dollar, Euro, Yen, British Pound) while displaying them to consumers at the same time and automatically updated in more currencies, including in BitCoin, reflecting the current market exchange rates. This system makes relatively costless price tracking in BitCoin to businesses that reflect up-to-date market exchange rate particularly for online sales. Another example includes instantaneous exchange facilities which enable retailers to accept BitCoin as payment without actually receiving BitCoins. This system involves a third party which intermediates the exchange of BitCoins, paid by consumers, to standard currency, which is received by the retailer (e.g. US dollar, Euro, Yen, British Pound). Because sellers never actually receive BitCoins, they avoid the BitCoins exchange risk; the exchange risk is beard by the intermediaries which receive a fee in comeback (Luther and White two thousand fourteen ).

In the context of BitCoin use as a unit of account, the utterly high price volatility reduces BitCoin power to convey accurately the relative prices of goods and services in the economy (Table two ). If, as argued by Luther and White ( two thousand fourteen ), financial developments are providing alleviating options for addressing the issue of BitCoin price volatility, then the adverse impacts of price volatility may decline in future.

Two.Four Store of value

The value of the money must remain stable over time to permit their use in exchanges in different points of time. We identify two features of BitCoin as store of value, which differ substantially from traditional currencies: non-inflationary supply and cyber security (see Table two for summary).

Two.Four.1 Non-inflationary supply

Standard currencies are usually inflationary, meaning that their value reduces over time and thus diminishes the capability of the currency to function as a store of value. In contrast, an significant advantage of BitCoin is the protection against inflation as a safe haven from government interference. Given that under the current mechanism the future number of BitCoins is capped to a maximum amount of twenty one million units with no expansion possible beyond this amount, BitCoin will be exposed to deflationary pressures if its use as an investment alternative or for use as a medium of exchange increases. Seven On the one palm, this is beneficial to BitCoin holders, as it will make them richer over time. On the other forearm, the expectation of higher future value of BitCoins may lead to its hoarding in anticipation of higher price which in turn may reduce its use in exchanges of goods and services. Meiklejohn et al. ( two thousand thirteen ) find that two thousand eleven was a break point for spending-versus hording of BitCoin. After this date the majority of BitCoins (more than sixty %) that were received were spent within less than one month period. The hording prior to two thousand eleven took many BitCoins out of circulation meaning that most of BitCoins from this period are in the “sink” addresses that have never been spent (Ron and Shamir two thousand thirteen ).

We may conclude that, in the absence of inflationary pressures, BitCoin popularity should increase compared to standard currencies, albeit deflationary pressure may act as a countervailing force (Table two ).

Two.Four.Two Cyber security

One of the main threats to BitCoin capability to preserve its value to the holders is the issue of cyber security. In the past, many BitCoin owners have lost their virtual money through online theft. Standard currencies give a possibility to protect it against theft either by physically hiding it (e.g. under mattresses, in a safe) or by depositing it in a bank. BitCoin is a virtual currency and thus it cannot be physically hidden. BitCoins must be held in computer accounts known as virtual wallets. The security of BitCoin wallets has often been a major problem. Even BitCoin exchanges were subject to cyber-attacks and thefts as they are common target of hackers. The most prominent example is the collapse of the largest BitCoin exchange, MtGox, in February two thousand fourteen allegedly leading to a loss of eight hundred fifty thousand BitCoins (Böhme et al. Two thousand fifteen ; Yermack two thousand fourteen ). Moore and Christin ( two thousand thirteen ) find that forty five % of the total studied BitCoin exchanges closed down, and forty six % of the exchanges that closed did not reimburse any claims to consumers.

In the context of BitCoin use as a store of value, BitCoin sensitivity to cyber-attacks and thefts reduces trust in BitCoin as a currency and hence impedes its growth and chances of becoming a global currency (Table two ).

Trio BitCoin price and its volatility

Among all BitCoin features identified and discussed above, eventually, price volatility is the one with the largest differences compared to the major world currencies, such as US dollar, Euro, Yen, British Pound. As shown in Fig. One , since its introduction in 2009, BitCoin price has shown utterly high volatility; it has enhanced from zero value at the time of its inception in two thousand nine to around $1100 at the end of two thousand thirteen when it reached the peak followed by a decline to around $225 (see Fig. One ). For comparison, the fluctuation of exchange rate inbetween US dollar and Euro has not exceeded a ±20 % strapped in the same six year period (see Fig. One ). Such market volatility with extreme price movements (±8000 %) makes BitCoin of little use as a unit of account.

One consequence of such extreme price volatility is that BitCoin may not succeed as a global currency to be accepted by suppliers and consumers, as it may fail to accurately convey relative prices of goods and services in the economy and may generate uncertainties to its holders due to its inability to preserve stable value over time. Given that BitCoin is a relatively fresh currency, its price formation is not well understood yet. In order to better understand reasons for such utterly high price volatility, in this section we attempt to identify the drivers of BitCoin price and estimate their relative importance econometrically.

Three.1 Drivers of BitCoin price

The existing studies in the literature (e.g. Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ), suggest three types of drivers determining BitCoin price formation: (1) market coerces of BitCoin supply and request, (Two) BitCoin attraction, and (Three) global macroeconomic and financial developments.

Trio.1.1 Driver 1: market coerces of BitCoin supply and request

According to Buchholz et al. ( two thousand twelve ), one of the key drivers of BitCoin price is the interaction inbetween BitCoin supply and request on the BitCoin market. According to the quantity theory of money, the BitCoin supply is determined by the total stock of BitCoin in circulation. The BitCoin request is represented by the size of BitCoin economy (i.e. its use in exchanges) and the velocity of BitCoin circulation. The BitCoin velocity measures the frequency at which one unit of BitCoin is used for purchase of goods and services. The quantity theory implies that the price of BitCoin decreases with the velocity and the stock of BitCoins, but increases with the size of BitCoin economy and the general price level.

The request for BitCoin is primarily driven by its value as a medium of exchange. BitCoin does not have intrinsic value like commodity currency such as gold standard. For example, the key difference inbetween the gold standard and BitCoin is that the request for BitCoin is driven solely by its value in future exchange, whereas the request for commodity currency is driven by both its intrinsic value and its value in future exchange. The BitCoin supply is given by the total stock of units put in circulation, which is publicly known and is motionless in the long-run. Whereas BitCoin supply is exogenous, the supply of gold is endogenous, as it responds to switches in production technology (e.g. mining technology for gold) and comebacks. Given the exogeneity of BitCoin supply, the primary driver of its price developments is expected to be the request side shocks. By altering expectations of future use in exchanges, such shocks to request have the potential to produce large swings in the BitCoin price (Luther and White two thousand fourteen ).

Three.1.Two Driver Two: BitCoin attraction

There are several BitCoin-specific factors which, in addition to traditional currency price determinants, such as market supply and request, determine its request. This is partially linked to the fact that BitCoin has been created relatively recently and to the nature of the currency. (Barber et al. Two thousand twelve ; Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ).

Very first, BitCoin price may be affected by the risk and uncertainty of the entire BitCoin system. Given that BitCoin is a fiat currency and thus intrinsically worthless, it does not have an underlying value derived from consumption or its use in production process (such as gold). The value of a fiat currency is based on trust that it will be valuable and accepted as a medium of exchange also in future (Greco two thousand one ; Kovenock and Vries two thousand two ). Eight The expectations about trust and acceptance are particularly relevant for BitCoin which, being a relatively fresh currency, is in the phase of establishing its market share by building credibility among market participants.

2nd, being a virtual currency, BitCoin is more vulnerable to cyber-attacks than traditional currencies, which can destabilize the entire BitCoin system and eventually lead to a collapse of BitCoin. Such attacks have been frequently occurring in the BitCoin system in the past (Barber et al. Two thousand twelve ; Moore and Christin two thousand thirteen ). As mentioned above, Moore and Christin ( two thousand thirteen ) examined forty BitCoin exchanges and found that eighteen have been closed down after cyber-attacks, while the world’s largest BitCoin exchange, collapsed in two thousand fourteen due to a cyber-attack.

Third, BitCoin price may be affected by its appeal as an investment chance for potential investors. According to Gervais et al. ( two thousand one ), Grullon et al. ( two thousand four ) and Barber and Odean ( two thousand eight ), potential investors’ decisions may be affected by an increase or decrease of attention in the news media. The role of information is particularly significant in the presence of many alternative investment choices and positive search costs. Given that investment request depends on the costs associated with searching for information for potential investment opportunities available on the market, such as, the stock exchange, those investment opportunities which are under a particular attention of news media may be preferred by potential investors, because they reduce search costs. In turn, enlargened investment request for BitCoin may exercise upward pressure on BitCoin price. Indeed, Lee ( two thousand fourteen ) finds such evidence for BitCoin, whereby the alteration of positive and negative news generated high price cycles. This implies that the attention-driven investment behavior can affect BitCoin price either positively or negatively, depending on the type of news that predominate in the media at a given point of time.

Trio.1.Three Driver Three: global macroeconomic and financial developments

Van Wijk ( two thousand thirteen ) stresses the role of global macroeconomic and financial development, captured by variables such as stock exchange indices, exchange rates, and oil prices measures in determining BitCoin price. The influence of macroeconomic and financial indicators on BitCoin price may work through several channels. For example, stock exchange indices may reflect general macroeconomic and financial developments of the global economy. Favorable macroeconomic and financial developments may stimulate the use of BitCoin in trade and exchanges and thus strengthen its request, which may have positive influence on BitCoin price.

Inflation and price indices are other type of indicators capturing significant macroeconomic and financial developments. According to Krugman and Obstfeld ( two thousand three ) and Palombizio and Morris ( two thousand twelve ), oil price is one of the main sources of request and cost pressures, and it provides an early indication of inflation development. Thus, when the price of oil signals potential switches in the general price level, this may lead to depreciation (or appreciation) of BitCoin price. Also the exchange rate may reflect inflation development and thus influence positively BitCoin price as indicated above.

According to Dimitrova ( two thousand five ), there could be also negative relation inbetween a currency’s price and macro financial indicators. A decline in the stock prices induces foreign investors to sell the financial assets they hold. This leads to a depreciation of the underlying currency, but may stimulate BitCoin price, if investors substitute investment in stocks for investment in BitCoin. Generally, investors’ comeback on stock exchange may capture chance costs of investing in BitCoin. Hence, in this case the stock exchange indices are expected to be positively related to BitCoin price.

Trio.Two Econometric treatment

The very first four variables pt, yt, vt, and bt account for request and supply drivers of BitCoin price (driver 1). Following the quantity theory of money we expect that coefficients β1 and βTwo would be positive, whereas βTrio and βFour would be negative. In addition, given that BitCoin supply is largely predefined, the total stock of BitCoins in circulation, bt, is a semi- exogenous variable, and implying that the influence of coefficient βFour on BitCoin price should be puny and/or statistically not significant. Variable at captures the BitCoin attraction (driver Two). As discussed above, the coefficient βFive associated to this variable can be either negative or positive, for example, as both positive and negative news attract investors’ and users’ attention. Ultimately, variable mt represents global macroeconomic and financial developments (driver Three). According to the previous findings discussed above, we expect the sign of the coefficient β6 to be either positive or negative depending on the type of macroeconomic variable.

The econometric model (1) contains mutually interdependent variables—BitCoin price and its explanatory variables. The estimation of non-linear interdependencies among interdependent time series in presence of mutually correlated variables is subject to an issue of endogeneity (Lütkepohl and Krätzig two thousand four ). To circumvent the issue of endogeneity, we go after the general treatment in the literature to analyze the causality inbetween endogenous time-series and specify a multivariate Vector Auto Regressive (VAR) model (Lütkepohl and Krätzig two thousand four ).

According to Engle and Granger ( one thousand nine hundred eighty seven ), regressions of interdependent and non-stationary time series may lead to spurious results. In order to avoid spurious regression, it is significant to test the properties of the time series involved. Therefore, in the very first step, the stationarity of time series is determined, for which we use two unit root tests: the augmented Dickey–Fuller (ADF) test and the Phillips–Perron (PP) test. The number of lags that we use for each dependent variable is determined by the Akaike Information Criterion (AIC). If two individual time series are non-stationary, their combination may be stationary (Engle and Granger one thousand nine hundred eighty seven ). In this special case, the time series are considered to be cointegrated, implying that there exists a long-run equilibrium relationship inbetween them.

In the 2nd step, we employ the Johansen’s cointegration method to examine the long-term relationship inbetween the price series. The number of cointegrating vectors is determined by the maximum eigenvalue test and the trace test. Both tests use eigenvalues to compute the associated test statistics. We go after the Pantula principle (Pantula one thousand nine hundred eighty nine ) to determine whether a time trend and a constant term should be included in the model.

In the third step, we estimate a vector error correction model for those series that are cointegrated (Johansen and Juselius’s one thousand nine hundred ninety ). It includes an error correction term indicating the speed of adjustment of any disequilibrium towards a long-term equilibrium state. In other words, the error correction term captures the switches in the BitCoin price required to eliminate the past deviation of the prices from the equilibrium levels.

As usual, in order to ensure the adequacy of the estimated models, we implement a series of specification tests: Lagrange-multiplier (LM) test for autocorrelation in the residuals; Jarque–Bera test to check if the residuals in the Vector-Error Correction (VEC) are normally distributed and a test of stability of the model.

Three.Trio Data and variable construction

In order to construct the dependent variable, we use data for BitCoin price, p t B , denominated in US dollar (BitCoin price). We use the historical number of total BitCoins (number of BitCoins) which have been mined to account for the total stock of BitCoins in circulation, bt. We use two alternative proxies for the size of BitCoin economy, yt: the total number of unique BitCoin transactions per day (number of transactions), and the number of unique BitCoin addresses used per day (number of addresses). Following Matonis ( two thousand twelve ), we proxy the monetary velocity of BitCoin circulation, vt, by BitCoin days demolished for any given transaction (days ruined). This variable is calculated by taking the number of BitCoins in transaction and multiplying it by the number of days since those coins were last spent. All these data are extracted from quandl.com. To measure the price level of global economy, pt, we use exchange rate inbetween the US dollar and the Euro (exchange rate) extracted from the European Central Bank. We use exchange rate inbetween the US dollar and the Euro, because in our data BitCoin price is denominated in US dollar. For example, if the US dollar would appreciate against the Euro, most likely it will also appreciate against the BitCoin. Consequently, an increase in the exchange rate inbetween the Euro and the US dollar would lead to a decrease in the amount of US dollar that have to be paid for one BitCoin, which decreases its price.

In order to capture BitCoin attraction, at, we go after Kristoufek ( two thousand thirteen ) and use the volume of daily BitCoin views on Wikipedia (views on Wikipedia). According to Kristoufek ( two thousand thirteen ), the frequency of searches related to the virtual currency is a good measure of potential investors’ interest in the currency. However, the online search queries, such as Wikipedia views, may measure both investors’ and users’ interest in BitCoin, as it captures information’s request about the currency but it does not differentiate on whether the information is used to guide investment decisions or online BitCoin denominated exchange of goods and services. In addition, we also construct a variable capturing the number of fresh members (fresh members) and fresh posts on online BitCoin forums (fresh posts) extracted from bitcointalk.org. As explained above, the variable fresh members captures the size of the BitCoin economy but also attention-driven investment behavior of fresh BitCoin members. The variable fresh posts captures the effect of trust and/or uncertainty, as it represents the power of discussions among members.

To account for global macroeconomic and financial developments, mt, we go after van Wijk ( two thousand thirteen ) and use oil price (oil price) and the Dow Jones stock market index (Dow Jones). Nine Oil prices are extracted from the US Energy Information Administration, and Dow Jones index is extracted from the Federal Research Bank of St. Louis.

Three.Four Estimation results

Specification of the empirically estimated models

Number of BitCoins

Number of transactions

Number of addresses

Views on Wikipedia

Short-run effects on BitCoin price for drivers 1, two and Trio

L1D BitCoin price

L2D BitCoin price

L3D BitCoin price

L4D BitCoin price

L1D number of BitCoins

L2D number of BitCoins

L3D number of BitCoins

L4D number of BitCoins

L1D number of transactions

L2D number of transactions

L3D number of transactions

L4D number of transactions

L1D number of addresses

L2D number of addresses

L3D number of addresses

L4D number of addresses

L1D days ruined

L2D days demolished

L3D days demolished

L4D days demolished

L1D exchange rate

L2D exchange rate

L3D exchange rate

L4D exchange rate

L1D views on Wikipedia

L2D views on Wikipedia

L3D views on Wikipedia

L4D views on Wikipedia

L1D fresh members

L2D fresh members

L3D fresh members

L4D fresh members

*** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero. L1D indicates one lag, L2D indicates two lags, L3D indicates three lags, and L4D indicates four lags

Short-run effects on BitCoin price for general models

L1D BitCoin price

L2D BitCoin price

L3D BitCoin price

L4D BitCoin price

L1D number of BitCoins

L2D number of BitCoins

L3D number of BitCoins

L4D number of BitCoins

L1D number of transactions

L2D number of transactions

L3D number of transactions

L4D number of transactions

L1D number of addresses

L2D number of addresses

L3D number of addresses

L4D number of addresses

L1D days demolished

L2D days ruined

L3D days ruined

L4D days demolished

L1D exchange rate

L2D exchange rate

L3D exchange rate

L4D exchange rate

L1D views on Wikipedia

L2D views on Wikipedia

L3D views on Wikipedia

L4D views on Wikipedia

L1D fresh members

L2D fresh members

L3D fresh members

L4D fresh members

*** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero. L1D indicates one lag, L2D indicates two lags, L3D indicates three lags, and L4D indicates four lags

Long-run effects on BitCoin price for drivers 1, two and Three

Number of BitCoins

Number of transactions

Number of addresses

views on Wikipedia

Dependent variable: BitCoin price. *** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero

Long-run effects on BitCoin price for general models

Number of BitCoins

Number of transactions

Number of addresses

Views on Wikipedia

Dependent variable: BitCoin price. *** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero

Tables six and seven showcase the long-run impacts of different determinants on BitCoin price. According to the results reported in Tables five and six , the long-run relationship inbetween BitCoin price and the explanatory variables considered in the estimated models is stronger than the short-run influence. In the following we discuss the long-run results with respect to the three drivers of BitCoin price formation.

Three.Four.1 Results for driver 1: market coerces of BitCoin supply and request

The very first major observation arising from the estimates reported in Tables six and seven is that the market compels of supply and request have an influence on BitCoin price. Generally, the request side variables (e.g. days ruined, number of addresses) emerge to exert a more pronounced influence on BitCoin price than the supply side drivers (e.g. number of BitCoins). According to the results reported in Table six , an increase in the stock of BitCoins (number of BitCoins) leads to a decrease in BitCoin price (model 1.Two), whereas an increase in the size of the BitCoin economy (number of addresses) and its velocity (days ruined) lead to an increase in BitCoin price (models 1.1, 1.Two, 1.Three, 1.Four, 1.Five, Four.Three, Four.Four, Four.7). These results are in line with our expectations. Contrary to our expectations, the alternative variable that captures the size of the BitCoin economy (number of transactions) has negative influence on BitCoin price in models 1.1 and 1.Five. However, this variable is not significant in the more general models (models Four.1–4.9), implying that its influence on BitCoin price is negligible.

Albeit, the sign of the estimated coefficients for market coerces of BitCoin supply and request is in line with our hypothesis (except for the variable number of transactions in models 1.1 and 1.Five), the statistical significance and magnitude of the estimated coefficients decreases in most models, when accounting for the influence of BitCoin appeal and global macroeconomic and financial developments (models Four.1–4.9 in Table seven ). The supply–demand variables are statistically significant in models Four.Trio, Four.Four and Four.7, but have a considerably lower magnitude of the estimated influence than in models 1.1–1.Five, which capture only market compels of supply and request. This could be explained by the fact that part of the BitCoin price variation explained by the supply–demand variables is absorbed by other variables in more general specifications (models Four.1–4.9).

Three.Four.Two Results for driver Two: BitCoin attraction

The strongest and statistically the most significant influence on BitCoin price is estimated for variables capturing the influence of BitCoin attraction: views on Wikipedia, fresh members and fresh posts (models Two.1 and models Four.1–4.9). Variable fresh members has negative influence on BitCoin price, implying that attention-driven investment behavior of fresh investors predominates. Variable fresh posts has positive influence on BitCoin price, reflecting an enhancing acceptance and trust in BitCoin captured by the energy of discussion inbetween BitCoin users. This may reflect declining transaction costs and uncertainty for investors, which increases investment request for BitCoin and hence its price.

Consistent with the findings of Kristoufek ( two thousand thirteen ), Wikipedia views have a statistically significant influence on BitCoin price. This variable is significant and has positive influence in all models (except for model Four.8). These results are in line with our expectations and with the remarkable Wikipedia article traffic statistics, according to which on average BitCoin article on Wikipedia is being viewed 200,000 times per month, and is ranked among the top one hundred most viewed articles. Only eighty eight English-language Wikipedia articles have more traffic than the BitCoin article, out of a total of approximately five million English-language articles. However, the interpretation of Wikipedia views is not straightforward, as it may capture various effects. On the one mitt, Wikipedia views may represent attention dedicated by potential users’ and/or investors’ to BitCoin likely induced by their reaction to media reporting on BitCoin. The attention effect may influence either positively or negatively the BitCoin price depending on the type of news. The positive estimated coefficient associated with Wikipedia views variable implies that the influence of positive news predominates. One of the most common negative news reported in media about BitCoin is related the security breach (cyber-attacks) against the currency. The positive estimated results indicate that the implication of security problem emerges to be offset by the positive news effect.

On the other forearm, Wikipedia views may measure investors’ or users’ interest in BitCoin, as it captures information’s request about the currency. It may reflect switches in the skill about BitCoin inbetween potential investors and/or users, thus leading to a higher acceptance and request for it either as an investment chance or as a medium of exchange. Significant is that the type of individuals searching information about BitCoin on Wikipedia likely are fresh BitCoin users/investors, because Wikipedia contains rather general information about BitCoin, which is known by incumbent investors or advanced BitCoin users. Both investors and users may search for information about BitCoin and thus the Wikipedia effect may capture either the users’ request for information with the aim to use it as a medium of exchange or may reflect investors’ information search who perceive BitCoin as an investment chance (i.e. for hoarding). Kristoufek ( two thousand thirteen ) argues that the second—the hording—effect predominates because, according to him, the market compels of BitCoin supply and request permitting for setting a ‘‘fair’’ price are missing, rather its price is driven by the investors’ faith in the future growth and is predominated by short-term investors, trend chasers, noise traders and speculators. This argument is also in line with the sceptics on BitCoin (e.g. Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ) who argue that hoarding is one of the key weaknesses of BitCoin as a currency (alongside the security problem) compared to standard currencies.

Trio.Four.Trio Results for driver Three: global macroeconomic and financial developments

Our findings suggest that, in contrast to previous studies (i.e. van Wijk two thousand thirteen ), global macro-financial developments such as the Dow Jones Index, exchange rate and oil price do not significantly affect BitCoin price in the long-run. Only in Model Three.1 the macro and financial variables (Dow Jones, oil price and exchange rate) are statistically significant (Table six ). Our results suggest that an increase in oil price leads to a decrease in the budget of consumers (and companies), implying that less money will be spent on other goods, including BitCoin. Consequently, this would lead to a decrease in request for BitCoin, decreasing its price. These results are in line with the estimates of van Wijk ( two thousand thirteen ), who also finds statistically significant influence of macro-financial variables on BitCoin price. However, van Wijk ( two thousand thirteen ) does not account for market compels of supply and request or BitCoin attraction indicators. When these factors are taken into consideration (models Four.1–4.9), their influence decreases considerably in all estimated models (except for model Four.1) (Table seven ). These finding support the argument of Yermack ( two thousand fourteen ) that BitCoin is relatively ineffective as a instrument for risk management against adverse market developments as its price is not responsive to macroeconomic variables meaning that it cannot be lightly hedged against other assets that are driven by macroeconomic developments.

Trio.Five Comparison with previous studies

The verification of our results through comparison with previous studies is a challenging task, as (1) the currency is relatively fresh and there are not many studies in the literature, which analyze BitCoin price formation; and (Two) there are significant differences inbetween the used empirical proxies for variable construction and model specification across different studies. Despite these difficulties, in this section we attempt to compare our results with those reported in the literature. Generally, the explore of Bouoiyour and Selmi ( two thousand fifteen ) seems to be the most comparable examine to ours in terms of the employed estimation technics, underlying data and explanatory variables. Similarly to our paper, they control for all three sets of drivers: market coerces, appeal indicators and macro variables. Other studies are less comparable to our treatment mainly due to differences in model specification, as most of them explore the influence of each BitCoin price driver separately; they do not consider interactions inbetween them. For example, Kristoufek ( two thousand thirteen ) considers only BitCoin appeal indicators, whereas Van Wijk ( two thousand thirteen ) includes only the macroeconomic drivers. Buchholz et al. ( two thousand twelve ) account for the impacts of BitCoin supply and request, and BitCoin appeal, but does not account for global macroeconomic and financial developments.

The impacts of BitCoin supply and request (driver 1) on BitCoin price are measured by the ratio of exchange and trade transactions and velocity as (the frequency at which one unit of BitCoin is used to purchase goods) in Bouoiyour and Selmi ( two thousand fifteen ). Buchholz et al. ( two thousand twelve ) measure BitCoin request by price volatility. According to Bouoiyour and Selmi ( two thousand fifteen ), the estimated short-run influence of transactions on BitCoin price is positive and statistically significant (0.129, Table seven in Bouoiyour and Selmi). Compared to our not significantly different from zero estimates (−0.018 to 0.003, models 1.1 and 1.Five in Table four ) they are also smaller by around one order of magnitude. These differences may be caused by the fact that Bouoiyour and Selmi ( two thousand fifteen ) proxy for transactions using the ratio of exchange and trade transactions, whereas our proxy for transaction variable is directly constructed from the number of BitCoin transactions. Note also that our short-run estimates are not statistically different from zero. The estimated long-run influence of transactions is also positive and statistically significant at the ninety % level in Bouoiyour and Selmi ( two thousand fifteen ) (0.018, Table seven in Bouoiyour and Selmi). Remarkably, the long-run influence of transactions is considerably lower (by one order of magnitude) than the short-run influence. In contrast, our estimates suggest that the long-run effect of transactions is negative and considerably higher (by three orders of magnitude) (−3.99 and −3.42, models 1.1 and 1.Five in Table six ) than the short-run effect. Note that our estimates are significant at the ninety five and ninety nine % levels, respectively). Given that our specification is more general, we use longer time series and our results are more significant, a negative long-run relationship inbetween the number of transactions and BitCoin price is more likely.

According to Bouoiyour and Selmi ( two thousand fifteen ), the estimated short-run influence of BitCoin velocity is positive and considerably higher (Two.741, Table seven in Bouoiyour and Selmi) compared to our negative and rather petite estimates (−0.011 to −0.002, models 1.1–1.Four in Table four ). Note, however, that the coefficient of Bouoiyour and Selmi ( two thousand fifteen ) is not statistically significant, whereas it is statistically significant for all estimated models with one and two lags in our results. Again, these differences may be explained by differences in the velocity variable construction and model specifications. Whereas Bouoiyour and Selmi ( two thousand fifteen ) calculate velocity as the frequency at which one unit of BitCoin is used to purchase goods, we measure velocity by the number of days needed until a BitCoin is ruined. The estimated long-run influence of transactions is positive in Bouoiyour and Selmi ( two thousand fifteen ) as well (0.004, Table seven in Bouoiyour and Selmi). However, it is not statistically significant. As in the case of transactions, the long-run influence of velocity is considerably lower (by three orders of magnitude) than the short-run influence in Bouoiyour and Selmi. In contrast, our estimates suggest that the long-run effect of velocity is positive and considerably higher (by two orders of magnitude) (Five.07–11.71, models 1.1, 1.Three, 1.Four and 1.Five in Table six ) than the short-run effect. Note that the magnitude of the estimated long-run coefficients is very stable across all for models, and they all are significant at the ninety nine % level. Given that our specification is more general, we use longer time series and our results are more significant, a positive long-run relationship inbetween velocity and BitCoin price is more likely.

Buchholz et al. ( two thousand twelve ) estimate the relationship inbetween BitCoin price and transactions, however they are interested in the switch roles relationship, as they attempt to response the question how BitCoin price shocks affect BitCoin transactions. Hence, their results are not directly comparable to our estimates. Buchholz et al. ( two thousand twelve ) also estimate the relationship inbetween BitCoin request and BitCoin price, whereas BitCoin request is proxied by price volatility. They find that an enhancing price volatility leads to higher BitCoin price. The coefficient from the GARCH-in-mean model is statistically significant. Moreover, the results are different inbetween the two analyzed periods (August 2010–June two thousand eleven vs. June 2011–March 2012): in the very first period the volatility led to a request for BitCoins, and after the bubble burst in June two thousand twelve the novelty of BitCoin decreased, only market participants who were reluctant to volatility stayed in the market, leading to no effect of volatility on BitCoin price in the 2nd period. However, as above, their proxy for market demand—price volatility—is very different from our proxy—number of transactions—and hence cannot be directly compared with our estimates. Moreover, Buchholz et al. ( two thousand twelve ) do not consider all types of BitCoin price drivers and interactions inbetween them, and they use weekly data for a rather brief period (82 weeks), whereas our analysis is based on daily data for a Five.Five year period, capturing the extreme price switches in two thousand thirteen and two thousand fourteen (see Fig. One ).

The influence of BitCoin attraction (driver Two) on BitCoin price is measured by BitCoin searches in Google in Bouoiyour and Selmi ( two thousand fifteen ) and Buchholz et al. ( two thousand twelve ), whereas Kristoufek ( two thousand thirteen ) uses both BitCoin searches in Google and Wikipedia views.

The estimated short-run influence of searches in Google is positive and sizeable in Bouoiyour and Selmi ( two thousand fifteen ), when compared to our negative and considerably smaller estimates. These differences could be explained by differences in variable construction and model specifications. While Bouoiyour and Selmi ( two thousand fifteen ) use the number of BitCoin searches in Google to proxy for investors’ appeal, we use the number of views of BitCoin site on Wikipedia. The estimated long-run influence of searches in Google is positive and smaller than the short-run estimates (0.143 and 0.527, respectively, Table seven in Bouoiyour and Selmi) but statistically not significantly different from zero. Again, our long-run estimates (1.94, model Two.1 in Table six ) are of higher order of magnitude with respect to our short-run estimates. Note that our model specification is more general, and our estimates are significant at the ninety nine % level, which suggests that they may be more precise.

Kristoufek ( two thousand thirteen ) studies the influence of BitCoin attraction on its price by using BitCoin searches in Google and Wikipedia views. He finds that the enlargened interest in BitCoin measured by BitCoin searches in Google increases its price. As the interest in BitCoin increases, the request increases as well causing the prices to increase. The estimated positive influence is in line with our results. The relationship inbetween Wikipedia views and BitCoin turns out to be statistically not significant in Kristoufek ( two thousand thirteen ). In contrast, our long-run results suggest a positive and statistically significant influence of Wikipedia views on BitCoin price (1.94, model Two.1 in Table six ). The statistically significant results of Kristoufek ( two thousand thirteen ) may be caused by the fact that his analysis is based on a rather brief period (Two years) compared to a Five.Five year period in our examine as well as they do not control for other BitCoin price drivers, such as BitCoin supply and request and macroeconomic developments. Note that Kristoufek ( two thousand thirteen ) finds that the relationship is positive and statistically significant in the opposite direction. The estimated coefficient is 0.05, which implies that a ten % switch in BitCoin price is connected to a 0.Five % permanent shift in the Wikipedia views.

Also Buchholz et al. ( two thousand twelve ) use BitCoin searches in Google to estimate the influence on BitCoin. However, they estimate the relationship inbetween Google hits and the number of BitCoin transactions, not BitCoin price. Hence, their results are not directly comparable to our estimates.

The influence of global macroeconomic and financial developments (driver Trio) on BitCoin price is measured by output volume, gold price and Chinese market index in Bouoiyour and Selmi ( two thousand fifteen ). van Wijk ( two thousand thirteen ) measures global macroeconomic and financial developments by the Dow Jones index, the value of the FTSE 100, the value of the Nikkei 225, the Euro—US dollar exchange rate, the Yen—US dollar exchange rate, and several oil price indices.

According to Bouoiyour and Selmi ( two thousand fifteen ), output volume and Chinese market index have positive and statistically significant influence on BitCoin price (0.079 and 0.380, respectively, Table seven in Bouoiyour and Selmi), whereas the influence of gold price is negative and statistically insignificant (−0.241, Table seven in Bouoiyour and Selmi). Our estimates are different, as they rely on different proxies capturing global macroeconomic and financial developments: Dow Jones index and oil price (−0.046 and 0.178, model Three.1 in Table four ). However, our estimates are not significantly different from zero. Both the sign and the magnitude of the estimated long-run coefficients on the output volume and Chinese market index are rather similar to short-run estimates (0.007 and 0.216, Table seven in Bouoiyour and Selmi). The long-run effect of the gold price coefficient decreases by two orders of magnitude (−0.002) compared to the short-run influence in Bouoiyour and Selmi. However, none of the long-run global macroeconomic and financial development coefficients is statistically significant in Bouoiyour and Selmi ( two thousand fifteen ). In contrast, both our global macroeconomic and financial development coefficients have a statistically significant influence on BitCoin price in the long-run. Whereas the influence of the Dow Jones index is positive (16.Ten, model Three.1 in Table six ), the influence of oil price is negative (−4.52, model Three.1 in Table six ). Note that signs of our long-run estimates are the same as those of Bouoiyour and Selmi ( two thousand fifteen ): Dow Jones index and Chinese market index have positive influence, whereas gold price and oil price have negative influence on BitCoin price. Given that our long-run estimates are statistically significant but those of Bouoiyour and Selmi ( two thousand fifteen ) are not, our estimates can be considered as more reliable.

van Wijk ( two thousand thirteen ) finds that the value of the Dow Jones has positive and statistically significant effect on BitCoin price in the long-run (0.525, Table Three.Three in van Wijk). The positive and statistically significant coefficient is in line with our results for models Three.1 and Four.1. The differences in the estimated elasticities may be caused by the fact that the analysis of van Wijk ( two thousand thirteen ) (1) does not consider other determinants of BitCoin price, such as BitCoin request and supply, and BitCoin attraction, and (Two) is based on a <Three year period, whereas our analysis is based on a Five.Five year period, capturing extreme price switches in two thousand thirteen and two thousand fourteen (see Fig. One ). The estimated influence of the Euro–US dollar exchange rate is negative and statistically significant (−0.335, Table Three.Three in van Wijk). The results of van Wijk ( two thousand thirteen ) suggest that, if the US dollar appreciates against Euro, it is most likely to be the case that it also appreciates against the BitCoin. Consequently, an increase in the Euro–US dollar exchange rate leads to a decrease in the amount of US dollar that have to be paid for one BitCoin, which decreases its value. In contrast, our estimates suggest a positive relationship inbetween the Euro–US dollar exchange rate and BitCoin price. However, none of our estimated models is statistically significant (1.Trio, 1.Five and Four.1). Given that the estimates of van Wijk ( two thousand thirteen ) are statistically significant and plausible from an economic theory point of view, likely they are more reliable. van Wijk ( two thousand thirteen ) also estimates the influence of oil price on BitCoin price. The estimated influence is negative and statistically significant (−0.242, Table Three.Trio in van Wijk). These results are in line with our estimates (−4.52, model Three.1 in Table six ). As explained above, differences inbetween the estimates may be caused by the fact that the times series of van Wijk ( two thousand thirteen ) are only half as long as ours, not capturing the extreme price switches in two thousand thirteen and 2014, as well as they do not control for other significant BitCoin price drivers.

Four Conclusions

Currency, or money in general, is typically defined as having three main functions: a medium of exchange, a unit of account, and a store of value. The present paper attempts to analyze BitCoin features, which may facilitate to become a global currency, as well as characteristics, which may impede the use of BitCoin as a medium of exchange, a unit of account, and a store of value.

The very first function of any currency is to intermediate the exchange of goods and services. Being not a legal tender, BitCoin is fully dependent on voluntary adoption by market participants as a medium of exchange. BitCoin primary advantage relative to standard currencies for its use in exchanges is lower costs of transfers as there are minimal costs linked to third-party intermediaries. Albeit, BitCoin has shown a phenomenal growth during the last years, it still has a negligible market presence globally as a medium of exchange. Moreover, evidence tends to support the view that many of BitCoin transactions involve transfers inbetween speculative investors and are not used in exchanges of goods and services. BitCoin faces also several other obstacles in becoming a widely used medium of exchange such as users’ difficulty of procuring fresh BitCoins, the problem of irreversibility of BitCoin transactions, which does not permit to treat disputes, inability to grant loans denominated in BitCoin, and a relatively high level of computer skill required for using and understanding BitCoin.

The 2nd function of a currency is to serve the function of a unit of account by being able to convey the relative value of goods and services in the economy. The major concern of BitCoin is its high price volatility which may reduce its power to fulfill this function accurately. However, entrepreneurial innovations such as market exchange pricing and instantaneous exchange facilities may remedy this problem but not fully eliminate it. Further, nearly-infinite sub-divisibility is a good advantage of BitCoin relative to standard currencies, however in some instances it may pose confusion among consumers given that the price quotation differentiated in the magnitude of several decimal places may be undistinguishable for them.

The third function of a currency is to serve as a store of value over time. Standard currencies are usually inflationary, while BitCoin is associated with deflationary pressures, if it becomes widely used. While this is beneficial to BitCoin holders, the expectation of higher future value of BitCoin may lead to its hoarding, which may reduce its use in exchanges of goods and services. One of the main threats to BitCoin capability to preserve the value to its holders is the security problem linked to cyber-attacks given that BitCoin is a virtual currency, its system is fully internet based and has no oversight institution entrusted to protect the system.

The identified BitCoin features are compared with traditional currencies and their possible influence on BitCoin functions as a currency. We identify several features of BitCoin as a medium of exchange, which differ substantially from traditional currencies: legal tender, motionless costs, network externalities, transaction costs, dispute resolution, credit market, anonymity and transparency. We also find two key characteristics of BitCoin as a unit of account, which differ substantially from traditional currencies: price volatility and divisibility. Eventually, we identify two features of BitCoin as store of value, which differ substantially from traditional currencies: non-inflationary supply and cyber security.

Among all BitCoin features identified and analyzed in this paper, eventually, price volatility is the one with the largest differences compared to standard currencies, such as US dollar, Euro, Yen, British Pound. As shown in Fig. One , since its introduction in 2009, BitCoin price has shown utterly high volatility. One consequence of such extreme price volatility is a threat for BitCoin being accepted as a global currency, as it may fail to accurately convey relative prices of goods and services in the economy and may generate uncertainties to its holders due to its inability to preserve stable value over time. Indeed, a desirable property of any monetary mean is that it holds its value over short-to-medium periods of time, in order not to create distortion when used as a medium of exchange in transactions, a unit of account, and a store of value. Large price movements alter the purchasing power causing risk and costs to firms and consumers. Given that BitCoin is a relatively fresh currency, its price formation is not well understood yet, as there are only few studies on BitCoin price formation available in the literature.

In order to better understand reasons for such enormously high price volatility, in the 2nd part of the paper we attempt to identify drivers of BitCoin price and estimate their importance econometrically. The previous literature suggests three types of drivers determining BitCoin price development: (1) market coerces of BitCoin supply and request, (Two) BitCoin appeal, and (Three) global macroeconomic and financial developments. We apply time-series analytical mechanisms to these drivers using daily data for the 2009–2014 period.

Our empirical results confirm that market coerces of BitCoin supply and request have an influence on BitCoin price, implying that, to a certain extent, the formation of BitCoin price can be explained in a standard economic model of currency price formation. In particular, the demand-side drivers, such as the size of the BitCoin economy and the velocity of BitCoin circulation, were found to influence BitCoin price. Given that BitCoin supply is exogenous, likely, the development of the request side drivers will be among the key determinants of BitCoin price also in the future potentially lading to deflationary pressures if BitCoin use expands.

However, the influence of demand-side drivers on the BitCoin price somehow reduces when controlling for BitCoin attraction, which implies that these drivers emerge to be relatively more significant. In fact, we cannot reject the hypothesis extensively emphasized in the literature that investor speculations are also affecting BitCoin price. The statistically significant influence of Wikipedia views on BitCoin price could be an indicator of speculative short-run behavior of investors, or it may capture the expansion of the request as a medium of exchange of the BitCoin. Additionally, we find that also fresh information influence BitCoin price positively, which may be a result of an enlargened trust among users. As such, speculative trading of BitCoins is not necessarily an undesirable activity, as it may generate benefits in terms of absorbing excess risk from risk reluctant participants and providing liquidity on the BitCoin market. A negative side of the short-run speculative investment is that it may increase price volatility and create price bubbles which has adverse implication for BitCoin users. A further negative side of the long-run speculative investment is more extensive hoarding of BitCoins which may reduce its use in exchanges. The success of BitCoin thus also hitches on its capability to reduce the potential negative implications of such speculations and expand the use of BitCoin in trade and commerce.

Ultimately, our econometric estimates do not support previous findings that the global macro-financial development may be driving BitCoin price. In fact, once we control for supply–demand variables and BitCoin’s attraction for investors linked to BitCoin, the influence of global macro-financial development captured by the Dow Jones Index, exchange rate and oil price becomes statistically insignificant. Because macroeconomic switches are not reflected in BitCoin price movements its price volatility cannot be lightly hedged. This is in contrast to standard currencies which are strenuously driven by macro developments and hedging option is available and widely used.

In summary, our investigate has shown that there is a disagreement in the literature on whether BitCoin can become a global currency. Overall, negative assessments about BitCoin as a currency tend to prevail given that several BitCoin characteristics identified in the paper impede its use as a currency and thus constrain its expansion globally. However, as outlined in the paper some BitCoin characteristics give its predisposition to be adopted at least in some segment of money market if not in a broader context. In particular, BitCoin may have a high relative comparative advantage with respect to standard currencies in countries with unstable financial system (e.g. in developing countries), and may provide an alternative to standard currencies in countries with poor and not widely available financial services, non-convertible currency, expensive financial services and high administrative cargo in opening an account. In addition, BitCoin may represent a cost-effective remittance system in developing countries, were traditional transfers are very expensive and the banking system is underdeveloped and unsecure. Given that BitCoin transfers can be done with relatively minimal cost and resource requirements and are independent of geographical location or banking system in place, they are ideally placed to serve as an efficient international remittance system. Ten

The BitCoin technology—BlockChain eleven —opens up several other possibilities and technological innovations, including micropayments, crowdfunding, distributed exchanges, clever property, property registry, ticketing and secure voting systems. For example, BlockChain technology could provide a way to track the unique history of individual devices, by recording a ledger of data exchanges inbetween it and other devices, web services, and human users. Similarly, BlockChain could enable clever devices to become independent agents, autonomously conducting a multiplicity of transactions. For example, a vending machine could not only monitor and report its own stock, but also solicit bids from distributors and pay for the delivery of fresh items automatically—based on the purchase history of its customers. Further, the disruptive innovation of BitCoin provides the potential to give citizens direct control over their financial activities by removing costly—and sometimes obscure—intermediation layers fostering financial inclusion .

Footnotes

Note that virtual currencies must be distinguished from electronic money. The key distinguishing feature of electronic money is that their link with traditional money is preserved and have the same unit of account as well as they have legal foundation and are regulated. This is not the case of virtual currencies including BitCoin (ECB two thousand twelve ).

Given that freshly issued BitCoins decrease over time, miners will have to rely more on transaction fees to recoup their investment in mining which may lead to higher transaction fees in the future (EBA two thousand fourteen ). For a theoretical analysis of the economics of BitCoin transaction fees see Kroll et al. ( two thousand thirteen ) and Houy ( two thousand fourteen ).

There was approximately $1.37 trillion in circulation as of June, two thousand fifteen (Federal Reserve System 2015).

Access to the BitCoin network requires downloading a BitCoin software on private computer and joining the BitCoin network, which permits users to engage in operations, and update and verify transactions.

A related problem pointed by Yermack ( two thousand fourteen ) is linked to relatively high diversity of BitCoin prices across different exchanges and web quotations at any given time. This variation in prices poses problem to establish a valid reference point for price setting for both consumers and businesses.

Moreover, BitCoins which were accidentally lost or demolished can never be substituted, resulting in shrinkage of the money base and leading to enhanced deflationary trend. For example, the bitomat exchange, one of the largest exchanges, lost around $200 thousand worth of BitCoins (at the exchange rate at the time) due to human error by accidentally erasing cloud server where the wallet was stored (Barber two thousand twelve ).

Given that people consider a currency valuable if they expect others to do so, for a decentralized currency, such as BitCoin, trust depends on a belief that the rules of the currency will be stable over time.

The Dow Jones Index is an industrial average that captures thirty major corporations on either the NYSE or the NASDAQ.

An example of the BitCoin based system for remittance transfers is BitPesa. BitPesa is an online payment platform that uses BitCoin to suggest money transfers to and from East Africa (Folkinshteyn et al. Two thousand fifteen ).

BlockChain is a public ledger of all BitCoin transactions that have ever been executed; it is permanently growing as ‘completed’ blocks are added to it with a fresh set of recordings.

Notes

Acknowledgments

The authors are grateful to Tony Tam for providing access to the BitCoin data of Bitcoinpulse. We gratefully acknowledge financial support from the research project VEGA1/0797/16. The authors are solely responsible for the content of the paper. The views voiced are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission.

The digital agenda of virtual currencies: Can BitCoin become a global currency, SpringerLink

The digital agenda of virtual currencies: Can BitCoin become a global currency?

  • Pavel Ciaian Email author
  • Miroslava Rajcaniova
  • d’Artis Kancs

Abstract

This paper identifies and analyzes BitCoin features which may facilitate BitCoin to become a global currency, as well as characteristics which may impede the use of BitCoin as a medium of exchange, a unit of account and a store of value, and compares BitCoin with standard currencies with respect to the main functions of money. Among all analyzed BitCoin features, the extreme price volatility stands out most clearly compared to standard currencies. In order to understand the reasons for such extreme price volatility, we attempt to identify drivers of BitCoin price formation and estimate their importance econometrically. We apply time-series analytical mechanisms to daily data for the 2009–2014 period. Our estimation results suggest that BitCoin attraction indicators are the strongest drivers of BitCoin price followed by market compels. In contrast, macro-financial developments do not determine BitCoin price in the long-run. Our findings suggest that as long as BitCoin price will be mainly driven by speculative investments, BitCoin will not be able to contest with standard currencies.

Keywords

1 Introduction

A broad range of virtual currencies have emerged during the last decade, such as BitCoin, LiteCoin, PeerCoin, AuroraCoin, DogeCoin and Ripple. The most successful among them is BitCoin, both in terms of its astounding growth in the number of currency users and popularity by retailers. Since its introduction in 2009, BitCoin has been characterized also by a phenomenal increase in the number of transactions and market capitalization, which surpassed five billion US dollar in two thousand fifteen and since then has recorded further growth.

In the public media and in scientific community there is ongoing a upbeat debate on wheather BitCoin can actually function as a substitute for standard currencies such as US dollar, Euro or Yen. There are two rivaling views in the literature on whether BitCoin fulfils the three criteria of a currency (a medium of exchange, a unit of account, and a store of value) (Mankiw two thousand seven ). One part of the literature argues that BitCoin does not behave as a real currency, but rather resembles speculative investments (e.g. Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ; Williams two thousand fourteen ). Other line of research, stresses the positive aspects of BitCoin and perceives it as a global virtual currency with strong future potential (e.g. Plassaras two thousand thirteen ; Satran two thousand thirteen ; Luther and White two thousand fourteen ; Folkinshteyn et al. Two thousand fifteen ). Also from a legal point of view, there is no global agreement on the status of BitCoin, as there are no international laws regulating BitCoin. Each country regards BitCoin differently and regulations are permanently evolving. For example, the Internal Revenue Service in the US has assigned BitCoin as barter on the grounds of the market-orientated treatment. Finland considers BitCoin as priced commodity. Germany has recognized it as private currency. Yet other EU Member States by and large are in the wait-and-see stage. On the other mitt, institutions such as European Central Bank (ECB) and European Banking Authority (EBA) define BitCoin as virtual currency.

In light of these and other open questions about BitCoin, in April two thousand fifteen the European Commission organized a Virtual Currencies Conference, one where among others the potential and challenges of BitCoin to become a global currency were discussed. The Virtual Currencies Conference acknowledged the need for further research to better understand the socio-economic and monetary aspects of the virtual currency ecosystem.

In order to shed light on these and related questions, the present paper attempts to identify and analyze BitCoin features which may facilitate BitCoin to become a global currency, as well as characteristics which may impede the use of BitCoin as a medium of exchange, a unit of account, or a store of value. Among the BitCoin features, which may facilitate its use as a currency, we have identified low transaction costs, high anonymity and privacy, learning spillover effects, infinite divisibility and no inflationary pressures. Among the BitCoin features, which may impede its use as a currency include the absence of a legal tender attribute, difficulty to procure BitCoins, relatively high immobile costs of adoption, dependence on network externalities, absence of an institution enforcing dispute resolution, absence of BitCoin denominated credits, deflationary pressure, utterly high price volatility, and issues with cyber security.

Our 2nd contribution to the literature is to undertake a comparative analysis of the identified BitCoin features with traditional currencies and their possible influence on BitCoin functions as a currency. Our analysis builds on previous studies, which have identified several advantages of BitCoin over traditional currencies. Very first, as noted by EPRS ( two thousand fourteen ), EBA ( two thousand fourteen ), and Folkinshteyn et al. ( two thousand fifteen ), BitCoin has no physical representation, such as paper bills or metal coins, which saves costs related to the production, transportation, and treating of physical currency. These costs can be substantial for standard currencies. Similarly, an advantage of BitCoin is that it is more convenient than standard payment mechanisms for small-value purchases, as it permits for money transfers at low costs and relatively swift (Hayes et al. One thousand nine hundred ninety six ). As noted by Folkinshteyn et al. ( two thousand fifteen ), the advantages of low transaction fees and brief execution time make BitCoin an ideal medium of exchange. In contrast, Grinberg ( two thousand eleven ) opposes this view and argues that the response of the traditional e-commerce sector to the competitive pressure from BitCoin may induce them to reduce their transaction costs and thus offset the advantage of BitCoin, while providing also other benefits such as higher security. 2nd, an extra advantage of BitCoin, compared to standard currencies, is that it may reduce opportunities for theft, such as bank robbery. BitCoin could help curtail vandalism of vending machines, public phones, etc., because there would be no cash to steal. Similarly, businesses who treat cash, such as taxi drivers and puny shops, could be much less vulnerable to robbery, if they would use BitCoin. Ultimately, BitCoin is a global currency, implying that there are no transaction costs related to currency exchange.

On the other palm, BitCoin faces several challenges compared to standard currencies. One of the main challenges of BitCoin is its security (Moore and Christin two thousand thirteen ; Böhme et al. Two thousand fifteen ; Yermack two thousand fourteen ). Given that BitCoin transactions take place exclusively over the internet, cyber-security is its main threat. Particularly vulnerable to cyber-attacks are large holdings of BitCoins as well as BitCoin exchanges. The security problem is largely attributed to the lack of an oversight institution that would ensure security of BitCoin transactions and BitCoin system (ECB two thousand twelve ; Plassaras two thousand thirteen ; Moore and Christin two thousand thirteen ). 2nd, the use of a fresh currency, such as BitCoin, requires an initial investment for both businesses and consumers (Velde two thousand thirteen ). This includes costs linked to getting acquainted with BitCoin system in general and to the adoption of the payment technology in particular. For example, BitCoin requires the use of electronic devices including specific software. Similarly, BitCoin may also suffer from information asymmetry, as its system is relatively complicated and therefore may not be lightly understood by all potential users (ECB two thousand twelve ). Third, Yermack ( two thousand fourteen ) argues that BitCoin may fail to become a global medium of exchange, as it is used in too few exchanges of goods and services; it has a negligible market presence globally. Presently, there are around 0.11 million BitCoin transactions executed per day (BlockChain two thousand fifteen ) and most of these BitCoin transactions involve transfers inbetween speculative investors. BitCoin use for purchases of goods and services is only minor—20 % of all BitCoin transactions (Yermack two thousand fourteen ). For example, if compared to two hundred ninety five million conventional payments and terminal transactions per day done in Europe alone, the market share of BitCoin shows up to be minuscule (EBA two thousand fourteen ). Fourth, Yermack ( two thousand fourteen ) argues that BitCoin is ineffective as a implement of risk management of price volatility as it is uncorrelated with major world currencies (i.e. US dollar, Euro, Yen, British Pound) or gold price, which is a common use of currencies by businesses for hedging the risks associated with currency volatility. Macroeconomic switches that cause adjustment in exchange rates of currencies are not reflected in any way in BitCoin price movements implying that its price volatility cannot be lightly hedged. Ultimately, BitCoin faces the problem of network externalities in its adoption (Gowrisankaran and Stavins one thousand nine hundred ninety nine ), as its benefit in exchange is positively correlated with the number of users (Plassaras two thousand thirteen ; ECB two thousand twelve ).

BitCoin and Euro price development in USD, 2009–2011. Source: BlockChain and Oanda

In order to better understand the reasons for such utterly high price volatility, we attempt to identify drivers of BitCoin price and estimate their importance econometrically. This is our third contribution to the literature. Previous studies (e.g. Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ) suggest three types of drivers determining BitCoin price development: (1) market coerces of BitCoin supply and request (Two) BitCoin attraction, and (Trio) global macroeconomic and financial developments. We apply time-series analytical mechanisms to these drivers using daily data for the 2009–2014 period. Our estimation results suggest that BitCoin attraction variables are most significant drivers of BitCoin price followed by market compels. Our estimates do not support previous findings that macro-financial developments are driving BitCoin price in the long-run.

Generally, our findings suggest that as long as BitCoin price will be mainly driven by speculative investments, BitCoin will not be able to rival with standard currencies.

The rest of the paper is structured as goes after. Section two provides background information about BitCoin, which is a relatively fresh virtual currency. In the context of the three currency criteria, the key features are discussed and contrasted with standard currencies to analyze whether BitCoin fulfills the main functions of money. Section three identifies the main factors affecting BitCoin price based on previous studies; outlines the econometric approach—a Vector Auto Regressive (VAR) model—and specifies the estimated model. It also details the data sources used in the empirical analysis, the construction of the estimable model’s variables, and discusses the estimation results. The final section concludes.

Two Is BitCoin a currency?

Two.1 Background of BitCoin

Type of currencies

Money (currency) format

Certain types of local currencies

Banknotes and coins

Commercial bank money (deposits)

The most popular virtual currency is BitCoin, created by the Japanese programmer Satoshi Nakamoto in two thousand nine (Nakamoto two thousand nine ). It was the very first open source virtual currency, as BitCoin is managed by an open source software algorithm that uses the global internet network both to create BitCoins as well as to record and verify its transactions. Being a cryptocurrency, BitCoin uses the principles of cryptography to control the creation and exchange of BitCoins. BitCoins can be stored in local wallets (e.g. individual computer, smartphone) using an open-source software or in an online wallet (Brito and Castillo two thousand thirteen ; Murphy two thousand thirteen ; CoinDesk 2015b ).

Compared to standard fiat currencies, such as US dollars or Euro, a distinguishing feature of BitCoin is that the quantity of units in circulation is not managed by a person, group, company, central authority, or government, but by a software algorithm. BitCoins are created in a ‘mining’ process, in which computer network participants, i.e. users who provide their computing power, verify and record payments into a public ledger called BlockChain. In come back for this service they receive transaction fees and freshly minted BitCoins. A immobilized amount of BitCoins is issued at a constant a priori defined and publicly known rate, according to which the stock of BitCoins increases at a decreasing rate. In two thousand one hundred forty the growth rate of BitCoin will converge to zero, when the maximum amount of BitCoins in circulation will reach twenty one million units; according to the current algorithm it will not switch after 2140.

BitCoin does not have physical representation. Instead, it is stored either on electronic devices (e.g. individual computer, mobile, tablet) or entrusted to an online service and is transferred via the internet. BitCoins can be spent on both goods and services, if accepted by the retailer. Users interact with each other directly and anonymously, without third-party intervention. There is neither a central clearing house nor any other intermediary institution involved in the transactions such as central bank or government agency (Table one ). Presently, BitCoins can be acquired either (1) by exchanging them for standard money (e.g. US dollar, Euro) on a BitCoin exchange or from a BitCoin dealer, (Two) by obtaining them from sales of goods or services denominated in BitCoin, or (Three) through a mining process (Plassaras two thousand thirteen ; CoinDesk 2015c ).

BitCoins can be used to buy goods or services worldwide, provided that the transaction fucking partner accepts BitCoin as a mean of payment. A transaction implies that the present holder of BitCoins transfers the ownership of a certain amount of BitCoins to a different market participant in exchange for other currencies, goods or services. A continuously growing number of companies accept BitCoins as payments for their goods and services, at the beginning of two thousand fifteen there were more than 100,000 venues accepting BitCoins (CoinDesk 2015a ; Cuthbertson two thousand fifteen ).

There are two rivaling views in the literature on whether BitCoin fulfils the three criteria of a currency (a medium of exchange, a unit of account, and a store of value). One part of the literature argues that BitCoin does not behave largely like a real currency, as it does not fulfil the main functions of a currency, but rather serves as a vehicle for speculative investments (Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ; Williams two thousand fourteen ). Other line of research stresses the potential of BitCoin and perceives it as a global currency with strong future potential (e.g. Plassaras two thousand thirteen ; Satran two thousand thirteen ; Luther and White two thousand fourteen ; Folkinshteyn et al. Two thousand fifteen ). In the following we attempt to identify and analyze the most significant advantages and disadvantages of BitCoin in the context of the three currency functions.

Two.Two Medium of exchange

Currency characteristics of BitCoin with respect to standard currencies

Advantage of BitCoin

Disadvantage of BitCoin

Medium of exchange

Anonymity and privacy

Not legal tender and difficulty to procure BitCoins

Learning spillover effect

Immobilized costs of adoption

Dispute resolution not available

Absence of BitCoin denominated credit

Unit of account

Relative price comparability problem

Source Authors’ presentation

Two.Two.1 Transaction costs

An appealing feature of BitCoin when used as a medium of exchange is its comparative advantage in transaction costs relative to standard currencies, as BitCoin transaction fees are considerably lower than comparable costs of traditional means of payment (e.g. payment cards or bank transfers). BitCoin transaction fees cover only the costs of maintaining the system (i.e. clearing system) paid to miners; there are no costs linked to third-party intermediaries, such as a regulatory authority, that would perform validation, storage or security functions; the BitCoin system (maintained by miners) is financed through transaction fees and freshly issued BitCoins. Three Transaction costs of standard currencies are considerably higher as they need to cover all costs of intermediaries. Besides the costs of clearing system, the fees charged for standard money transfers need to also offset the cost of storage, authentication, transport, security, etc. (Šurda two thousand twelve ; EBA two thousand fourteen ). Average transaction fees per transaction for BitCoin transfers are inbetween zero and one %, while traditional online payment systems charge fees inbetween two and five % or even more (EPRS two thousand fourteen ; EBA two thousand fourteen ; Folkinshteyn, Lennon and Reilly two thousand fifteen ). Estimates for the US dollar display that these costs (costs of processing and accounting of money, storage, transport, and security) account to as much as $60 billion annually (ca. Four.Four % of the US dollar value). Four Further, BitCoin offers swifter transaction execution than traditional online payment systems. The total processing time of BitCoin transfers is inbetween ten and sixty min (EBA two thousand fourteen ).

We may conclude that comparably low transaction and maintenance costs together with rapid execution of transactions may facilitate the acceptance and use of BitCoin as a currency (Table two ).

Two.Two.Two Anonymity and transparency

An area, where BitCoin has a relatively high appeal, involves exchanges linked to illegal and criminal activities (e.g. money laundering; narcotic trade, tax evasion). While this was not the intent of BitCoin creators, it is rather the way individuals choose to use it, as certain BitCoin characteristics give its predisposition to be advantageous in such activities. For example, BitCoin payment transactions are anonymous and do not require the provision of individual identity information. Further, BitCoin permits for effortless international transfers (e.g. money laundering) without supervision, as BitCoin transactions are peer-to-peer and require just internet access. With its infrastructure being spread across the globe, it is difficult to intercept individual transactions (Bryans two thousand fourteen ; EBA two thousand fourteen ). These features grant BitCoin a rather high predisposition to be used in illegal activities. For example, the internet portal Silk Road, created in January 2011, provided online marketplace for the sale of illegal narcotics and weapons using BitCoins for payment, accounting for as much as half of all BitCoin transactions. Silk Road was shut down following an FBI investigation in 2013. Often it is argued that this event generated benefit to BitCoin by boosting its popularity (EBA two thousand fourteen ; Yermack two thousand fourteen ).

Gambling is other area where BitCoin shows up to be growing as a medium of exchange, as it protects costumer privacy and permits receiving funds from consumers not being able to use other payment methods. The most popular BitCoin denominated online gambling site is Satoshi Dice operating since two thousand twelve (Böhme et al. Two thousand fifteen ). Generally, there is a growing number of online gambling sites, the BitCoin Wiki ( two thousand fifteen ) lists around one hundred BitCoin based casinos, poker sites, bingo games, betting services and lotteries.

The BitCoin platform is see-through and public, meaning that anyone is able to go after the chain of transaction. All BitCoin payments have a traceable history that can be viewed by anyone. A single anonymity breach can uncover a user’s entire BitCoin transaction history. For example, the skill of the identity of any user from any transaction (e.g. obtained from the mailing address used for delivery of purchased goods with BitCoin, or from the bank account used to purchase BitCoin) permit to track that user’s transactions backward and forward through the BlockChain history. Hence, BitCoin system is often referred as pseudonymous in the sense that total history of all transactions and every BitCoin is preserved on the publicly ledger. However, several options are available to overcome this anonymity problem and make transactions non-traceable, for example, by using fresh addresses for each payment received, using switch addresses when sending payments, or using BitCoin mixer services to break the link inbetween an user and its BitCoins (Bitcoinhelp two thousand fourteen ; Crawford two thousand fourteen ; BitCoin two thousand fifteen ; Böhme et al. Two thousand fifteen ).

We may conclude that relatively high anonymity of BitCoin users and transparency of BitCoin transactions may facilitate the popularity and use of BitCoin as a currency (Table two ).

Two.Two.Trio Legal tender

One distinguishing feature of BitCoin is that it is not a legal tender (as opposed to standard currencies which would imply its mandatory acceptance in exchanges). Private or public businesses are not legally obliged to accept BitCoin as a payment form for goods and services they trade. The use of BitCoin as a medium of exchange is thus fully dependent on its voluntary adoption by market participants (EBA two thousand fourteen ).

According to Yermack ( two thousand fourteen ), an extra obstacle for BitCoin to become a widely used medium of exchange, arises from the difficulty of procuring fresh BitCoins. BitCoins can be obtained only from online exchanges or dealers (except for successful BitCoin miners).

We may conclude that, generally, the absence of legal tender may impede BitCoin use as a currency, because any business can determine individually on the acceptance/not acceptance of BitCoin (Table two ).

Two.Two.Four Immobile costs

The use of a fresh currency, such as BitCoin, requires an initial investment for both businesses and consumers. This includes costs linked to getting acquainted with the BitCoin system in general and to the adoption of the payment technology in particular. For example, BitCoin requires the use of electronic devices including specific software. Five

In addition, a relatively high level of computer skill is required for understanding and using BitCoins, which may represent a barrier to many potential users and may constraint its broader adoption (i.e. information asymmetry problem). BitCoin is based on a complicated code, which is understood by only a few persons and is managed by even fewer individuals (programmers), who oversee the entire system without accountability, arbitration or recourse (Velde two thousand thirteen ; Yermack two thousand fourteen ).

On the other arm, BitCoin use may increase users’ interaction with virtual currencies and thus may have learning spillover effects by helping to improve their abilities and skill in virtual transactions and online financial applications (Hayes et al. One thousand nine hundred ninety six ; Berentsten one thousand nine hundred ninety eight ; Plassaras two thousand thirteen ).

We may conclude that in presence of high motionless costs, the popularity and use of BitCoin as a currency may be gravely impeded at least in the short-run, as presently the technical skill about BitCoin is still rather low in society. However, BitCoin may generate some extra benefits to its users through learning spillover effects (Table two ).

Two.Two.Five Network externalities

The incentives for market participants to use BitCoin depend on the number of existing users. That is, the benefit of making the investment and using it in exchanges depends on the number of other suppliers and consumers of goods and services using BitCoin. If only a few businesses accept BitCoins, there are little incentives for consumers to acquire them. On the other palm, if only a few consumers use BitCoins, businesses have little incentives to invest into the equipment for processing BitCoin payments for their goods and services. This is a well-known problem of network externalities (Gowrisankaran and Stavins one thousand nine hundred ninety nine ). Hence, one of the main BitCoin challenges in becoming a global currency is to woo users to use it in their purchases and businesses to accept it as payment form for their goods and services (Berentsten one thousand nine hundred ninety eight ; ECB two thousand twelve ; Plassaras two thousand thirteen ).

Yermack ( two thousand fourteen ) argues that BitCoin largely fails to be a global medium of exchange as it is used to a limited extent to intermediate the exchange of goods and services. Presently there are around 0.11 million BitCoin transactions executed per day (BlockChain two thousand fifteen ) which is insignificant compared, for example, to two hundred ninety five million conventional payments and terminal transactions per day done in Europe alone (EBA two thousand fourteen ). According to BlockChain ( two thousand fifteen ), the use of BitCoin as a medium of exchange is continuously enhancing over time as more and more businesses tend to accept it as a form of payment. Hence, despite the rather low BitCoin invasion in the global exchanges presently may not be an obstacle for BitCoin to become a global medium of exchange in future. Very first, given that BitCoin is a relatively fresh currency, it may take some time till it gets adopted in a broader context. 2nd, there exist examples of puny countries with own currencies (e.g. Fiji) which intermediate the exchange of goods and services within the national boundaries but have only a minor share globally.

We may conclude that the issue of network externalities may impede growth of BitCoin use as a currency in the short-run. However, these issues may become less relevant in future, as BitCoin use continuously increases (Table two ).

Two.Two.6 Dispute resolution

The use of BitCoin in market exchanges bears a certain risk, because of the absence of any protection against disputes inbetween parties involved in the exchange. Once a BitCoin transaction is realized, it is irreversible and cannot be disputed. There is no centralized mechanism available to revert an erroneous transaction or to treat the disputes with the aim to provide protection against human errors of fraud that may occur in exchanges (e.g. protection against disputes over non-fulfillment of contract). Presently, the correction of an erroneous transaction is possible only through a voluntary agreement of the parties involved in the exchange (EBA two thousand fourteen ; Böhme et al. Two thousand fifteen ).

We may conclude that the absence of an institution regulating and enforcing BitCoin related disputes, the popularity and use of BitCoin as a currency may be impeded, particularly for risk-averse market participants (Table two ).

Two.Two.7 Credit market

Another obstacle for BitCoin to become a widely used medium of exchange may be that BitCoin cannot be used to take loans because, under the current system, every loan would need to be made in BitCoin. Whereas the basis of standard banking system is fractional-reserve, which determine how much fresh money can be created through loans, BitCoins are unique and cannot be duplicated, they only exist as an electronic analog a kind of physical coin. This thresholds the expansion of BitCoin (Hanley two thousand fourteen ). For example, the absence of BitCoin denominated credit cards and consumer loans do not permit purchases on credit—a method widely used in most developed retail markets, which boundaries the expansion of BitCoin as a medium of exchange (Yermack two thousand fourteen ).

We may conclude that in the absence of BitCoin denominated credit, BitCoin growth is gravely constrained, as the share of credit transactions is rather high (and growing) in modern societies (Table two ).

Two.Three Unit of account

To serve the function of a unit of account, as any other currency, BitCoin should be able to measure the relative value of goods and services and other transactions (e.g. debts) in the economy. We identify two key characteristics of BitCoin as a unit of account, which differ substantially from traditional currencies: divisibility and price volatility (see Table two for summary).

Two.Trio.1 Divisibility

An significant distinguishing feature of BitCoin is its almost infinite divisibility, implying that prices may be quoted in four or more decimal places. Indeed, divisibility is a necessary characteristic of a currency to accommodate the valuation for all types and sizes of transactions.

On the other mitt, the price differences in the magnitude of several decimal places (e.g. four or more) may be confusing to consumers and may pose problems to them to comprehend and compare relative prices of goods and services. For comparison most world currencies use no more than two decimal points for price quotations (Yermack two thousand fourteen ).

We may conclude that, generally, due to almost infinite divisibility possibilities of BitCoin its use and popularity may grow compared to standard currencies, albeit puny price denominations may reduce capability of consumers to accurately distinguish relative prices (Table two ).

Two.Three.Two Price volatility

BitCoin prices demonstrate enormously high short-run volatility, which diminishes its capability in signifying an effective unit of account (Yermack two thousand fourteen ). The frequent BitCoin price switches cause direct and indirect costs to businesses and consumers. Businesses that use BitCoin have to adjust prices frequently, otherwise they may realize decrease in comes back (because of underpriced goods and services) or loss of competitiveness (because of overpriced goods and services). This is particularly problematic for businesses trading outputs in BitCoins, while paying for production factors and intermediate inputs in local standard currency (e.g. US dollar, Euro, Yen, British Pound), causing discrepancy in relative prices inbetween outputs and inputs in presence of high BitCoin price volatility. Frequent price switches in turn become confusing to consumers, as it becomes more difficult to spot the true relative prices of goods and services. Six

Albeit, high price volatility of BitCoin may inflict risk to its holders, entrepreneurial innovations provide alternative solutions which diminish the price volatility risk. For example, market exchange pricing may facilitate price setting to businesses (retailers) and widen spending options to consumers. The market exchange pricing enable retailers to set prices in one currency (e.g. US dollar, Euro, Yen, British Pound) while displaying them to consumers at the same time and automatically updated in more currencies, including in BitCoin, reflecting the current market exchange rates. This system makes relatively costless price tracking in BitCoin to businesses that reflect up-to-date market exchange rate particularly for online sales. Another example includes instantaneous exchange facilities which enable retailers to accept BitCoin as payment without actually receiving BitCoins. This system involves a third party which intermediates the exchange of BitCoins, paid by consumers, to standard currency, which is received by the retailer (e.g. US dollar, Euro, Yen, British Pound). Because sellers never actually receive BitCoins, they avoid the BitCoins exchange risk; the exchange risk is beard by the intermediaries which receive a fee in come back (Luther and White two thousand fourteen ).

In the context of BitCoin use as a unit of account, the utterly high price volatility reduces BitCoin power to convey accurately the relative prices of goods and services in the economy (Table two ). If, as argued by Luther and White ( two thousand fourteen ), financial developments are providing alleviating options for addressing the issue of BitCoin price volatility, then the adverse impacts of price volatility may decline in future.

Two.Four Store of value

The value of the money must remain stable over time to permit their use in exchanges in different points of time. We identify two features of BitCoin as store of value, which differ substantially from traditional currencies: non-inflationary supply and cyber security (see Table two for summary).

Two.Four.1 Non-inflationary supply

Standard currencies are usually inflationary, meaning that their value reduces over time and thus diminishes the capability of the currency to function as a store of value. In contrast, an significant advantage of BitCoin is the protection against inflation as a safe haven from government interference. Given that under the current mechanism the future number of BitCoins is capped to a maximum amount of twenty one million units with no expansion possible beyond this amount, BitCoin will be exposed to deflationary pressures if its use as an investment alternative or for use as a medium of exchange increases. Seven On the one mitt, this is beneficial to BitCoin holders, as it will make them richer over time. On the other forearm, the expectation of higher future value of BitCoins may lead to its hoarding in anticipation of higher price which in turn may reduce its use in exchanges of goods and services. Meiklejohn et al. ( two thousand thirteen ) find that two thousand eleven was a break point for spending-versus hording of BitCoin. After this date the majority of BitCoins (more than sixty %) that were received were spent within less than one month period. The hording prior to two thousand eleven took many BitCoins out of circulation meaning that most of BitCoins from this period are in the “sink” addresses that have never been spent (Ron and Shamir two thousand thirteen ).

We may conclude that, in the absence of inflationary pressures, BitCoin popularity should increase compared to standard currencies, albeit deflationary pressure may act as a countervailing force (Table two ).

Two.Four.Two Cyber security

One of the main threats to BitCoin capability to preserve its value to the holders is the issue of cyber security. In the past, many BitCoin owners have lost their virtual money through online theft. Standard currencies give a possibility to protect it against theft either by physically hiding it (e.g. under mattresses, in a safe) or by depositing it in a bank. BitCoin is a virtual currency and thus it cannot be physically hidden. BitCoins must be held in computer accounts known as virtual wallets. The security of BitCoin wallets has often been a major problem. Even BitCoin exchanges were subject to cyber-attacks and thefts as they are common target of hackers. The most prominent example is the collapse of the largest BitCoin exchange, MtGox, in February two thousand fourteen allegedly leading to a loss of eight hundred fifty thousand BitCoins (Böhme et al. Two thousand fifteen ; Yermack two thousand fourteen ). Moore and Christin ( two thousand thirteen ) find that forty five % of the total studied BitCoin exchanges closed down, and forty six % of the exchanges that closed did not reimburse any claims to consumers.

In the context of BitCoin use as a store of value, BitCoin sensitivity to cyber-attacks and thefts reduces trust in BitCoin as a currency and hence impedes its growth and chances of becoming a global currency (Table two ).

Three BitCoin price and its volatility

Among all BitCoin features identified and discussed above, eventually, price volatility is the one with the largest differences compared to the major world currencies, such as US dollar, Euro, Yen, British Pound. As shown in Fig. One , since its introduction in 2009, BitCoin price has shown utterly high volatility; it has enlargened from zero value at the time of its inception in two thousand nine to around $1100 at the end of two thousand thirteen when it reached the peak followed by a decline to around $225 (see Fig. One ). For comparison, the fluctuation of exchange rate inbetween US dollar and Euro has not exceeded a ±20 % strapped in the same six year period (see Fig. One ). Such market volatility with extreme price movements (±8000 %) makes BitCoin of little use as a unit of account.

One consequence of such extreme price volatility is that BitCoin may not succeed as a global currency to be accepted by suppliers and consumers, as it may fail to accurately convey relative prices of goods and services in the economy and may generate uncertainties to its holders due to its inability to preserve stable value over time. Given that BitCoin is a relatively fresh currency, its price formation is not well understood yet. In order to better understand reasons for such enormously high price volatility, in this section we attempt to identify the drivers of BitCoin price and estimate their relative importance econometrically.

Three.1 Drivers of BitCoin price

The existing studies in the literature (e.g. Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ), suggest three types of drivers determining BitCoin price formation: (1) market compels of BitCoin supply and request, (Two) BitCoin appeal, and (Three) global macroeconomic and financial developments.

Three.1.1 Driver 1: market coerces of BitCoin supply and request

According to Buchholz et al. ( two thousand twelve ), one of the key drivers of BitCoin price is the interaction inbetween BitCoin supply and request on the BitCoin market. According to the quantity theory of money, the BitCoin supply is determined by the total stock of BitCoin in circulation. The BitCoin request is represented by the size of BitCoin economy (i.e. its use in exchanges) and the velocity of BitCoin circulation. The BitCoin velocity measures the frequency at which one unit of BitCoin is used for purchase of goods and services. The quantity theory implies that the price of BitCoin decreases with the velocity and the stock of BitCoins, but increases with the size of BitCoin economy and the general price level.

The request for BitCoin is primarily driven by its value as a medium of exchange. BitCoin does not have intrinsic value like commodity currency such as gold standard. For example, the key difference inbetween the gold standard and BitCoin is that the request for BitCoin is driven solely by its value in future exchange, whereas the request for commodity currency is driven by both its intrinsic value and its value in future exchange. The BitCoin supply is given by the total stock of units put in circulation, which is publicly known and is motionless in the long-run. Whereas BitCoin supply is exogenous, the supply of gold is endogenous, as it responds to switches in production technology (e.g. mining technology for gold) and comebacks. Given the exogeneity of BitCoin supply, the primary driver of its price developments is expected to be the request side shocks. By altering expectations of future use in exchanges, such shocks to request have the potential to produce large swings in the BitCoin price (Luther and White two thousand fourteen ).

Trio.1.Two Driver Two: BitCoin appeal

There are several BitCoin-specific factors which, in addition to traditional currency price determinants, such as market supply and request, determine its request. This is partially linked to the fact that BitCoin has been created relatively recently and to the nature of the currency. (Barber et al. Two thousand twelve ; Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ).

Very first, BitCoin price may be affected by the risk and uncertainty of the entire BitCoin system. Given that BitCoin is a fiat currency and thus intrinsically worthless, it does not have an underlying value derived from consumption or its use in production process (such as gold). The value of a fiat currency is based on trust that it will be valuable and accepted as a medium of exchange also in future (Greco two thousand one ; Kovenock and Vries two thousand two ). Eight The expectations about trust and acceptance are particularly relevant for BitCoin which, being a relatively fresh currency, is in the phase of establishing its market share by building credibility among market participants.

2nd, being a virtual currency, BitCoin is more vulnerable to cyber-attacks than traditional currencies, which can destabilize the entire BitCoin system and eventually lead to a collapse of BitCoin. Such attacks have been frequently occurring in the BitCoin system in the past (Barber et al. Two thousand twelve ; Moore and Christin two thousand thirteen ). As mentioned above, Moore and Christin ( two thousand thirteen ) examined forty BitCoin exchanges and found that eighteen have been closed down after cyber-attacks, while the world’s largest BitCoin exchange, collapsed in two thousand fourteen due to a cyber-attack.

Third, BitCoin price may be affected by its appeal as an investment chance for potential investors. According to Gervais et al. ( two thousand one ), Grullon et al. ( two thousand four ) and Barber and Odean ( two thousand eight ), potential investors’ decisions may be affected by an increase or decrease of attention in the news media. The role of information is particularly significant in the presence of many alternative investment choices and positive search costs. Given that investment request depends on the costs associated with searching for information for potential investment opportunities available on the market, such as, the stock exchange, those investment opportunities which are under a particular attention of news media may be preferred by potential investors, because they reduce search costs. In turn, enhanced investment request for BitCoin may exercise upward pressure on BitCoin price. Indeed, Lee ( two thousand fourteen ) finds such evidence for BitCoin, whereby the alteration of positive and negative news generated high price cycles. This implies that the attention-driven investment behavior can affect BitCoin price either positively or negatively, depending on the type of news that predominate in the media at a given point of time.

Trio.1.Three Driver Three: global macroeconomic and financial developments

Van Wijk ( two thousand thirteen ) stresses the role of global macroeconomic and financial development, captured by variables such as stock exchange indices, exchange rates, and oil prices measures in determining BitCoin price. The influence of macroeconomic and financial indicators on BitCoin price may work through several channels. For example, stock exchange indices may reflect general macroeconomic and financial developments of the global economy. Favorable macroeconomic and financial developments may stimulate the use of BitCoin in trade and exchanges and thus strengthen its request, which may have positive influence on BitCoin price.

Inflation and price indices are other type of indicators capturing significant macroeconomic and financial developments. According to Krugman and Obstfeld ( two thousand three ) and Palombizio and Morris ( two thousand twelve ), oil price is one of the main sources of request and cost pressures, and it provides an early indication of inflation development. Thus, when the price of oil signals potential switches in the general price level, this may lead to depreciation (or appreciation) of BitCoin price. Also the exchange rate may reflect inflation development and thus influence positively BitCoin price as indicated above.

According to Dimitrova ( two thousand five ), there could be also negative relation inbetween a currency’s price and macro financial indicators. A decline in the stock prices induces foreign investors to sell the financial assets they hold. This leads to a depreciation of the underlying currency, but may stimulate BitCoin price, if investors substitute investment in stocks for investment in BitCoin. Generally, investors’ comeback on stock exchange may capture chance costs of investing in BitCoin. Hence, in this case the stock exchange indices are expected to be positively related to BitCoin price.

Trio.Two Econometric treatment

The very first four variables pt, yt, vt, and bt account for request and supply drivers of BitCoin price (driver 1). Following the quantity theory of money we expect that coefficients β1 and βTwo would be positive, whereas βTrio and βFour would be negative. In addition, given that BitCoin supply is largely predefined, the total stock of BitCoins in circulation, bt, is a semi- exogenous variable, and implying that the influence of coefficient βFour on BitCoin price should be puny and/or statistically not significant. Variable at captures the BitCoin attraction (driver Two). As discussed above, the coefficient βFive associated to this variable can be either negative or positive, for example, as both positive and negative news attract investors’ and users’ attention. Ultimately, variable mt represents global macroeconomic and financial developments (driver Three). According to the previous findings discussed above, we expect the sign of the coefficient β6 to be either positive or negative depending on the type of macroeconomic variable.

The econometric model (1) contains mutually interdependent variables—BitCoin price and its explanatory variables. The estimation of non-linear interdependencies among interdependent time series in presence of mutually correlated variables is subject to an issue of endogeneity (Lütkepohl and Krätzig two thousand four ). To circumvent the issue of endogeneity, we go after the general treatment in the literature to analyze the causality inbetween endogenous time-series and specify a multivariate Vector Auto Regressive (VAR) model (Lütkepohl and Krätzig two thousand four ).

According to Engle and Granger ( one thousand nine hundred eighty seven ), regressions of interdependent and non-stationary time series may lead to spurious results. In order to avoid spurious regression, it is significant to test the properties of the time series involved. Therefore, in the very first step, the stationarity of time series is determined, for which we use two unit root tests: the augmented Dickey–Fuller (ADF) test and the Phillips–Perron (PP) test. The number of lags that we use for each dependent variable is determined by the Akaike Information Criterion (AIC). If two individual time series are non-stationary, their combination may be stationary (Engle and Granger one thousand nine hundred eighty seven ). In this special case, the time series are considered to be cointegrated, implying that there exists a long-run equilibrium relationship inbetween them.

In the 2nd step, we employ the Johansen’s cointegration method to examine the long-term relationship inbetween the price series. The number of cointegrating vectors is determined by the maximum eigenvalue test and the trace test. Both tests use eigenvalues to compute the associated test statistics. We go after the Pantula principle (Pantula one thousand nine hundred eighty nine ) to determine whether a time trend and a constant term should be included in the model.

In the third step, we estimate a vector error correction model for those series that are cointegrated (Johansen and Juselius’s one thousand nine hundred ninety ). It includes an error correction term indicating the speed of adjustment of any disequilibrium towards a long-term equilibrium state. In other words, the error correction term captures the switches in the BitCoin price required to eliminate the past deviation of the prices from the equilibrium levels.

As usual, in order to ensure the adequacy of the estimated models, we implement a series of specification tests: Lagrange-multiplier (LM) test for autocorrelation in the residuals; Jarque–Bera test to check if the residuals in the Vector-Error Correction (VEC) are normally distributed and a test of stability of the model.

Three.Three Data and variable construction

In order to construct the dependent variable, we use data for BitCoin price, p t B , denominated in US dollar (BitCoin price). We use the historical number of total BitCoins (number of BitCoins) which have been mined to account for the total stock of BitCoins in circulation, bt. We use two alternative proxies for the size of BitCoin economy, yt: the total number of unique BitCoin transactions per day (number of transactions), and the number of unique BitCoin addresses used per day (number of addresses). Following Matonis ( two thousand twelve ), we proxy the monetary velocity of BitCoin circulation, vt, by BitCoin days ruined for any given transaction (days demolished). This variable is calculated by taking the number of BitCoins in transaction and multiplying it by the number of days since those coins were last spent. All these data are extracted from quandl.com. To measure the price level of global economy, pt, we use exchange rate inbetween the US dollar and the Euro (exchange rate) extracted from the European Central Bank. We use exchange rate inbetween the US dollar and the Euro, because in our data BitCoin price is denominated in US dollar. For example, if the US dollar would appreciate against the Euro, most likely it will also appreciate against the BitCoin. Consequently, an increase in the exchange rate inbetween the Euro and the US dollar would lead to a decrease in the amount of US dollar that have to be paid for one BitCoin, which decreases its price.

In order to capture BitCoin appeal, at, we go after Kristoufek ( two thousand thirteen ) and use the volume of daily BitCoin views on Wikipedia (views on Wikipedia). According to Kristoufek ( two thousand thirteen ), the frequency of searches related to the virtual currency is a good measure of potential investors’ interest in the currency. However, the online search queries, such as Wikipedia views, may measure both investors’ and users’ interest in BitCoin, as it captures information’s request about the currency but it does not differentiate on whether the information is used to guide investment decisions or online BitCoin denominated exchange of goods and services. In addition, we also construct a variable capturing the number of fresh members (fresh members) and fresh posts on online BitCoin forums (fresh posts) extracted from bitcointalk.org. As explained above, the variable fresh members captures the size of the BitCoin economy but also attention-driven investment behavior of fresh BitCoin members. The variable fresh posts captures the effect of trust and/or uncertainty, as it represents the force of discussions among members.

To account for global macroeconomic and financial developments, mt, we go after van Wijk ( two thousand thirteen ) and use oil price (oil price) and the Dow Jones stock market index (Dow Jones). Nine Oil prices are extracted from the US Energy Information Administration, and Dow Jones index is extracted from the Federal Research Bank of St. Louis.

Three.Four Estimation results

Specification of the empirically estimated models

Number of BitCoins

Number of transactions

Number of addresses

Views on Wikipedia

Short-run effects on BitCoin price for drivers 1, two and Trio

L1D BitCoin price

L2D BitCoin price

L3D BitCoin price

L4D BitCoin price

L1D number of BitCoins

L2D number of BitCoins

L3D number of BitCoins

L4D number of BitCoins

L1D number of transactions

L2D number of transactions

L3D number of transactions

L4D number of transactions

L1D number of addresses

L2D number of addresses

L3D number of addresses

L4D number of addresses

L1D days ruined

L2D days ruined

L3D days demolished

L4D days ruined

L1D exchange rate

L2D exchange rate

L3D exchange rate

L4D exchange rate

L1D views on Wikipedia

L2D views on Wikipedia

L3D views on Wikipedia

L4D views on Wikipedia

L1D fresh members

L2D fresh members

L3D fresh members

L4D fresh members

*** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero. L1D indicates one lag, L2D indicates two lags, L3D indicates three lags, and L4D indicates four lags

Short-run effects on BitCoin price for general models

L1D BitCoin price

L2D BitCoin price

L3D BitCoin price

L4D BitCoin price

L1D number of BitCoins

L2D number of BitCoins

L3D number of BitCoins

L4D number of BitCoins

L1D number of transactions

L2D number of transactions

L3D number of transactions

L4D number of transactions

L1D number of addresses

L2D number of addresses

L3D number of addresses

L4D number of addresses

L1D days demolished

L2D days demolished

L3D days ruined

L4D days demolished

L1D exchange rate

L2D exchange rate

L3D exchange rate

L4D exchange rate

L1D views on Wikipedia

L2D views on Wikipedia

L3D views on Wikipedia

L4D views on Wikipedia

L1D fresh members

L2D fresh members

L3D fresh members

L4D fresh members

*** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero. L1D indicates one lag, L2D indicates two lags, L3D indicates three lags, and L4D indicates four lags

Long-run effects on BitCoin price for drivers 1, two and Three

Number of BitCoins

Number of transactions

Number of addresses

views on Wikipedia

Dependent variable: BitCoin price. *** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero

Long-run effects on BitCoin price for general models

Number of BitCoins

Number of transactions

Number of addresses

Views on Wikipedia

Dependent variable: BitCoin price. *** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero

Tables six and seven display the long-run impacts of different determinants on BitCoin price. According to the results reported in Tables five and six , the long-run relationship inbetween BitCoin price and the explanatory variables considered in the estimated models is stronger than the short-run influence. In the following we discuss the long-run results with respect to the three drivers of BitCoin price formation.

Three.Four.1 Results for driver 1: market coerces of BitCoin supply and request

The very first major observation arising from the estimates reported in Tables six and seven is that the market compels of supply and request have an influence on BitCoin price. Generally, the request side variables (e.g. days ruined, number of addresses) emerge to exert a more pronounced influence on BitCoin price than the supply side drivers (e.g. number of BitCoins). According to the results reported in Table six , an increase in the stock of BitCoins (number of BitCoins) leads to a decrease in BitCoin price (model 1.Two), whereas an increase in the size of the BitCoin economy (number of addresses) and its velocity (days ruined) lead to an increase in BitCoin price (models 1.1, 1.Two, 1.Three, 1.Four, 1.Five, Four.Trio, Four.Four, Four.7). These results are in line with our expectations. Contrary to our expectations, the alternative variable that captures the size of the BitCoin economy (number of transactions) has negative influence on BitCoin price in models 1.1 and 1.Five. However, this variable is not significant in the more general models (models Four.1–4.9), implying that its influence on BitCoin price is negligible.

Albeit, the sign of the estimated coefficients for market coerces of BitCoin supply and request is in line with our hypothesis (except for the variable number of transactions in models 1.1 and 1.Five), the statistical significance and magnitude of the estimated coefficients decreases in most models, when accounting for the influence of BitCoin appeal and global macroeconomic and financial developments (models Four.1–4.9 in Table seven ). The supply–demand variables are statistically significant in models Four.Three, Four.Four and Four.7, but have a considerably lower magnitude of the estimated influence than in models 1.1–1.Five, which capture only market coerces of supply and request. This could be explained by the fact that part of the BitCoin price variation explained by the supply–demand variables is absorbed by other variables in more general specifications (models Four.1–4.9).

Trio.Four.Two Results for driver Two: BitCoin attraction

The strongest and statistically the most significant influence on BitCoin price is estimated for variables capturing the influence of BitCoin appeal: views on Wikipedia, fresh members and fresh posts (models Two.1 and models Four.1–4.9). Variable fresh members has negative influence on BitCoin price, implying that attention-driven investment behavior of fresh investors predominates. Variable fresh posts has positive influence on BitCoin price, reflecting an enlargening acceptance and trust in BitCoin captured by the strength of discussion inbetween BitCoin users. This may reflect declining transaction costs and uncertainty for investors, which increases investment request for BitCoin and hence its price.

Consistent with the findings of Kristoufek ( two thousand thirteen ), Wikipedia views have a statistically significant influence on BitCoin price. This variable is significant and has positive influence in all models (except for model Four.8). These results are in line with our expectations and with the remarkable Wikipedia article traffic statistics, according to which on average BitCoin article on Wikipedia is being viewed 200,000 times per month, and is ranked among the top one hundred most viewed articles. Only eighty eight English-language Wikipedia articles have more traffic than the BitCoin article, out of a total of approximately five million English-language articles. However, the interpretation of Wikipedia views is not straightforward, as it may capture various effects. On the one forearm, Wikipedia views may represent attention faithful by potential users’ and/or investors’ to BitCoin likely induced by their reaction to media reporting on BitCoin. The attention effect may influence either positively or negatively the BitCoin price depending on the type of news. The positive estimated coefficient associated with Wikipedia views variable implies that the influence of positive news predominates. One of the most common negative news reported in media about BitCoin is related the security breach (cyber-attacks) against the currency. The positive estimated results indicate that the implication of security problem shows up to be offset by the positive news effect.

On the other arm, Wikipedia views may measure investors’ or users’ interest in BitCoin, as it captures information’s request about the currency. It may reflect switches in the skill about BitCoin inbetween potential investors and/or users, thus leading to a higher acceptance and request for it either as an investment chance or as a medium of exchange. Significant is that the type of individuals searching information about BitCoin on Wikipedia likely are fresh BitCoin users/investors, because Wikipedia contains rather general information about BitCoin, which is known by incumbent investors or advanced BitCoin users. Both investors and users may search for information about BitCoin and thus the Wikipedia effect may capture either the users’ request for information with the aim to use it as a medium of exchange or may reflect investors’ information search who perceive BitCoin as an investment chance (i.e. for hoarding). Kristoufek ( two thousand thirteen ) argues that the second—the hording—effect predominates because, according to him, the market compels of BitCoin supply and request permitting for setting a ‘‘fair’’ price are missing, rather its price is driven by the investors’ faith in the future growth and is predominated by short-term investors, trend chasers, noise traders and speculators. This argument is also in line with the sceptics on BitCoin (e.g. Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ) who argue that hoarding is one of the key weaknesses of BitCoin as a currency (alongside the security problem) compared to standard currencies.

Three.Four.Three Results for driver Trio: global macroeconomic and financial developments

Our findings suggest that, in contrast to previous studies (i.e. van Wijk two thousand thirteen ), global macro-financial developments such as the Dow Jones Index, exchange rate and oil price do not significantly affect BitCoin price in the long-run. Only in Model Trio.1 the macro and financial variables (Dow Jones, oil price and exchange rate) are statistically significant (Table six ). Our results suggest that an increase in oil price leads to a decrease in the budget of consumers (and companies), implying that less money will be spent on other goods, including BitCoin. Consequently, this would lead to a decrease in request for BitCoin, decreasing its price. These results are in line with the estimates of van Wijk ( two thousand thirteen ), who also finds statistically significant influence of macro-financial variables on BitCoin price. However, van Wijk ( two thousand thirteen ) does not account for market compels of supply and request or BitCoin attraction indicators. When these factors are taken into consideration (models Four.1–4.9), their influence decreases considerably in all estimated models (except for model Four.1) (Table seven ). These finding support the argument of Yermack ( two thousand fourteen ) that BitCoin is relatively ineffective as a instrument for risk management against adverse market developments as its price is not responsive to macroeconomic variables meaning that it cannot be lightly hedged against other assets that are driven by macroeconomic developments.

Three.Five Comparison with previous studies

The verification of our results through comparison with previous studies is a challenging task, as (1) the currency is relatively fresh and there are not many studies in the literature, which analyze BitCoin price formation; and (Two) there are significant differences inbetween the used empirical proxies for variable construction and model specification across different studies. Despite these difficulties, in this section we attempt to compare our results with those reported in the literature. Generally, the probe of Bouoiyour and Selmi ( two thousand fifteen ) seems to be the most comparable examine to ours in terms of the employed estimation mechanisms, underlying data and explanatory variables. Similarly to our paper, they control for all three sets of drivers: market compels, appeal indicators and macro variables. Other studies are less comparable to our treatment mainly due to differences in model specification, as most of them explore the influence of each BitCoin price driver separately; they do not consider interactions inbetween them. For example, Kristoufek ( two thousand thirteen ) considers only BitCoin appeal indicators, whereas Van Wijk ( two thousand thirteen ) includes only the macroeconomic drivers. Buchholz et al. ( two thousand twelve ) account for the impacts of BitCoin supply and request, and BitCoin appeal, but does not account for global macroeconomic and financial developments.

The impacts of BitCoin supply and request (driver 1) on BitCoin price are measured by the ratio of exchange and trade transactions and velocity as (the frequency at which one unit of BitCoin is used to purchase goods) in Bouoiyour and Selmi ( two thousand fifteen ). Buchholz et al. ( two thousand twelve ) measure BitCoin request by price volatility. According to Bouoiyour and Selmi ( two thousand fifteen ), the estimated short-run influence of transactions on BitCoin price is positive and statistically significant (0.129, Table seven in Bouoiyour and Selmi). Compared to our not significantly different from zero estimates (−0.018 to 0.003, models 1.1 and 1.Five in Table four ) they are also smaller by around one order of magnitude. These differences may be caused by the fact that Bouoiyour and Selmi ( two thousand fifteen ) proxy for transactions using the ratio of exchange and trade transactions, whereas our proxy for transaction variable is directly constructed from the number of BitCoin transactions. Note also that our short-run estimates are not statistically different from zero. The estimated long-run influence of transactions is also positive and statistically significant at the ninety % level in Bouoiyour and Selmi ( two thousand fifteen ) (0.018, Table seven in Bouoiyour and Selmi). Remarkably, the long-run influence of transactions is considerably lower (by one order of magnitude) than the short-run influence. In contrast, our estimates suggest that the long-run effect of transactions is negative and considerably higher (by three orders of magnitude) (−3.99 and −3.42, models 1.1 and 1.Five in Table six ) than the short-run effect. Note that our estimates are significant at the ninety five and ninety nine % levels, respectively). Given that our specification is more general, we use longer time series and our results are more significant, a negative long-run relationship inbetween the number of transactions and BitCoin price is more likely.

According to Bouoiyour and Selmi ( two thousand fifteen ), the estimated short-run influence of BitCoin velocity is positive and considerably higher (Two.741, Table seven in Bouoiyour and Selmi) compared to our negative and rather puny estimates (−0.011 to −0.002, models 1.1–1.Four in Table four ). Note, however, that the coefficient of Bouoiyour and Selmi ( two thousand fifteen ) is not statistically significant, whereas it is statistically significant for all estimated models with one and two lags in our results. Again, these differences may be explained by differences in the velocity variable construction and model specifications. Whereas Bouoiyour and Selmi ( two thousand fifteen ) calculate velocity as the frequency at which one unit of BitCoin is used to purchase goods, we measure velocity by the number of days needed until a BitCoin is ruined. The estimated long-run influence of transactions is positive in Bouoiyour and Selmi ( two thousand fifteen ) as well (0.004, Table seven in Bouoiyour and Selmi). However, it is not statistically significant. As in the case of transactions, the long-run influence of velocity is considerably lower (by three orders of magnitude) than the short-run influence in Bouoiyour and Selmi. In contrast, our estimates suggest that the long-run effect of velocity is positive and considerably higher (by two orders of magnitude) (Five.07–11.71, models 1.1, 1.Three, 1.Four and 1.Five in Table six ) than the short-run effect. Note that the magnitude of the estimated long-run coefficients is very stable across all for models, and they all are significant at the ninety nine % level. Given that our specification is more general, we use longer time series and our results are more significant, a positive long-run relationship inbetween velocity and BitCoin price is more likely.

Buchholz et al. ( two thousand twelve ) estimate the relationship inbetween BitCoin price and transactions, however they are interested in the switch sides relationship, as they attempt to reaction the question how BitCoin price shocks affect BitCoin transactions. Hence, their results are not directly comparable to our estimates. Buchholz et al. ( two thousand twelve ) also estimate the relationship inbetween BitCoin request and BitCoin price, whereas BitCoin request is proxied by price volatility. They find that an enlargening price volatility leads to higher BitCoin price. The coefficient from the GARCH-in-mean model is statistically significant. Moreover, the results are different inbetween the two analyzed periods (August 2010–June two thousand eleven vs. June 2011–March 2012): in the very first period the volatility led to a request for BitCoins, and after the bubble burst in June two thousand twelve the novelty of BitCoin decreased, only market participants who were reluctant to volatility stayed in the market, leading to no effect of volatility on BitCoin price in the 2nd period. However, as above, their proxy for market demand—price volatility—is very different from our proxy—number of transactions—and hence cannot be directly compared with our estimates. Moreover, Buchholz et al. ( two thousand twelve ) do not consider all types of BitCoin price drivers and interactions inbetween them, and they use weekly data for a rather brief period (82 weeks), whereas our analysis is based on daily data for a Five.Five year period, capturing the extreme price switches in two thousand thirteen and two thousand fourteen (see Fig. One ).

The influence of BitCoin appeal (driver Two) on BitCoin price is measured by BitCoin searches in Google in Bouoiyour and Selmi ( two thousand fifteen ) and Buchholz et al. ( two thousand twelve ), whereas Kristoufek ( two thousand thirteen ) uses both BitCoin searches in Google and Wikipedia views.

The estimated short-run influence of searches in Google is positive and sizeable in Bouoiyour and Selmi ( two thousand fifteen ), when compared to our negative and considerably smaller estimates. These differences could be explained by differences in variable construction and model specifications. While Bouoiyour and Selmi ( two thousand fifteen ) use the number of BitCoin searches in Google to proxy for investors’ appeal, we use the number of views of BitCoin site on Wikipedia. The estimated long-run influence of searches in Google is positive and smaller than the short-run estimates (0.143 and 0.527, respectively, Table seven in Bouoiyour and Selmi) but statistically not significantly different from zero. Again, our long-run estimates (1.94, model Two.1 in Table six ) are of higher order of magnitude with respect to our short-run estimates. Note that our model specification is more general, and our estimates are significant at the ninety nine % level, which suggests that they may be more precise.

Kristoufek ( two thousand thirteen ) studies the influence of BitCoin appeal on its price by using BitCoin searches in Google and Wikipedia views. He finds that the enlargened interest in BitCoin measured by BitCoin searches in Google increases its price. As the interest in BitCoin increases, the request increases as well causing the prices to increase. The estimated positive influence is in line with our results. The relationship inbetween Wikipedia views and BitCoin turns out to be statistically not significant in Kristoufek ( two thousand thirteen ). In contrast, our long-run results suggest a positive and statistically significant influence of Wikipedia views on BitCoin price (1.94, model Two.1 in Table six ). The statistically significant results of Kristoufek ( two thousand thirteen ) may be caused by the fact that his analysis is based on a rather brief period (Two years) compared to a Five.Five year period in our explore as well as they do not control for other BitCoin price drivers, such as BitCoin supply and request and macroeconomic developments. Note that Kristoufek ( two thousand thirteen ) finds that the relationship is positive and statistically significant in the opposite direction. The estimated coefficient is 0.05, which implies that a ten % switch in BitCoin price is connected to a 0.Five % permanent shift in the Wikipedia views.

Also Buchholz et al. ( two thousand twelve ) use BitCoin searches in Google to estimate the influence on BitCoin. However, they estimate the relationship inbetween Google hits and the number of BitCoin transactions, not BitCoin price. Hence, their results are not directly comparable to our estimates.

The influence of global macroeconomic and financial developments (driver Three) on BitCoin price is measured by output volume, gold price and Chinese market index in Bouoiyour and Selmi ( two thousand fifteen ). van Wijk ( two thousand thirteen ) measures global macroeconomic and financial developments by the Dow Jones index, the value of the FTSE 100, the value of the Nikkei 225, the Euro—US dollar exchange rate, the Yen—US dollar exchange rate, and several oil price indices.

According to Bouoiyour and Selmi ( two thousand fifteen ), output volume and Chinese market index have positive and statistically significant influence on BitCoin price (0.079 and 0.380, respectively, Table seven in Bouoiyour and Selmi), whereas the influence of gold price is negative and statistically insignificant (−0.241, Table seven in Bouoiyour and Selmi). Our estimates are different, as they rely on different proxies capturing global macroeconomic and financial developments: Dow Jones index and oil price (−0.046 and 0.178, model Trio.1 in Table four ). However, our estimates are not significantly different from zero. Both the sign and the magnitude of the estimated long-run coefficients on the output volume and Chinese market index are rather similar to short-run estimates (0.007 and 0.216, Table seven in Bouoiyour and Selmi). The long-run effect of the gold price coefficient decreases by two orders of magnitude (−0.002) compared to the short-run influence in Bouoiyour and Selmi. However, none of the long-run global macroeconomic and financial development coefficients is statistically significant in Bouoiyour and Selmi ( two thousand fifteen ). In contrast, both our global macroeconomic and financial development coefficients have a statistically significant influence on BitCoin price in the long-run. Whereas the influence of the Dow Jones index is positive (16.Ten, model Three.1 in Table six ), the influence of oil price is negative (−4.52, model Trio.1 in Table six ). Note that signs of our long-run estimates are the same as those of Bouoiyour and Selmi ( two thousand fifteen ): Dow Jones index and Chinese market index have positive influence, whereas gold price and oil price have negative influence on BitCoin price. Given that our long-run estimates are statistically significant but those of Bouoiyour and Selmi ( two thousand fifteen ) are not, our estimates can be considered as more reliable.

van Wijk ( two thousand thirteen ) finds that the value of the Dow Jones has positive and statistically significant effect on BitCoin price in the long-run (0.525, Table Trio.Three in van Wijk). The positive and statistically significant coefficient is in line with our results for models Trio.1 and Four.1. The differences in the estimated elasticities may be caused by the fact that the analysis of van Wijk ( two thousand thirteen ) (1) does not consider other determinants of BitCoin price, such as BitCoin request and supply, and BitCoin appeal, and (Two) is based on a <Three year period, whereas our analysis is based on a Five.Five year period, capturing extreme price switches in two thousand thirteen and two thousand fourteen (see Fig. One ). The estimated influence of the Euro–US dollar exchange rate is negative and statistically significant (−0.335, Table Three.Trio in van Wijk). The results of van Wijk ( two thousand thirteen ) suggest that, if the US dollar appreciates against Euro, it is most likely to be the case that it also appreciates against the BitCoin. Consequently, an increase in the Euro–US dollar exchange rate leads to a decrease in the amount of US dollar that have to be paid for one BitCoin, which decreases its value. In contrast, our estimates suggest a positive relationship inbetween the Euro–US dollar exchange rate and BitCoin price. However, none of our estimated models is statistically significant (1.Trio, 1.Five and Four.1). Given that the estimates of van Wijk ( two thousand thirteen ) are statistically significant and plausible from an economic theory point of view, likely they are more reliable. van Wijk ( two thousand thirteen ) also estimates the influence of oil price on BitCoin price. The estimated influence is negative and statistically significant (−0.242, Table Trio.Three in van Wijk). These results are in line with our estimates (−4.52, model Trio.1 in Table six ). As explained above, differences inbetween the estimates may be caused by the fact that the times series of van Wijk ( two thousand thirteen ) are only half as long as ours, not capturing the extreme price switches in two thousand thirteen and 2014, as well as they do not control for other significant BitCoin price drivers.

Four Conclusions

Currency, or money in general, is typically defined as having three main functions: a medium of exchange, a unit of account, and a store of value. The present paper attempts to analyze BitCoin features, which may facilitate to become a global currency, as well as characteristics, which may impede the use of BitCoin as a medium of exchange, a unit of account, and a store of value.

The very first function of any currency is to intermediate the exchange of goods and services. Being not a legal tender, BitCoin is fully dependent on voluntary adoption by market participants as a medium of exchange. BitCoin primary advantage relative to standard currencies for its use in exchanges is lower costs of transfers as there are minimal costs linked to third-party intermediaries. Albeit, BitCoin has shown a phenomenal growth during the last years, it still has a negligible market presence globally as a medium of exchange. Moreover, evidence tends to support the view that many of BitCoin transactions involve transfers inbetween speculative investors and are not used in exchanges of goods and services. BitCoin faces also several other obstacles in becoming a widely used medium of exchange such as users’ difficulty of procuring fresh BitCoins, the problem of irreversibility of BitCoin transactions, which does not permit to treat disputes, inability to grant loans denominated in BitCoin, and a relatively high level of computer skill required for using and understanding BitCoin.

The 2nd function of a currency is to serve the function of a unit of account by being able to convey the relative value of goods and services in the economy. The major concern of BitCoin is its high price volatility which may reduce its power to fulfill this function accurately. However, entrepreneurial innovations such as market exchange pricing and instantaneous exchange facilities may remedy this problem but not fully eliminate it. Further, nearly-infinite sub-divisibility is a good advantage of BitCoin relative to standard currencies, however in some instances it may pose confusion among consumers given that the price quotation differentiated in the magnitude of several decimal places may be undistinguishable for them.

The third function of a currency is to serve as a store of value over time. Standard currencies are usually inflationary, while BitCoin is associated with deflationary pressures, if it becomes widely used. While this is beneficial to BitCoin holders, the expectation of higher future value of BitCoin may lead to its hoarding, which may reduce its use in exchanges of goods and services. One of the main threats to BitCoin capability to preserve the value to its holders is the security problem linked to cyber-attacks given that BitCoin is a virtual currency, its system is fully internet based and has no oversight institution entrusted to protect the system.

The identified BitCoin features are compared with traditional currencies and their possible influence on BitCoin functions as a currency. We identify several features of BitCoin as a medium of exchange, which differ substantially from traditional currencies: legal tender, stationary costs, network externalities, transaction costs, dispute resolution, credit market, anonymity and transparency. We also find two key characteristics of BitCoin as a unit of account, which differ substantially from traditional currencies: price volatility and divisibility. Eventually, we identify two features of BitCoin as store of value, which differ substantially from traditional currencies: non-inflationary supply and cyber security.

Among all BitCoin features identified and analyzed in this paper, eventually, price volatility is the one with the largest differences compared to standard currencies, such as US dollar, Euro, Yen, British Pound. As shown in Fig. One , since its introduction in 2009, BitCoin price has shown enormously high volatility. One consequence of such extreme price volatility is a threat for BitCoin being accepted as a global currency, as it may fail to accurately convey relative prices of goods and services in the economy and may generate uncertainties to its holders due to its inability to preserve stable value over time. Indeed, a desirable property of any monetary mean is that it holds its value over short-to-medium periods of time, in order not to create distortion when used as a medium of exchange in transactions, a unit of account, and a store of value. Large price movements alter the purchasing power causing risk and costs to firms and consumers. Given that BitCoin is a relatively fresh currency, its price formation is not well understood yet, as there are only few studies on BitCoin price formation available in the literature.

In order to better understand reasons for such enormously high price volatility, in the 2nd part of the paper we attempt to identify drivers of BitCoin price and estimate their importance econometrically. The previous literature suggests three types of drivers determining BitCoin price development: (1) market coerces of BitCoin supply and request, (Two) BitCoin appeal, and (Trio) global macroeconomic and financial developments. We apply time-series analytical mechanisms to these drivers using daily data for the 2009–2014 period.

Our empirical results confirm that market coerces of BitCoin supply and request have an influence on BitCoin price, implying that, to a certain extent, the formation of BitCoin price can be explained in a standard economic model of currency price formation. In particular, the demand-side drivers, such as the size of the BitCoin economy and the velocity of BitCoin circulation, were found to influence BitCoin price. Given that BitCoin supply is exogenous, likely, the development of the request side drivers will be among the key determinants of BitCoin price also in the future potentially lading to deflationary pressures if BitCoin use expands.

However, the influence of demand-side drivers on the BitCoin price somehow reduces when controlling for BitCoin appeal, which implies that these drivers show up to be relatively more significant. In fact, we cannot reject the hypothesis extensively emphasized in the literature that investor speculations are also affecting BitCoin price. The statistically significant influence of Wikipedia views on BitCoin price could be an indicator of speculative short-run behavior of investors, or it may capture the expansion of the request as a medium of exchange of the BitCoin. Additionally, we find that also fresh information influence BitCoin price positively, which may be a result of an enlargened trust among users. As such, speculative trading of BitCoins is not necessarily an undesirable activity, as it may generate benefits in terms of absorbing excess risk from risk reluctant participants and providing liquidity on the BitCoin market. A negative side of the short-run speculative investment is that it may increase price volatility and create price bubbles which has adverse implication for BitCoin users. A further negative side of the long-run speculative investment is more extensive hoarding of BitCoins which may reduce its use in exchanges. The success of BitCoin thus also hitches on its capability to reduce the potential negative implications of such speculations and expand the use of BitCoin in trade and commerce.

Eventually, our econometric estimates do not support previous findings that the global macro-financial development may be driving BitCoin price. In fact, once we control for supply–demand variables and BitCoin’s appeal for investors linked to BitCoin, the influence of global macro-financial development captured by the Dow Jones Index, exchange rate and oil price becomes statistically insignificant. Because macroeconomic switches are not reflected in BitCoin price movements its price volatility cannot be lightly hedged. This is in contrast to standard currencies which are powerfully driven by macro developments and hedging option is available and widely used.

In summary, our investigate has shown that there is a disagreement in the literature on whether BitCoin can become a global currency. Overall, negative assessments about BitCoin as a currency tend to prevail given that several BitCoin characteristics identified in the paper impede its use as a currency and thus constrain its expansion globally. However, as outlined in the paper some BitCoin characteristics give its predisposition to be adopted at least in some segment of money market if not in a broader context. In particular, BitCoin may have a high relative comparative advantage with respect to standard currencies in countries with unstable financial system (e.g. in developing countries), and may provide an alternative to standard currencies in countries with poor and not widely available financial services, non-convertible currency, expensive financial services and high administrative cargo in opening an account. In addition, BitCoin may represent a cost-effective remittance system in developing countries, were traditional transfers are very expensive and the banking system is underdeveloped and unsecure. Given that BitCoin transfers can be done with relatively minimal cost and resource requirements and are independent of geographical location or banking system in place, they are ideally placed to serve as an efficient international remittance system. Ten

The BitCoin technology—BlockChain eleven —opens up several other possibilities and technological innovations, including micropayments, crowdfunding, distributed exchanges, brainy property, property registry, ticketing and secure voting systems. For example, BlockChain technology could provide a way to track the unique history of individual devices, by recording a ledger of data exchanges inbetween it and other devices, web services, and human users. Similarly, BlockChain could enable clever devices to become independent agents, autonomously conducting a multitude of transactions. For example, a vending machine could not only monitor and report its own stock, but also solicit bids from distributors and pay for the delivery of fresh items automatically—based on the purchase history of its customers. Further, the disruptive innovation of BitCoin provides the potential to give citizens direct control over their financial activities by removing costly—and sometimes obscure—intermediation layers fostering financial inclusion .

Footnotes

Note that virtual currencies must be distinguished from electronic money. The key distinguishing feature of electronic money is that their link with traditional money is preserved and have the same unit of account as well as they have legal foundation and are regulated. This is not the case of virtual currencies including BitCoin (ECB two thousand twelve ).

Given that freshly issued BitCoins decrease over time, miners will have to rely more on transaction fees to recoup their investment in mining which may lead to higher transaction fees in the future (EBA two thousand fourteen ). For a theoretical analysis of the economics of BitCoin transaction fees see Kroll et al. ( two thousand thirteen ) and Houy ( two thousand fourteen ).

There was approximately $1.37 trillion in circulation as of June, two thousand fifteen (Federal Reserve System 2015).

Access to the BitCoin network requires downloading a BitCoin software on individual computer and joining the BitCoin network, which permits users to engage in operations, and update and verify transactions.

A related problem pointed by Yermack ( two thousand fourteen ) is linked to relatively high diversity of BitCoin prices across different exchanges and web quotations at any given time. This variation in prices poses problem to establish a valid reference point for price setting for both consumers and businesses.

Moreover, BitCoins which were accidentally lost or demolished can never be substituted, resulting in shrinkage of the money base and leading to enhanced deflationary trend. For example, the bitomat exchange, one of the largest exchanges, lost around $200 thousand worth of BitCoins (at the exchange rate at the time) due to human error by accidentally erasing cloud server where the wallet was stored (Barber two thousand twelve ).

Given that people consider a currency valuable if they expect others to do so, for a decentralized currency, such as BitCoin, trust depends on a belief that the rules of the currency will be stable over time.

The Dow Jones Index is an industrial average that captures thirty major corporations on either the NYSE or the NASDAQ.

An example of the BitCoin based system for remittance transfers is BitPesa. BitPesa is an online payment platform that uses BitCoin to suggest money transfers to and from East Africa (Folkinshteyn et al. Two thousand fifteen ).

BlockChain is a public ledger of all BitCoin transactions that have ever been executed; it is permanently growing as ‘completed’ blocks are added to it with a fresh set of recordings.

Notes

Acknowledgments

The authors are grateful to Tony Tam for providing access to the BitCoin data of Bitcoinpulse. We gratefully acknowledge financial support from the research project VEGA1/0797/16. The authors are solely responsible for the content of the paper. The views voiced are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission.

The digital agenda of virtual currencies: Can BitCoin become a global currency, SpringerLink

The digital agenda of virtual currencies: Can BitCoin become a global currency?

  • Pavel Ciaian Email author
  • Miroslava Rajcaniova
  • d’Artis Kancs

Abstract

This paper identifies and analyzes BitCoin features which may facilitate BitCoin to become a global currency, as well as characteristics which may impede the use of BitCoin as a medium of exchange, a unit of account and a store of value, and compares BitCoin with standard currencies with respect to the main functions of money. Among all analyzed BitCoin features, the extreme price volatility stands out most clearly compared to standard currencies. In order to understand the reasons for such extreme price volatility, we attempt to identify drivers of BitCoin price formation and estimate their importance econometrically. We apply time-series analytical mechanisms to daily data for the 2009–2014 period. Our estimation results suggest that BitCoin appeal indicators are the strongest drivers of BitCoin price followed by market compels. In contrast, macro-financial developments do not determine BitCoin price in the long-run. Our findings suggest that as long as BitCoin price will be mainly driven by speculative investments, BitCoin will not be able to contest with standard currencies.

Keywords

1 Introduction

A broad range of virtual currencies have emerged during the last decade, such as BitCoin, LiteCoin, PeerCoin, AuroraCoin, DogeCoin and Ripple. The most successful among them is BitCoin, both in terms of its extraordinaire growth in the number of currency users and popularity by retailers. Since its introduction in 2009, BitCoin has been characterized also by a phenomenal increase in the number of transactions and market capitalization, which surpassed five billion US dollar in two thousand fifteen and since then has recorded further growth.

In the public media and in scientific community there is ongoing a upbeat debate on wheather BitCoin can actually function as a substitute for standard currencies such as US dollar, Euro or Yen. There are two contesting views in the literature on whether BitCoin fulfils the three criteria of a currency (a medium of exchange, a unit of account, and a store of value) (Mankiw two thousand seven ). One part of the literature argues that BitCoin does not behave as a real currency, but rather resembles speculative investments (e.g. Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ; Williams two thousand fourteen ). Other line of research, stresses the positive aspects of BitCoin and perceives it as a global virtual currency with strong future potential (e.g. Plassaras two thousand thirteen ; Satran two thousand thirteen ; Luther and White two thousand fourteen ; Folkinshteyn et al. Two thousand fifteen ). Also from a legal point of view, there is no global agreement on the status of BitCoin, as there are no international laws regulating BitCoin. Each country regards BitCoin differently and regulations are permanently evolving. For example, the Internal Revenue Service in the US has assigned BitCoin as barter on the grounds of the market-orientated treatment. Finland considers BitCoin as priced commodity. Germany has recognized it as private currency. Yet other EU Member States by and large are in the wait-and-see stage. On the other mitt, institutions such as European Central Bank (ECB) and European Banking Authority (EBA) define BitCoin as virtual currency.

In light of these and other open questions about BitCoin, in April two thousand fifteen the European Commission organized a Virtual Currencies Conference, one where among others the potential and challenges of BitCoin to become a global currency were discussed. The Virtual Currencies Conference acknowledged the need for further research to better understand the socio-economic and monetary aspects of the virtual currency ecosystem.

In order to shed light on these and related questions, the present paper attempts to identify and analyze BitCoin features which may facilitate BitCoin to become a global currency, as well as characteristics which may impede the use of BitCoin as a medium of exchange, a unit of account, or a store of value. Among the BitCoin features, which may facilitate its use as a currency, we have identified low transaction costs, high anonymity and privacy, learning spillover effects, infinite divisibility and no inflationary pressures. Among the BitCoin features, which may impede its use as a currency include the absence of a legal tender attribute, difficulty to procure BitCoins, relatively high immovable costs of adoption, dependence on network externalities, absence of an institution enforcing dispute resolution, absence of BitCoin denominated credits, deflationary pressure, utterly high price volatility, and issues with cyber security.

Our 2nd contribution to the literature is to undertake a comparative analysis of the identified BitCoin features with traditional currencies and their possible influence on BitCoin functions as a currency. Our analysis builds on previous studies, which have identified several advantages of BitCoin over traditional currencies. Very first, as noted by EPRS ( two thousand fourteen ), EBA ( two thousand fourteen ), and Folkinshteyn et al. ( two thousand fifteen ), BitCoin has no physical representation, such as paper bills or metal coins, which saves costs related to the production, transportation, and treating of physical currency. These costs can be substantial for standard currencies. Similarly, an advantage of BitCoin is that it is more convenient than standard payment mechanisms for small-value purchases, as it permits for money transfers at low costs and relatively prompt (Hayes et al. One thousand nine hundred ninety six ). As noted by Folkinshteyn et al. ( two thousand fifteen ), the advantages of low transaction fees and brief execution time make BitCoin an ideal medium of exchange. In contrast, Grinberg ( two thousand eleven ) opposes this view and argues that the response of the traditional e-commerce sector to the competitive pressure from BitCoin may induce them to reduce their transaction costs and thus offset the advantage of BitCoin, while providing also other benefits such as higher security. 2nd, an extra advantage of BitCoin, compared to standard currencies, is that it may reduce opportunities for theft, such as bank robbery. BitCoin could help curtail vandalism of vending machines, public phones, etc., because there would be no cash to steal. Similarly, businesses who treat cash, such as taxi drivers and puny shops, could be much less vulnerable to robbery, if they would use BitCoin. Ultimately, BitCoin is a global currency, implying that there are no transaction costs related to currency exchange.

On the other palm, BitCoin faces several challenges compared to standard currencies. One of the main challenges of BitCoin is its security (Moore and Christin two thousand thirteen ; Böhme et al. Two thousand fifteen ; Yermack two thousand fourteen ). Given that BitCoin transactions take place exclusively over the internet, cyber-security is its main threat. Particularly vulnerable to cyber-attacks are large holdings of BitCoins as well as BitCoin exchanges. The security problem is largely attributed to the lack of an oversight institution that would ensure security of BitCoin transactions and BitCoin system (ECB two thousand twelve ; Plassaras two thousand thirteen ; Moore and Christin two thousand thirteen ). 2nd, the use of a fresh currency, such as BitCoin, requires an initial investment for both businesses and consumers (Velde two thousand thirteen ). This includes costs linked to getting acquainted with BitCoin system in general and to the adoption of the payment technology in particular. For example, BitCoin requires the use of electronic devices including specific software. Similarly, BitCoin may also suffer from information asymmetry, as its system is relatively sophisticated and therefore may not be lightly understood by all potential users (ECB two thousand twelve ). Third, Yermack ( two thousand fourteen ) argues that BitCoin may fail to become a global medium of exchange, as it is used in too few exchanges of goods and services; it has a negligible market presence globally. Presently, there are around 0.11 million BitCoin transactions executed per day (BlockChain two thousand fifteen ) and most of these BitCoin transactions involve transfers inbetween speculative investors. BitCoin use for purchases of goods and services is only minor—20 % of all BitCoin transactions (Yermack two thousand fourteen ). For example, if compared to two hundred ninety five million conventional payments and terminal transactions per day done in Europe alone, the market share of BitCoin shows up to be minuscule (EBA two thousand fourteen ). Fourth, Yermack ( two thousand fourteen ) argues that BitCoin is ineffective as a device of risk management of price volatility as it is uncorrelated with major world currencies (i.e. US dollar, Euro, Yen, British Pound) or gold price, which is a common use of currencies by businesses for hedging the risks associated with currency volatility. Macroeconomic switches that cause adjustment in exchange rates of currencies are not reflected in any way in BitCoin price movements implying that its price volatility cannot be lightly hedged. Eventually, BitCoin faces the problem of network externalities in its adoption (Gowrisankaran and Stavins one thousand nine hundred ninety nine ), as its benefit in exchange is positively correlated with the number of users (Plassaras two thousand thirteen ; ECB two thousand twelve ).

BitCoin and Euro price development in USD, 2009–2011. Source: BlockChain and Oanda

In order to better understand the reasons for such utterly high price volatility, we attempt to identify drivers of BitCoin price and estimate their importance econometrically. This is our third contribution to the literature. Previous studies (e.g. Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ) suggest three types of drivers determining BitCoin price development: (1) market compels of BitCoin supply and request (Two) BitCoin appeal, and (Three) global macroeconomic and financial developments. We apply time-series analytical mechanisms to these drivers using daily data for the 2009–2014 period. Our estimation results suggest that BitCoin appeal variables are most significant drivers of BitCoin price followed by market compels. Our estimates do not support previous findings that macro-financial developments are driving BitCoin price in the long-run.

Generally, our findings suggest that as long as BitCoin price will be mainly driven by speculative investments, BitCoin will not be able to challenge with standard currencies.

The rest of the paper is structured as goes after. Section two provides background information about BitCoin, which is a relatively fresh virtual currency. In the context of the three currency criteria, the key features are discussed and contrasted with standard currencies to analyze whether BitCoin fulfills the main functions of money. Section three identifies the main factors affecting BitCoin price based on previous studies; outlines the econometric approach—a Vector Auto Regressive (VAR) model—and specifies the estimated model. It also details the data sources used in the empirical analysis, the construction of the estimable model’s variables, and discusses the estimation results. The final section concludes.

Two Is BitCoin a currency?

Two.1 Background of BitCoin

Type of currencies

Money (currency) format

Certain types of local currencies

Banknotes and coins

Commercial bank money (deposits)

The most popular virtual currency is BitCoin, created by the Japanese programmer Satoshi Nakamoto in two thousand nine (Nakamoto two thousand nine ). It was the very first open source virtual currency, as BitCoin is managed by an open source software algorithm that uses the global internet network both to create BitCoins as well as to record and verify its transactions. Being a cryptocurrency, BitCoin uses the principles of cryptography to control the creation and exchange of BitCoins. BitCoins can be stored in local wallets (e.g. individual computer, smartphone) using an open-source software or in an online wallet (Brito and Castillo two thousand thirteen ; Murphy two thousand thirteen ; CoinDesk 2015b ).

Compared to standard fiat currencies, such as US dollars or Euro, a distinguishing feature of BitCoin is that the quantity of units in circulation is not managed by a person, group, company, central authority, or government, but by a software algorithm. BitCoins are created in a ‘mining’ process, in which computer network participants, i.e. users who provide their computing power, verify and record payments into a public ledger called BlockChain. In comeback for this service they receive transaction fees and freshly minted BitCoins. A stationary amount of BitCoins is issued at a constant a priori defined and publicly known rate, according to which the stock of BitCoins increases at a decreasing rate. In two thousand one hundred forty the growth rate of BitCoin will converge to zero, when the maximum amount of BitCoins in circulation will reach twenty one million units; according to the current algorithm it will not switch after 2140.

BitCoin does not have physical representation. Instead, it is stored either on electronic devices (e.g. private computer, mobile, tablet) or entrusted to an online service and is transferred via the internet. BitCoins can be spent on both goods and services, if accepted by the retailer. Users interact with each other directly and anonymously, without third-party intervention. There is neither a central clearing house nor any other intermediary institution involved in the transactions such as central bank or government agency (Table one ). Presently, BitCoins can be acquired either (1) by exchanging them for standard money (e.g. US dollar, Euro) on a BitCoin exchange or from a BitCoin dealer, (Two) by obtaining them from sales of goods or services denominated in BitCoin, or (Trio) through a mining process (Plassaras two thousand thirteen ; CoinDesk 2015c ).

BitCoins can be used to buy goods or services worldwide, provided that the transaction fucking partner accepts BitCoin as a mean of payment. A transaction implies that the present holder of BitCoins transfers the ownership of a certain amount of BitCoins to a different market participant in exchange for other currencies, goods or services. A continuously growing number of companies accept BitCoins as payments for their goods and services, at the beginning of two thousand fifteen there were more than 100,000 venues accepting BitCoins (CoinDesk 2015a ; Cuthbertson two thousand fifteen ).

There are two contesting views in the literature on whether BitCoin fulfils the three criteria of a currency (a medium of exchange, a unit of account, and a store of value). One part of the literature argues that BitCoin does not behave largely like a real currency, as it does not fulfil the main functions of a currency, but rather serves as a vehicle for speculative investments (Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ; Williams two thousand fourteen ). Other line of research stresses the potential of BitCoin and perceives it as a global currency with strong future potential (e.g. Plassaras two thousand thirteen ; Satran two thousand thirteen ; Luther and White two thousand fourteen ; Folkinshteyn et al. Two thousand fifteen ). In the following we attempt to identify and analyze the most significant advantages and disadvantages of BitCoin in the context of the three currency functions.

Two.Two Medium of exchange

Currency characteristics of BitCoin with respect to standard currencies

Advantage of BitCoin

Disadvantage of BitCoin

Medium of exchange

Anonymity and privacy

Not legal tender and difficulty to procure BitCoins

Learning spillover effect

Stationary costs of adoption

Dispute resolution not available

Absence of BitCoin denominated credit

Unit of account

Relative price comparability problem

Source Authors’ presentation

Two.Two.1 Transaction costs

An appealing feature of BitCoin when used as a medium of exchange is its comparative advantage in transaction costs relative to standard currencies, as BitCoin transaction fees are considerably lower than comparable costs of traditional means of payment (e.g. payment cards or bank transfers). BitCoin transaction fees cover only the costs of maintaining the system (i.e. clearing system) paid to miners; there are no costs linked to third-party intermediaries, such as a regulatory authority, that would perform validation, storage or security functions; the BitCoin system (maintained by miners) is financed through transaction fees and freshly issued BitCoins. Three Transaction costs of standard currencies are considerably higher as they need to cover all costs of intermediaries. Besides the costs of clearing system, the fees charged for standard money transfers need to also offset the cost of storage, authentication, transport, security, etc. (Šurda two thousand twelve ; EBA two thousand fourteen ). Average transaction fees per transaction for BitCoin transfers are inbetween zero and one %, while traditional online payment systems charge fees inbetween two and five % or even more (EPRS two thousand fourteen ; EBA two thousand fourteen ; Folkinshteyn, Lennon and Reilly two thousand fifteen ). Estimates for the US dollar display that these costs (costs of processing and accounting of money, storage, transport, and security) account to as much as $60 billion annually (ca. Four.Four % of the US dollar value). Four Further, BitCoin offers quicker transaction execution than traditional online payment systems. The total processing time of BitCoin transfers is inbetween ten and sixty min (EBA two thousand fourteen ).

We may conclude that comparably low transaction and maintenance costs together with rapid execution of transactions may facilitate the acceptance and use of BitCoin as a currency (Table two ).

Two.Two.Two Anonymity and transparency

An area, where BitCoin has a relatively high appeal, involves exchanges linked to illegal and criminal activities (e.g. money laundering; narcotic trade, tax evasion). While this was not the intent of BitCoin creators, it is rather the way individuals choose to use it, as certain BitCoin characteristics give its predisposition to be advantageous in such activities. For example, BitCoin payment transactions are anonymous and do not require the provision of individual identity information. Further, BitCoin permits for effortless international transfers (e.g. money laundering) without supervision, as BitCoin transactions are peer-to-peer and require just internet access. With its infrastructure being spread across the globe, it is difficult to intercept individual transactions (Bryans two thousand fourteen ; EBA two thousand fourteen ). These features grant BitCoin a rather high predisposition to be used in illegal activities. For example, the internet portal Silk Road, created in January 2011, provided online marketplace for the sale of illegal narcotics and weapons using BitCoins for payment, accounting for as much as half of all BitCoin transactions. Silk Road was shut down following an FBI investigation in 2013. Often it is argued that this event generated benefit to BitCoin by boosting its popularity (EBA two thousand fourteen ; Yermack two thousand fourteen ).

Gambling is other area where BitCoin shows up to be growing as a medium of exchange, as it protects costumer privacy and permits receiving funds from consumers not being able to use other payment methods. The most popular BitCoin denominated online gambling site is Satoshi Dice operating since two thousand twelve (Böhme et al. Two thousand fifteen ). Generally, there is a growing number of online gambling sites, the BitCoin Wiki ( two thousand fifteen ) lists around one hundred BitCoin based casinos, poker sites, bingo games, betting services and lotteries.

The BitCoin platform is semi-transparent and public, meaning that anyone is able to go after the chain of transaction. All BitCoin payments have a traceable history that can be viewed by anyone. A single anonymity breach can uncover a user’s entire BitCoin transaction history. For example, the skill of the identity of any user from any transaction (e.g. obtained from the mailing address used for delivery of purchased goods with BitCoin, or from the bank account used to purchase BitCoin) permit to track that user’s transactions backward and forward through the BlockChain history. Hence, BitCoin system is often referred as pseudonymous in the sense that total history of all transactions and every BitCoin is preserved on the publicly ledger. However, several options are available to overcome this anonymity problem and make transactions non-traceable, for example, by using fresh addresses for each payment received, using switch addresses when sending payments, or using BitCoin mixer services to break the link inbetween an user and its BitCoins (Bitcoinhelp two thousand fourteen ; Crawford two thousand fourteen ; BitCoin two thousand fifteen ; Böhme et al. Two thousand fifteen ).

We may conclude that relatively high anonymity of BitCoin users and transparency of BitCoin transactions may facilitate the popularity and use of BitCoin as a currency (Table two ).

Two.Two.Three Legal tender

One distinguishing feature of BitCoin is that it is not a legal tender (as opposed to standard currencies which would imply its mandatory acceptance in exchanges). Private or public businesses are not legally obliged to accept BitCoin as a payment form for goods and services they trade. The use of BitCoin as a medium of exchange is thus fully dependent on its voluntary adoption by market participants (EBA two thousand fourteen ).

According to Yermack ( two thousand fourteen ), an extra obstacle for BitCoin to become a widely used medium of exchange, arises from the difficulty of procuring fresh BitCoins. BitCoins can be obtained only from online exchanges or dealers (except for successful BitCoin miners).

We may conclude that, generally, the absence of legal tender may impede BitCoin use as a currency, because any business can determine individually on the acceptance/not acceptance of BitCoin (Table two ).

Two.Two.Four Stationary costs

The use of a fresh currency, such as BitCoin, requires an initial investment for both businesses and consumers. This includes costs linked to getting acquainted with the BitCoin system in general and to the adoption of the payment technology in particular. For example, BitCoin requires the use of electronic devices including specific software. Five

In addition, a relatively high level of computer skill is required for understanding and using BitCoins, which may represent a barrier to many potential users and may constraint its broader adoption (i.e. information asymmetry problem). BitCoin is based on a elaborate code, which is understood by only a few persons and is managed by even fewer individuals (programmers), who oversee the entire system without accountability, arbitration or recourse (Velde two thousand thirteen ; Yermack two thousand fourteen ).

On the other mitt, BitCoin use may increase users’ interaction with virtual currencies and thus may have learning spillover effects by helping to improve their abilities and skill in virtual transactions and online financial applications (Hayes et al. One thousand nine hundred ninety six ; Berentsten one thousand nine hundred ninety eight ; Plassaras two thousand thirteen ).

We may conclude that in presence of high immobile costs, the popularity and use of BitCoin as a currency may be gravely impeded at least in the short-run, as presently the technical skill about BitCoin is still rather low in society. However, BitCoin may generate some extra benefits to its users through learning spillover effects (Table two ).

Two.Two.Five Network externalities

The incentives for market participants to use BitCoin depend on the number of existing users. That is, the benefit of making the investment and using it in exchanges depends on the number of other suppliers and consumers of goods and services using BitCoin. If only a few businesses accept BitCoins, there are little incentives for consumers to acquire them. On the other forearm, if only a few consumers use BitCoins, businesses have little incentives to invest into the equipment for processing BitCoin payments for their goods and services. This is a well-known problem of network externalities (Gowrisankaran and Stavins one thousand nine hundred ninety nine ). Hence, one of the main BitCoin challenges in becoming a global currency is to coax users to use it in their purchases and businesses to accept it as payment form for their goods and services (Berentsten one thousand nine hundred ninety eight ; ECB two thousand twelve ; Plassaras two thousand thirteen ).

Yermack ( two thousand fourteen ) argues that BitCoin largely fails to be a global medium of exchange as it is used to a limited extent to intermediate the exchange of goods and services. Presently there are around 0.11 million BitCoin transactions executed per day (BlockChain two thousand fifteen ) which is insignificant compared, for example, to two hundred ninety five million conventional payments and terminal transactions per day done in Europe alone (EBA two thousand fourteen ). According to BlockChain ( two thousand fifteen ), the use of BitCoin as a medium of exchange is continuously enlargening over time as more and more businesses tend to accept it as a form of payment. Hence, despite the rather low BitCoin invasion in the global exchanges presently may not be an obstacle for BitCoin to become a global medium of exchange in future. Very first, given that BitCoin is a relatively fresh currency, it may take some time till it gets adopted in a broader context. 2nd, there exist examples of puny countries with own currencies (e.g. Fiji) which intermediate the exchange of goods and services within the national boundaries but have only a minor share globally.

We may conclude that the issue of network externalities may impede growth of BitCoin use as a currency in the short-run. However, these issues may become less relevant in future, as BitCoin use continuously increases (Table two ).

Two.Two.6 Dispute resolution

The use of BitCoin in market exchanges bears a certain risk, because of the absence of any protection against disputes inbetween parties involved in the exchange. Once a BitCoin transaction is realized, it is irreversible and cannot be disputed. There is no centralized mechanism available to revert an erroneous transaction or to treat the disputes with the aim to provide protection against human errors of fraud that may occur in exchanges (e.g. protection against disputes over non-fulfillment of contract). Presently, the correction of an erroneous transaction is possible only through a voluntary agreement of the parties involved in the exchange (EBA two thousand fourteen ; Böhme et al. Two thousand fifteen ).

We may conclude that the absence of an institution regulating and enforcing BitCoin related disputes, the popularity and use of BitCoin as a currency may be impeded, particularly for risk-averse market participants (Table two ).

Two.Two.7 Credit market

Another obstacle for BitCoin to become a widely used medium of exchange may be that BitCoin cannot be used to take loans because, under the current system, every loan would need to be made in BitCoin. Whereas the basis of standard banking system is fractional-reserve, which determine how much fresh money can be created through loans, BitCoins are unique and cannot be duplicated, they only exist as an electronic analog a kind of physical coin. This boundaries the expansion of BitCoin (Hanley two thousand fourteen ). For example, the absence of BitCoin denominated credit cards and consumer loans do not permit purchases on credit—a method widely used in most developed retail markets, which thresholds the expansion of BitCoin as a medium of exchange (Yermack two thousand fourteen ).

We may conclude that in the absence of BitCoin denominated credit, BitCoin growth is gravely constrained, as the share of credit transactions is rather high (and growing) in modern societies (Table two ).

Two.Trio Unit of account

To serve the function of a unit of account, as any other currency, BitCoin should be able to measure the relative value of goods and services and other transactions (e.g. debts) in the economy. We identify two key characteristics of BitCoin as a unit of account, which differ substantially from traditional currencies: divisibility and price volatility (see Table two for summary).

Two.Trio.1 Divisibility

An significant distinguishing feature of BitCoin is its almost infinite divisibility, implying that prices may be quoted in four or more decimal places. Indeed, divisibility is a necessary characteristic of a currency to accommodate the valuation for all types and sizes of transactions.

On the other palm, the price differences in the magnitude of several decimal places (e.g. four or more) may be confusing to consumers and may pose problems to them to comprehend and compare relative prices of goods and services. For comparison most world currencies use no more than two decimal points for price quotations (Yermack two thousand fourteen ).

We may conclude that, generally, due to almost infinite divisibility possibilities of BitCoin its use and popularity may grow compared to standard currencies, albeit petite price denominations may reduce capability of consumers to accurately distinguish relative prices (Table two ).

Two.Three.Two Price volatility

BitCoin prices demonstrate enormously high short-run volatility, which diminishes its capability in signifying an effective unit of account (Yermack two thousand fourteen ). The frequent BitCoin price switches cause direct and indirect costs to businesses and consumers. Businesses that use BitCoin have to adjust prices frequently, otherwise they may realize decrease in comebacks (because of underpriced goods and services) or loss of competitiveness (because of overpriced goods and services). This is particularly problematic for businesses trading outputs in BitCoins, while paying for production factors and intermediate inputs in local standard currency (e.g. US dollar, Euro, Yen, British Pound), causing discrepancy in relative prices inbetween outputs and inputs in presence of high BitCoin price volatility. Frequent price switches in turn become confusing to consumers, as it becomes more difficult to spot the true relative prices of goods and services. Six

Albeit, high price volatility of BitCoin may inflict risk to its holders, entrepreneurial innovations provide alternative solutions which diminish the price volatility risk. For example, market exchange pricing may facilitate price setting to businesses (retailers) and widen spending options to consumers. The market exchange pricing enable retailers to set prices in one currency (e.g. US dollar, Euro, Yen, British Pound) while displaying them to consumers at the same time and automatically updated in more currencies, including in BitCoin, reflecting the current market exchange rates. This system makes relatively costless price tracking in BitCoin to businesses that reflect up-to-date market exchange rate particularly for online sales. Another example includes instantaneous exchange facilities which enable retailers to accept BitCoin as payment without actually receiving BitCoins. This system involves a third party which intermediates the exchange of BitCoins, paid by consumers, to standard currency, which is received by the retailer (e.g. US dollar, Euro, Yen, British Pound). Because sellers never actually receive BitCoins, they avoid the BitCoins exchange risk; the exchange risk is beard by the intermediaries which receive a fee in comeback (Luther and White two thousand fourteen ).

In the context of BitCoin use as a unit of account, the enormously high price volatility reduces BitCoin power to convey accurately the relative prices of goods and services in the economy (Table two ). If, as argued by Luther and White ( two thousand fourteen ), financial developments are providing alleviating options for addressing the issue of BitCoin price volatility, then the adverse impacts of price volatility may decline in future.

Two.Four Store of value

The value of the money must remain stable over time to permit their use in exchanges in different points of time. We identify two features of BitCoin as store of value, which differ substantially from traditional currencies: non-inflationary supply and cyber security (see Table two for summary).

Two.Four.1 Non-inflationary supply

Standard currencies are usually inflationary, meaning that their value reduces over time and thus diminishes the capability of the currency to function as a store of value. In contrast, an significant advantage of BitCoin is the protection against inflation as a safe haven from government interference. Given that under the current mechanism the future number of BitCoins is capped to a maximum amount of twenty one million units with no expansion possible beyond this amount, BitCoin will be exposed to deflationary pressures if its use as an investment alternative or for use as a medium of exchange increases. Seven On the one forearm, this is beneficial to BitCoin holders, as it will make them richer over time. On the other forearm, the expectation of higher future value of BitCoins may lead to its hoarding in anticipation of higher price which in turn may reduce its use in exchanges of goods and services. Meiklejohn et al. ( two thousand thirteen ) find that two thousand eleven was a break point for spending-versus hording of BitCoin. After this date the majority of BitCoins (more than sixty %) that were received were spent within less than one month period. The hording prior to two thousand eleven took many BitCoins out of circulation meaning that most of BitCoins from this period are in the “sink” addresses that have never been spent (Ron and Shamir two thousand thirteen ).

We may conclude that, in the absence of inflationary pressures, BitCoin popularity should increase compared to standard currencies, albeit deflationary pressure may act as a countervailing force (Table two ).

Two.Four.Two Cyber security

One of the main threats to BitCoin capability to preserve its value to the holders is the issue of cyber security. In the past, many BitCoin owners have lost their virtual money through online theft. Standard currencies give a possibility to protect it against theft either by physically hiding it (e.g. under mattresses, in a safe) or by depositing it in a bank. BitCoin is a virtual currency and thus it cannot be physically hidden. BitCoins must be held in computer accounts known as virtual wallets. The security of BitCoin wallets has often been a major problem. Even BitCoin exchanges were subject to cyber-attacks and thefts as they are common target of hackers. The most prominent example is the collapse of the largest BitCoin exchange, MtGox, in February two thousand fourteen allegedly leading to a loss of eight hundred fifty thousand BitCoins (Böhme et al. Two thousand fifteen ; Yermack two thousand fourteen ). Moore and Christin ( two thousand thirteen ) find that forty five % of the total studied BitCoin exchanges closed down, and forty six % of the exchanges that closed did not reimburse any claims to consumers.

In the context of BitCoin use as a store of value, BitCoin sensitivity to cyber-attacks and thefts reduces trust in BitCoin as a currency and hence impedes its growth and chances of becoming a global currency (Table two ).

Trio BitCoin price and its volatility

Among all BitCoin features identified and discussed above, eventually, price volatility is the one with the largest differences compared to the major world currencies, such as US dollar, Euro, Yen, British Pound. As shown in Fig. One , since its introduction in 2009, BitCoin price has shown utterly high volatility; it has enlargened from zero value at the time of its inception in two thousand nine to around $1100 at the end of two thousand thirteen when it reached the peak followed by a decline to around $225 (see Fig. One ). For comparison, the fluctuation of exchange rate inbetween US dollar and Euro has not exceeded a ±20 % strapped in the same six year period (see Fig. One ). Such market volatility with extreme price movements (±8000 %) makes BitCoin of little use as a unit of account.

One consequence of such extreme price volatility is that BitCoin may not succeed as a global currency to be accepted by suppliers and consumers, as it may fail to accurately convey relative prices of goods and services in the economy and may generate uncertainties to its holders due to its inability to preserve stable value over time. Given that BitCoin is a relatively fresh currency, its price formation is not well understood yet. In order to better understand reasons for such utterly high price volatility, in this section we attempt to identify the drivers of BitCoin price and estimate their relative importance econometrically.

Trio.1 Drivers of BitCoin price

The existing studies in the literature (e.g. Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ), suggest three types of drivers determining BitCoin price formation: (1) market coerces of BitCoin supply and request, (Two) BitCoin appeal, and (Three) global macroeconomic and financial developments.

Trio.1.1 Driver 1: market coerces of BitCoin supply and request

According to Buchholz et al. ( two thousand twelve ), one of the key drivers of BitCoin price is the interaction inbetween BitCoin supply and request on the BitCoin market. According to the quantity theory of money, the BitCoin supply is determined by the total stock of BitCoin in circulation. The BitCoin request is represented by the size of BitCoin economy (i.e. its use in exchanges) and the velocity of BitCoin circulation. The BitCoin velocity measures the frequency at which one unit of BitCoin is used for purchase of goods and services. The quantity theory implies that the price of BitCoin decreases with the velocity and the stock of BitCoins, but increases with the size of BitCoin economy and the general price level.

The request for BitCoin is primarily driven by its value as a medium of exchange. BitCoin does not have intrinsic value like commodity currency such as gold standard. For example, the key difference inbetween the gold standard and BitCoin is that the request for BitCoin is driven solely by its value in future exchange, whereas the request for commodity currency is driven by both its intrinsic value and its value in future exchange. The BitCoin supply is given by the total stock of units put in circulation, which is publicly known and is immobile in the long-run. Whereas BitCoin supply is exogenous, the supply of gold is endogenous, as it responds to switches in production technology (e.g. mining technology for gold) and comebacks. Given the exogeneity of BitCoin supply, the primary driver of its price developments is expected to be the request side shocks. By altering expectations of future use in exchanges, such shocks to request have the potential to produce large swings in the BitCoin price (Luther and White two thousand fourteen ).

Trio.1.Two Driver Two: BitCoin attraction

There are several BitCoin-specific factors which, in addition to traditional currency price determinants, such as market supply and request, determine its request. This is partially linked to the fact that BitCoin has been created relatively recently and to the nature of the currency. (Barber et al. Two thousand twelve ; Buchholz et al. Two thousand twelve ; Kristoufek two thousand thirteen ; van Wijk two thousand thirteen ; Bouoiyour and Selmi two thousand fifteen ).

Very first, BitCoin price may be affected by the risk and uncertainty of the entire BitCoin system. Given that BitCoin is a fiat currency and thus intrinsically worthless, it does not have an underlying value derived from consumption or its use in production process (such as gold). The value of a fiat currency is based on trust that it will be valuable and accepted as a medium of exchange also in future (Greco two thousand one ; Kovenock and Vries two thousand two ). Eight The expectations about trust and acceptance are particularly relevant for BitCoin which, being a relatively fresh currency, is in the phase of establishing its market share by building credibility among market participants.

2nd, being a virtual currency, BitCoin is more vulnerable to cyber-attacks than traditional currencies, which can destabilize the entire BitCoin system and eventually lead to a collapse of BitCoin. Such attacks have been frequently occurring in the BitCoin system in the past (Barber et al. Two thousand twelve ; Moore and Christin two thousand thirteen ). As mentioned above, Moore and Christin ( two thousand thirteen ) examined forty BitCoin exchanges and found that eighteen have been closed down after cyber-attacks, while the world’s largest BitCoin exchange, collapsed in two thousand fourteen due to a cyber-attack.

Third, BitCoin price may be affected by its appeal as an investment chance for potential investors. According to Gervais et al. ( two thousand one ), Grullon et al. ( two thousand four ) and Barber and Odean ( two thousand eight ), potential investors’ decisions may be affected by an increase or decrease of attention in the news media. The role of information is particularly significant in the presence of many alternative investment choices and positive search costs. Given that investment request depends on the costs associated with searching for information for potential investment opportunities available on the market, such as, the stock exchange, those investment opportunities which are under a particular attention of news media may be preferred by potential investors, because they reduce search costs. In turn, enlargened investment request for BitCoin may exercise upward pressure on BitCoin price. Indeed, Lee ( two thousand fourteen ) finds such evidence for BitCoin, whereby the alteration of positive and negative news generated high price cycles. This implies that the attention-driven investment behavior can affect BitCoin price either positively or negatively, depending on the type of news that predominate in the media at a given point of time.

Trio.1.Three Driver Three: global macroeconomic and financial developments

Van Wijk ( two thousand thirteen ) stresses the role of global macroeconomic and financial development, captured by variables such as stock exchange indices, exchange rates, and oil prices measures in determining BitCoin price. The influence of macroeconomic and financial indicators on BitCoin price may work through several channels. For example, stock exchange indices may reflect general macroeconomic and financial developments of the global economy. Favorable macroeconomic and financial developments may stimulate the use of BitCoin in trade and exchanges and thus strengthen its request, which may have positive influence on BitCoin price.

Inflation and price indices are other type of indicators capturing significant macroeconomic and financial developments. According to Krugman and Obstfeld ( two thousand three ) and Palombizio and Morris ( two thousand twelve ), oil price is one of the main sources of request and cost pressures, and it provides an early indication of inflation development. Thus, when the price of oil signals potential switches in the general price level, this may lead to depreciation (or appreciation) of BitCoin price. Also the exchange rate may reflect inflation development and thus influence positively BitCoin price as indicated above.

According to Dimitrova ( two thousand five ), there could be also negative relation inbetween a currency’s price and macro financial indicators. A decline in the stock prices induces foreign investors to sell the financial assets they hold. This leads to a depreciation of the underlying currency, but may stimulate BitCoin price, if investors substitute investment in stocks for investment in BitCoin. Generally, investors’ come back on stock exchange may capture chance costs of investing in BitCoin. Hence, in this case the stock exchange indices are expected to be positively related to BitCoin price.

Trio.Two Econometric treatment

The very first four variables pt, yt, vt, and bt account for request and supply drivers of BitCoin price (driver 1). Following the quantity theory of money we expect that coefficients β1 and βTwo would be positive, whereas βTrio and βFour would be negative. In addition, given that BitCoin supply is largely predefined, the total stock of BitCoins in circulation, bt, is a semi- exogenous variable, and implying that the influence of coefficient βFour on BitCoin price should be petite and/or statistically not significant. Variable at captures the BitCoin appeal (driver Two). As discussed above, the coefficient βFive associated to this variable can be either negative or positive, for example, as both positive and negative news attract investors’ and users’ attention. Eventually, variable mt represents global macroeconomic and financial developments (driver Trio). According to the previous findings discussed above, we expect the sign of the coefficient β6 to be either positive or negative depending on the type of macroeconomic variable.

The econometric model (1) contains mutually interdependent variables—BitCoin price and its explanatory variables. The estimation of non-linear interdependencies among interdependent time series in presence of mutually correlated variables is subject to an issue of endogeneity (Lütkepohl and Krätzig two thousand four ). To circumvent the issue of endogeneity, we go after the general treatment in the literature to analyze the causality inbetween endogenous time-series and specify a multivariate Vector Auto Regressive (VAR) model (Lütkepohl and Krätzig two thousand four ).

According to Engle and Granger ( one thousand nine hundred eighty seven ), regressions of interdependent and non-stationary time series may lead to spurious results. In order to avoid spurious regression, it is significant to test the properties of the time series involved. Therefore, in the very first step, the stationarity of time series is determined, for which we use two unit root tests: the augmented Dickey–Fuller (ADF) test and the Phillips–Perron (PP) test. The number of lags that we use for each dependent variable is determined by the Akaike Information Criterion (AIC). If two individual time series are non-stationary, their combination may be stationary (Engle and Granger one thousand nine hundred eighty seven ). In this special case, the time series are considered to be cointegrated, implying that there exists a long-run equilibrium relationship inbetween them.

In the 2nd step, we employ the Johansen’s cointegration method to examine the long-term relationship inbetween the price series. The number of cointegrating vectors is determined by the maximum eigenvalue test and the trace test. Both tests use eigenvalues to compute the associated test statistics. We go after the Pantula principle (Pantula one thousand nine hundred eighty nine ) to determine whether a time trend and a constant term should be included in the model.

In the third step, we estimate a vector error correction model for those series that are cointegrated (Johansen and Juselius’s one thousand nine hundred ninety ). It includes an error correction term indicating the speed of adjustment of any disequilibrium towards a long-term equilibrium state. In other words, the error correction term captures the switches in the BitCoin price required to eliminate the past deviation of the prices from the equilibrium levels.

As usual, in order to ensure the adequacy of the estimated models, we implement a series of specification tests: Lagrange-multiplier (LM) test for autocorrelation in the residuals; Jarque–Bera test to check if the residuals in the Vector-Error Correction (VEC) are normally distributed and a test of stability of the model.

Trio.Trio Data and variable construction

In order to construct the dependent variable, we use data for BitCoin price, p t B , denominated in US dollar (BitCoin price). We use the historical number of total BitCoins (number of BitCoins) which have been mined to account for the total stock of BitCoins in circulation, bt. We use two alternative proxies for the size of BitCoin economy, yt: the total number of unique BitCoin transactions per day (number of transactions), and the number of unique BitCoin addresses used per day (number of addresses). Following Matonis ( two thousand twelve ), we proxy the monetary velocity of BitCoin circulation, vt, by BitCoin days demolished for any given transaction (days ruined). This variable is calculated by taking the number of BitCoins in transaction and multiplying it by the number of days since those coins were last spent. All these data are extracted from quandl.com. To measure the price level of global economy, pt, we use exchange rate inbetween the US dollar and the Euro (exchange rate) extracted from the European Central Bank. We use exchange rate inbetween the US dollar and the Euro, because in our data BitCoin price is denominated in US dollar. For example, if the US dollar would appreciate against the Euro, most likely it will also appreciate against the BitCoin. Consequently, an increase in the exchange rate inbetween the Euro and the US dollar would lead to a decrease in the amount of US dollar that have to be paid for one BitCoin, which decreases its price.

In order to capture BitCoin attraction, at, we go after Kristoufek ( two thousand thirteen ) and use the volume of daily BitCoin views on Wikipedia (views on Wikipedia). According to Kristoufek ( two thousand thirteen ), the frequency of searches related to the virtual currency is a good measure of potential investors’ interest in the currency. However, the online search queries, such as Wikipedia views, may measure both investors’ and users’ interest in BitCoin, as it captures information’s request about the currency but it does not differentiate on whether the information is used to guide investment decisions or online BitCoin denominated exchange of goods and services. In addition, we also construct a variable capturing the number of fresh members (fresh members) and fresh posts on online BitCoin forums (fresh posts) extracted from bitcointalk.org. As explained above, the variable fresh members captures the size of the BitCoin economy but also attention-driven investment behavior of fresh BitCoin members. The variable fresh posts captures the effect of trust and/or uncertainty, as it represents the force of discussions among members.

To account for global macroeconomic and financial developments, mt, we go after van Wijk ( two thousand thirteen ) and use oil price (oil price) and the Dow Jones stock market index (Dow Jones). Nine Oil prices are extracted from the US Energy Information Administration, and Dow Jones index is extracted from the Federal Research Bank of St. Louis.

Trio.Four Estimation results

Specification of the empirically estimated models

Number of BitCoins

Number of transactions

Number of addresses

Views on Wikipedia

Short-run effects on BitCoin price for drivers 1, two and Three

L1D BitCoin price

L2D BitCoin price

L3D BitCoin price

L4D BitCoin price

L1D number of BitCoins

L2D number of BitCoins

L3D number of BitCoins

L4D number of BitCoins

L1D number of transactions

L2D number of transactions

L3D number of transactions

L4D number of transactions

L1D number of addresses

L2D number of addresses

L3D number of addresses

L4D number of addresses

L1D days ruined

L2D days ruined

L3D days ruined

L4D days demolished

L1D exchange rate

L2D exchange rate

L3D exchange rate

L4D exchange rate

L1D views on Wikipedia

L2D views on Wikipedia

L3D views on Wikipedia

L4D views on Wikipedia

L1D fresh members

L2D fresh members

L3D fresh members

L4D fresh members

*** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero. L1D indicates one lag, L2D indicates two lags, L3D indicates three lags, and L4D indicates four lags

Short-run effects on BitCoin price for general models

L1D BitCoin price

L2D BitCoin price

L3D BitCoin price

L4D BitCoin price

L1D number of BitCoins

L2D number of BitCoins

L3D number of BitCoins

L4D number of BitCoins

L1D number of transactions

L2D number of transactions

L3D number of transactions

L4D number of transactions

L1D number of addresses

L2D number of addresses

L3D number of addresses

L4D number of addresses

L1D days demolished

L2D days ruined

L3D days demolished

L4D days ruined

L1D exchange rate

L2D exchange rate

L3D exchange rate

L4D exchange rate

L1D views on Wikipedia

L2D views on Wikipedia

L3D views on Wikipedia

L4D views on Wikipedia

L1D fresh members

L2D fresh members

L3D fresh members

L4D fresh members

*** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero. L1D indicates one lag, L2D indicates two lags, L3D indicates three lags, and L4D indicates four lags

Long-run effects on BitCoin price for drivers 1, two and Trio

Number of BitCoins

Number of transactions

Number of addresses

views on Wikipedia

Dependent variable: BitCoin price. *** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero

Long-run effects on BitCoin price for general models

Number of BitCoins

Number of transactions

Number of addresses

Views on Wikipedia

Dependent variable: BitCoin price. *** Significant at one % level, ** Significant at five % level, * Significant at ten % level. “–” Indicates either absence of a variable in the respective model or the coefficient is not significantly different from zero

Tables six and seven showcase the long-run impacts of different determinants on BitCoin price. According to the results reported in Tables five and six , the long-run relationship inbetween BitCoin price and the explanatory variables considered in the estimated models is stronger than the short-run influence. In the following we discuss the long-run results with respect to the three drivers of BitCoin price formation.

Three.Four.1 Results for driver 1: market compels of BitCoin supply and request

The very first major observation arising from the estimates reported in Tables six and seven is that the market compels of supply and request have an influence on BitCoin price. Generally, the request side variables (e.g. days ruined, number of addresses) emerge to exert a more pronounced influence on BitCoin price than the supply side drivers (e.g. number of BitCoins). According to the results reported in Table six , an increase in the stock of BitCoins (number of BitCoins) leads to a decrease in BitCoin price (model 1.Two), whereas an increase in the size of the BitCoin economy (number of addresses) and its velocity (days demolished) lead to an increase in BitCoin price (models 1.1, 1.Two, 1.Trio, 1.Four, 1.Five, Four.Three, Four.Four, Four.7). These results are in line with our expectations. Contrary to our expectations, the alternative variable that captures the size of the BitCoin economy (number of transactions) has negative influence on BitCoin price in models 1.1 and 1.Five. However, this variable is not significant in the more general models (models Four.1–4.9), implying that its influence on BitCoin price is negligible.

Albeit, the sign of the estimated coefficients for market coerces of BitCoin supply and request is in line with our hypothesis (except for the variable number of transactions in models 1.1 and 1.Five), the statistical significance and magnitude of the estimated coefficients decreases in most models, when accounting for the influence of BitCoin attraction and global macroeconomic and financial developments (models Four.1–4.9 in Table seven ). The supply–demand variables are statistically significant in models Four.Trio, Four.Four and Four.7, but have a considerably lower magnitude of the estimated influence than in models 1.1–1.Five, which capture only market coerces of supply and request. This could be explained by the fact that part of the BitCoin price variation explained by the supply–demand variables is absorbed by other variables in more general specifications (models Four.1–4.9).

Trio.Four.Two Results for driver Two: BitCoin attraction

The strongest and statistically the most significant influence on BitCoin price is estimated for variables capturing the influence of BitCoin appeal: views on Wikipedia, fresh members and fresh posts (models Two.1 and models Four.1–4.9). Variable fresh members has negative influence on BitCoin price, implying that attention-driven investment behavior of fresh investors predominates. Variable fresh posts has positive influence on BitCoin price, reflecting an enhancing acceptance and trust in BitCoin captured by the energy of discussion inbetween BitCoin users. This may reflect declining transaction costs and uncertainty for investors, which increases investment request for BitCoin and hence its price.

Consistent with the findings of Kristoufek ( two thousand thirteen ), Wikipedia views have a statistically significant influence on BitCoin price. This variable is significant and has positive influence in all models (except for model Four.8). These results are in line with our expectations and with the remarkable Wikipedia article traffic statistics, according to which on average BitCoin article on Wikipedia is being viewed 200,000 times per month, and is ranked among the top one hundred most viewed articles. Only eighty eight English-language Wikipedia articles have more traffic than the BitCoin article, out of a total of approximately five million English-language articles. However, the interpretation of Wikipedia views is not straightforward, as it may capture various effects. On the one mitt, Wikipedia views may represent attention dedicated by potential users’ and/or investors’ to BitCoin likely induced by their reaction to media reporting on BitCoin. The attention effect may influence either positively or negatively the BitCoin price depending on the type of news. The positive estimated coefficient associated with Wikipedia views variable implies that the influence of positive news predominates. One of the most common negative news reported in media about BitCoin is related the security breach (cyber-attacks) against the currency. The positive estimated results indicate that the implication of security problem shows up to be offset by the positive news effect.

On the other forearm, Wikipedia views may measure investors’ or users’ interest in BitCoin, as it captures information’s request about the currency. It may reflect switches in the skill about BitCoin inbetween potential investors and/or users, thus leading to a higher acceptance and request for it either as an investment chance or as a medium of exchange. Significant is that the type of individuals searching information about BitCoin on Wikipedia likely are fresh BitCoin users/investors, because Wikipedia contains rather general information about BitCoin, which is known by incumbent investors or advanced BitCoin users. Both investors and users may search for information about BitCoin and thus the Wikipedia effect may capture either the users’ request for information with the aim to use it as a medium of exchange or may reflect investors’ information search who perceive BitCoin as an investment chance (i.e. for hoarding). Kristoufek ( two thousand thirteen ) argues that the second—the hording—effect predominates because, according to him, the market coerces of BitCoin supply and request permitting for setting a ‘‘fair’’ price are missing, rather its price is driven by the investors’ faith in the future growth and is predominated by short-term investors, trend chasers, noise traders and speculators. This argument is also in line with the sceptics on BitCoin (e.g. Velde two thousand thirteen ; Hanley two thousand fourteen ; Yermack two thousand fourteen ) who argue that hoarding is one of the key weaknesses of BitCoin as a currency (alongside the security problem) compared to standard currencies.

Three.Four.Trio Results for driver Three: global macroeconomic and financial developments

Our findings suggest that, in contrast to previous studies (i.e. van Wijk two thousand thirteen ), global macro-financial developments such as the Dow Jones Index, exchange rate and oil price do not significantly affect BitCoin price in the long-run. Only in Model Three.1 the macro and financial variables (Dow Jones, oil price and exchange rate) are statistically significant (Table six ). Our results suggest that an increase in oil price leads to a decrease in the budget of consumers (and companies), implying that less money will be spent on other goods, including BitCoin. Consequently, this would lead to a decrease in request for BitCoin, decreasing its price. These results are in line with the estimates of van Wijk ( two thousand thirteen ), who also finds statistically significant influence of macro-financial variables on BitCoin price. However, van Wijk ( two thousand thirteen ) does not account for market coerces of supply and request or BitCoin appeal indicators. When these factors are taken into consideration (models Four.1–4.9), their influence decreases considerably in all estimated models (except for model Four.1) (Table seven ). These finding support the argument of Yermack ( two thousand fourteen ) that BitCoin is relatively ineffective as a contraption for risk management against adverse market developments as its price is not responsive to macroeconomic variables meaning that it cannot be lightly hedged against other assets that are driven by macroeconomic developments.

Three.Five Comparison with previous studies

The verification of our results through comparison with previous studies is a challenging task, as (1) the currency is relatively fresh and there are not many studies in the literature, which analyze BitCoin price formation; and (Two) there are significant differences inbetween the used empirical proxies for variable construction and model specification across different studies. Despite these difficulties, in this section we attempt to compare our results with those reported in the literature. Generally, the explore of Bouoiyour and Selmi ( two thousand fifteen ) seems to be the most comparable probe to ours in terms of the employed estimation mechanisms, underlying data and explanatory variables. Similarly to our paper, they control for all three sets of drivers: market coerces, attraction indicators and macro variables. Other studies are less comparable to our treatment mainly due to differences in model specification, as most of them investigate the influence of each BitCoin price driver separately; they do not consider interactions inbetween them. For example, Kristoufek ( two thousand thirteen ) considers only BitCoin attraction indicators, whereas Van Wijk ( two thousand thirteen ) includes only the macroeconomic drivers. Buchholz et al. ( two thousand twelve ) account for the impacts of BitCoin supply and request, and BitCoin attraction, but does not account for global macroeconomic and financial developments.

The impacts of BitCoin supply and request (driver 1) on BitCoin price are measured by the ratio of exchange and trade transactions and velocity as (the frequency at which one unit of BitCoin is used to purchase goods) in Bouoiyour and Selmi ( two thousand fifteen ). Buchholz et al. ( two thousand twelve ) measure BitCoin request by price volatility. According to Bouoiyour and Selmi ( two thousand fifteen ), the estimated short-run influence of transactions on BitCoin price is positive and statistically significant (0.129, Table seven in Bouoiyour and Selmi). Compared to our not significantly different from zero estimates (−0.018 to 0.003, models 1.1 and 1.Five in Table four ) they are also smaller by around one order of magnitude. These differences may be caused by the fact that Bouoiyour and Selmi ( two thousand fifteen ) proxy for transactions using the ratio of exchange and trade transactions, whereas our proxy for transaction variable is directly constructed from the number of BitCoin transactions. Note also that our short-run estimates are not statistically different from zero. The estimated long-run influence of transactions is also positive and statistically significant at the ninety % level in Bouoiyour and Selmi ( two thousand fifteen ) (0.018, Table seven in Bouoiyour and Selmi). Remarkably, the long-run influence of transactions is considerably lower (by one order of magnitude) than the short-run influence. In contrast, our estimates suggest that the long-run effect of transactions is negative and considerably higher (by three orders of magnitude) (−3.99 and −3.42, models 1.1 and 1.Five in Table six ) than the short-run effect. Note that our estimates are significant at the ninety five and ninety nine % levels, respectively). Given that our specification is more general, we use longer time series and our results are more significant, a negative long-run relationship inbetween the number of transactions and BitCoin price is more likely.

According to Bouoiyour and Selmi ( two thousand fifteen ), the estimated short-run influence of BitCoin velocity is positive and considerably higher (Two.741, Table seven in Bouoiyour and Selmi) compared to our negative and rather petite estimates (−0.011 to −0.002, models 1.1–1.Four in Table four ). Note, however, that the coefficient of Bouoiyour and Selmi ( two thousand fifteen ) is not statistically significant, whereas it is statistically significant for all estimated models with one and two lags in our results. Again, these differences may be explained by differences in the velocity variable construction and model specifications. Whereas Bouoiyour and Selmi ( two thousand fifteen ) calculate velocity as the frequency at which one unit of BitCoin is used to purchase goods, we measure velocity by the number of days needed until a BitCoin is demolished. The estimated long-run influence of transactions is positive in Bouoiyour and Selmi ( two thousand fifteen ) as well (0.004, Table seven in Bouoiyour and Selmi). However, it is not statistically significant. As in the case of transactions, the long-run influence of velocity is considerably lower (by three orders of magnitude) than the short-run influence in Bouoiyour and Selmi. In contrast, our estimates suggest that the long-run effect of velocity is positive and considerably higher (by two orders of magnitude) (Five.07–11.71, models 1.1, 1.Trio, 1.Four and 1.Five in Table six ) than the short-run effect. Note that the magnitude of the estimated long-run coefficients is very stable across all for models, and they all are significant at the ninety nine % level. Given that our specification is more general, we use longer time series and our results are more significant, a positive long-run relationship inbetween velocity and BitCoin price is more likely.

Buchholz et al. ( two thousand twelve ) estimate the relationship inbetween BitCoin price and transactions, tho’ they are interested in the switch roles relationship, as they attempt to response the question how BitCoin price shocks affect BitCoin transactions. Hence, their results are not directly comparable to our estimates. Buchholz et al. ( two thousand twelve ) also estimate the relationship inbetween BitCoin request and BitCoin price, whereas BitCoin request is proxied by price volatility. They find that an enhancing price volatility leads to higher BitCoin price. The coefficient from the GARCH-in-mean model is statistically significant. Moreover, the results are different inbetween the two analyzed periods (August 2010–June two thousand eleven vs. June 2011–March 2012): in the very first period the volatility led to a request for BitCoins, and after the bubble burst in June two thousand twelve the novelty of BitCoin decreased, only market participants who were reluctant to volatility stayed in the market, leading to no effect of volatility on BitCoin price in the 2nd period. However, as above, their proxy for market demand—price volatility—is very different from our proxy—number of transactions—and hence cannot be directly compared with our estimates. Moreover, Buchholz et al. ( two thousand twelve ) do not consider all types of BitCoin price drivers and interactions inbetween them, and they use weekly data for a rather brief period (82 weeks), whereas our analysis is based on daily data for a Five.Five year period, capturing the extreme price switches in two thousand thirteen and two thousand fourteen (see Fig. One ).

The influence of BitCoin appeal (driver Two) on BitCoin price is measured by BitCoin searches in Google in Bouoiyour and Selmi ( two thousand fifteen ) and Buchholz et al. ( two thousand twelve ), whereas Kristoufek ( two thousand thirteen ) uses both BitCoin searches in Google and Wikipedia views.

The estimated short-run influence of searches in Google is positive and sizeable in Bouoiyour and Selmi ( two thousand fifteen ), when compared to our negative and considerably smaller estimates. These differences could be explained by differences in variable construction and model specifications. While Bouoiyour and Selmi ( two thousand fifteen ) use the number of BitCoin searches in Google to proxy for investors’ attraction, we use the number of views of BitCoin site on Wikipedia. The estimated long-run influence of searches in Google is positive and smaller than the short-run estimates (0.143 and 0.527, respectively, Table seven in Bouoiyour and Selmi) but statistically not significantly different from zero. Again, our long-run estimates (1.94, model Two.1 in Table six ) are of higher order of magnitude with respect to our short-run estimates. Note that our model specification is more general, and our estimates are significant at the ninety nine % level, which suggests that they may be more precise.

Kristoufek ( two thousand thirteen ) studies the influence of BitCoin attraction on its price by using BitCoin searches in Google and Wikipedia views. He finds that the enhanced interest in BitCoin measured by BitCoin searches in Google increases its price. As the interest in BitCoin increases, the request increases as well causing the prices to increase. The estimated positive influence is in line with our results. The relationship inbetween Wikipedia views and BitCoin turns out to be statistically not significant in Kristoufek ( two thousand thirteen ). In contrast, our long-run results suggest a positive and statistically significant influence of Wikipedia views on BitCoin price (1.94, model Two.1 in Table six ). The statistically significant results of Kristoufek ( two thousand thirteen ) may be caused by the fact that his analysis is based on a rather brief period (Two years) compared to a Five.Five year period in our examine as well as they do not control for other BitCoin price drivers, such as BitCoin supply and request and macroeconomic developments. Note that Kristoufek ( two thousand thirteen ) finds that the relationship is positive and statistically significant in the opposite direction. The estimated coefficient is 0.05, which implies that a ten % switch in BitCoin price is connected to a 0.Five % permanent shift in the Wikipedia views.

Also Buchholz et al. ( two thousand twelve ) use BitCoin searches in Google to estimate the influence on BitCoin. However, they estimate the relationship inbetween Google hits and the number of BitCoin transactions, not BitCoin price. Hence, their results are not directly comparable to our estimates.

The influence of global macroeconomic and financial developments (driver Three) on BitCoin price is measured by output volume, gold price and Chinese market index in Bouoiyour and Selmi ( two thousand fifteen ). van Wijk ( two thousand thirteen ) measures global macroeconomic and financial developments by the Dow Jones index, the value of the FTSE 100, the value of the Nikkei 225, the Euro—US dollar exchange rate, the Yen—US dollar exchange rate, and several oil price indices.

According to Bouoiyour and Selmi ( two thousand fifteen ), output volume and Chinese market index have positive and statistically significant influence on BitCoin price (0.079 and 0.380, respectively, Table seven in Bouoiyour and Selmi), whereas the influence of gold price is negative and statistically insignificant (−0.241, Table seven in Bouoiyour and Selmi). Our estimates are different, as they rely on different proxies capturing global macroeconomic and financial developments: Dow Jones index and oil price (−0.046 and 0.178, model Trio.1 in Table four ). However, our estimates are not significantly different from zero. Both the sign and the magnitude of the estimated long-run coefficients on the output volume and Chinese market index are rather similar to short-run estimates (0.007 and 0.216, Table seven in Bouoiyour and Selmi). The long-run effect of the gold price coefficient decreases by two orders of magnitude (−0.002) compared to the short-run influence in Bouoiyour and Selmi. However, none of the long-run global macroeconomic and financial development coefficients is statistically significant in Bouoiyour and Selmi ( two thousand fifteen ). In contrast, both our global macroeconomic and financial development coefficients have a statistically significant influence on BitCoin price in the long-run. Whereas the influence of the Dow Jones index is positive (16.Ten, model Trio.1 in Table six ), the influence of oil price is negative (−4.52, model Three.1 in Table six ). Note that signs of our long-run estimates are the same as those of Bouoiyour and Selmi ( two thousand fifteen ): Dow Jones index and Chinese market index have positive influence, whereas gold price and oil price have negative influence on BitCoin price. Given that our long-run estimates are statistically significant but those of Bouoiyour and Selmi ( two thousand fifteen ) are not, our estimates can be considered as more reliable.

van Wijk ( two thousand thirteen ) finds that the value of the Dow Jones has positive and statistically significant effect on BitCoin price in the long-run (0.525, Table Trio.Trio in van Wijk). The positive and statistically significant coefficient is in line with our results for models Three.1 and Four.1. The differences in the estimated elasticities may be caused by the fact that the analysis of van Wijk ( two thousand thirteen ) (1) does not consider other determinants of BitCoin price, such as BitCoin request and supply, and BitCoin attraction, and (Two) is based on a <Three year period, whereas our analysis is based on a Five.Five year period, capturing extreme price switches in two thousand thirteen and two thousand fourteen (see Fig. One ). The estimated influence of the Euro–US dollar exchange rate is negative and statistically significant (−0.335, Table Three.Trio in van Wijk). The results of van Wijk ( two thousand thirteen ) suggest that, if the US dollar appreciates against Euro, it is most likely to be the case that it also appreciates against the BitCoin. Consequently, an increase in the Euro–US dollar exchange rate leads to a decrease in the amount of US dollar that have to be paid for one BitCoin, which decreases its value. In contrast, our estimates suggest a positive relationship inbetween the Euro–US dollar exchange rate and BitCoin price. However, none of our estimated models is statistically significant (1.Trio, 1.Five and Four.1). Given that the estimates of van Wijk ( two thousand thirteen ) are statistically significant and plausible from an economic theory point of view, likely they are more reliable. van Wijk ( two thousand thirteen ) also estimates the influence of oil price on BitCoin price. The estimated influence is negative and statistically significant (−0.242, Table Trio.Trio in van Wijk). These results are in line with our estimates (−4.52, model Three.1 in Table six ). As explained above, differences inbetween the estimates may be caused by the fact that the times series of van Wijk ( two thousand thirteen ) are only half as long as ours, not capturing the extreme price switches in two thousand thirteen and 2014, as well as they do not control for other significant BitCoin price drivers.

Four Conclusions

Currency, or money in general, is typically defined as having three main functions: a medium of exchange, a unit of account, and a store of value. The present paper attempts to analyze BitCoin features, which may facilitate to become a global currency, as well as characteristics, which may impede the use of BitCoin as a medium of exchange, a unit of account, and a store of value.

The very first function of any currency is to intermediate the exchange of goods and services. Being not a legal tender, BitCoin is fully dependent on voluntary adoption by market participants as a medium of exchange. BitCoin primary advantage relative to standard currencies for its use in exchanges is lower costs of transfers as there are minimal costs linked to third-party intermediaries. Albeit, BitCoin has shown a phenomenal growth during the last years, it still has a negligible market presence globally as a medium of exchange. Moreover, evidence tends to support the view that many of BitCoin transactions involve transfers inbetween speculative investors and are not used in exchanges of goods and services. BitCoin faces also several other obstacles in becoming a widely used medium of exchange such as users’ difficulty of procuring fresh BitCoins, the problem of irreversibility of BitCoin transactions, which does not permit to treat disputes, inability to grant loans denominated in BitCoin, and a relatively high level of computer skill required for using and understanding BitCoin.

The 2nd function of a currency is to serve the function of a unit of account by being able to convey the relative value of goods and services in the economy. The major concern of BitCoin is its high price volatility which may reduce its power to fulfill this function accurately. However, entrepreneurial innovations such as market exchange pricing and instantaneous exchange facilities may remedy this problem but not fully eliminate it. Further, nearly-infinite sub-divisibility is a excellent advantage of BitCoin relative to standard currencies, however in some instances it may pose confusion among consumers given that the price quotation differentiated in the magnitude of several decimal places may be undistinguishable for them.

The third function of a currency is to serve as a store of value over time. Standard currencies are usually inflationary, while BitCoin is associated with deflationary pressures, if it becomes widely used. While this is beneficial to BitCoin holders, the expectation of higher future value of BitCoin may lead to its hoarding, which may reduce its use in exchanges of goods and services. One of the main threats to BitCoin capability to preserve the value to its holders is the security problem linked to cyber-attacks given that BitCoin is a virtual currency, its system is fully internet based and has no oversight institution entrusted to protect the system.

The identified BitCoin features are compared with traditional currencies and their possible influence on BitCoin functions as a currency. We identify several features of BitCoin as a medium of exchange, which differ substantially from traditional currencies: legal tender, immobile costs, network externalities, transaction costs, dispute resolution, credit market, anonymity and transparency. We also find two key characteristics of BitCoin as a unit of account, which differ substantially from traditional currencies: price volatility and divisibility. Eventually, we identify two features of BitCoin as store of value, which differ substantially from traditional currencies: non-inflationary supply and cyber security.

Among all BitCoin features identified and analyzed in this paper, eventually, price volatility is the one with the largest differences compared to standard currencies, such as US dollar, Euro, Yen, British Pound. As shown in Fig. One , since its introduction in 2009, BitCoin price has shown utterly high volatility. One consequence of such extreme price volatility is a threat for BitCoin being accepted as a global currency, as it may fail to accurately convey relative prices of goods and services in the economy and may generate uncertainties to its holders due to its inability to preserve stable value over time. Indeed, a desirable property of any monetary mean is that it holds its value over short-to-medium periods of time, in order not to create distortion when used as a medium of exchange in transactions, a unit of account, and a store of value. Large price movements alter the purchasing power causing risk and costs to firms and consumers. Given that BitCoin is a relatively fresh currency, its price formation is not well understood yet, as there are only few studies on BitCoin price formation available in the literature.

In order to better understand reasons for such enormously high price volatility, in the 2nd part of the paper we attempt to identify drivers of BitCoin price and estimate their importance econometrically. The previous literature suggests three types of drivers determining BitCoin price development: (1) market compels of BitCoin supply and request, (Two) BitCoin appeal, and (Trio) global macroeconomic and financial developments. We apply time-series analytical mechanisms to these drivers using daily data for the 2009–2014 period.

Our empirical results confirm that market compels of BitCoin supply and request have an influence on BitCoin price, implying that, to a certain extent, the formation of BitCoin price can be explained in a standard economic model of currency price formation. In particular, the demand-side drivers, such as the size of the BitCoin economy and the velocity of BitCoin circulation, were found to influence BitCoin price. Given that BitCoin supply is exogenous, likely, the development of the request side drivers will be among the key determinants of BitCoin price also in the future potentially lading to deflationary pressures if BitCoin use expands.

However, the influence of demand-side drivers on the BitCoin price somehow reduces when controlling for BitCoin attraction, which implies that these drivers emerge to be relatively more significant. In fact, we cannot reject the hypothesis extensively emphasized in the literature that investor speculations are also affecting BitCoin price. The statistically significant influence of Wikipedia views on BitCoin price could be an indicator of speculative short-run behavior of investors, or it may capture the expansion of the request as a medium of exchange of the BitCoin. Additionally, we find that also fresh information influence BitCoin price positively, which may be a result of an enhanced trust among users. As such, speculative trading of BitCoins is not necessarily an undesirable activity, as it may generate benefits in terms of absorbing excess risk from risk reluctant participants and providing liquidity on the BitCoin market. A negative side of the short-run speculative investment is that it may increase price volatility and create price bubbles which has adverse implication for BitCoin users. A further negative side of the long-run speculative investment is more extensive hoarding of BitCoins which may reduce its use in exchanges. The success of BitCoin thus also hitches on its capability to reduce the potential negative implications of such speculations and expand the use of BitCoin in trade and commerce.

Eventually, our econometric estimates do not support previous findings that the global macro-financial development may be driving BitCoin price. In fact, once we control for supply–demand variables and BitCoin’s attraction for investors linked to BitCoin, the influence of global macro-financial development captured by the Dow Jones Index, exchange rate and oil price becomes statistically insignificant. Because macroeconomic switches are not reflected in BitCoin price movements its price volatility cannot be lightly hedged. This is in contrast to standard currencies which are strongly driven by macro developments and hedging option is available and widely used.

In summary, our probe has shown that there is a disagreement in the literature on whether BitCoin can become a global currency. Overall, negative assessments about BitCoin as a currency tend to prevail given that several BitCoin characteristics identified in the paper impede its use as a currency and thus constrain its expansion globally. However, as outlined in the paper some BitCoin characteristics give its predisposition to be adopted at least in some segment of money market if not in a broader context. In particular, BitCoin may have a high relative comparative advantage with respect to standard currencies in countries with unstable financial system (e.g. in developing countries), and may provide an alternative to standard currencies in countries with poor and not widely available financial services, non-convertible currency, expensive financial services and high administrative cargo in opening an account. In addition, BitCoin may represent a cost-effective remittance system in developing countries, were traditional transfers are very expensive and the banking system is underdeveloped and unsecure. Given that BitCoin transfers can be done with relatively minimal cost and resource requirements and are independent of geographical location or banking system in place, they are ideally placed to serve as an efficient international remittance system. Ten

The BitCoin technology—BlockChain eleven —opens up several other possibilities and technological innovations, including micropayments, crowdfunding, distributed exchanges, brainy property, property registry, ticketing and secure voting systems. For example, BlockChain technology could provide a way to track the unique history of individual devices, by recording a ledger of data exchanges inbetween it and other devices, web services, and human users. Similarly, BlockChain could enable wise devices to become independent agents, autonomously conducting a diversity of transactions. For example, a vending machine could not only monitor and report its own stock, but also solicit bids from distributors and pay for the delivery of fresh items automatically—based on the purchase history of its customers. Further, the disruptive innovation of BitCoin provides the potential to give citizens direct control over their financial activities by removing costly—and sometimes obscure—intermediation layers fostering financial inclusion .

Footnotes

Note that virtual currencies must be distinguished from electronic money. The key distinguishing feature of electronic money is that their link with traditional money is preserved and have the same unit of account as well as they have legal foundation and are regulated. This is not the case of virtual currencies including BitCoin (ECB two thousand twelve ).

Given that freshly issued BitCoins decrease over time, miners will have to rely more on transaction fees to recoup their investment in mining which may lead to higher transaction fees in the future (EBA two thousand fourteen ). For a theoretical analysis of the economics of BitCoin transaction fees see Kroll et al. ( two thousand thirteen ) and Houy ( two thousand fourteen ).

There was approximately $1.37 trillion in circulation as of June, two thousand fifteen (Federal Reserve System 2015).

Access to the BitCoin network requires downloading a BitCoin software on private computer and joining the BitCoin network, which permits users to engage in operations, and update and verify transactions.

A related problem pointed by Yermack ( two thousand fourteen ) is linked to relatively high diversity of BitCoin prices across different exchanges and web quotations at any given time. This variation in prices poses problem to establish a valid reference point for price setting for both consumers and businesses.

Moreover, BitCoins which were accidentally lost or ruined can never be substituted, resulting in shrinkage of the money base and leading to enhanced deflationary trend. For example, the bitomat exchange, one of the largest exchanges, lost around $200 thousand worth of BitCoins (at the exchange rate at the time) due to human error by accidentally erasing cloud server where the wallet was stored (Barber two thousand twelve ).

Given that people consider a currency valuable if they expect others to do so, for a decentralized currency, such as BitCoin, trust depends on a belief that the rules of the currency will be stable over time.

The Dow Jones Index is an industrial average that captures thirty major corporations on either the NYSE or the NASDAQ.

An example of the BitCoin based system for remittance transfers is BitPesa. BitPesa is an online payment platform that uses BitCoin to suggest money transfers to and from East Africa (Folkinshteyn et al. Two thousand fifteen ).

BlockChain is a public ledger of all BitCoin transactions that have ever been executed; it is permanently growing as ‘completed’ blocks are added to it with a fresh set of recordings.

Notes

Acknowledgments

The authors are grateful to Tony Tam for providing access to the BitCoin data of Bitcoinpulse. We gratefully acknowledge financial support from the research project VEGA1/0797/16. The authors are solely responsible for the content of the paper. The views voiced are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission.

Related video:


Leave a Reply

Your email address will not be published. Required fields are marked *