BSc thesis Economics and Business
Economics and Finance
The effect of regulation on the price of Bitcoin
An event study on the bitcoin market
Written by: Vincent Zuurveen Student Number: 1034539
Table of contents
1. Introduction
3
2. Virtues and shortcomings of Bitcoin
4
2a. Advantages of Bitcoin
5
2b. Disadvantages of Bitcoin
6
3. Reviewing Regulation
8
3a. Regulatory State in the U.S.
8
3b. Regulatory State in Japan
10
3c. Regulatory State in China
11
3d. The Future of Regulation
13
4. Methodology and Data Analysis
15
4a. Timeframe
15
4b. Calculating Abnormal Returns
16
4c. Significance Testing and Results
19
4d. Discussion
25
5. Concluding Remarks
26
Bibliography
27
Statement of Originality
This document is written by Student Vincent Zuurveen who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
1. Introduction
In the past few years, and months especially, bitcoin and other cryptocurrencies have witnessed a tremendous rise in popularity. Though where some argue cryptocurrencies to be the future of payment methods (Kaplanov, 2012), others remain skeptical (Kleiman, 2013). Due to the high rate of criminal activity associated with cryptocurrencies, fueled by the anonymity it provides, cryptocurrencies have become a grave concern for financial regulators worldwide (Ponsford, 2015). Though many governments are still careful in their approach on
cryptocurrency regulation, China has proven to be more decisive. China has been imposing restrictions on the bitcoin market in multiple occasions already (Ponsford, 2015). Arguably the most significant example of this has been the shut down on all domestic cryptocurrency exchanges in September 2017.
While regulation is globally still in its infancy, it can be expected that more countries will start taking a focused directive with regards to the regulation of cryptocurrency markets. While South-Korea is already seemingly following China’s example (Titcomb, 2018), a collective of European countries have the ambition to regulate bitcoin at the G-20 level (Buergin, Jennen & Follain, 2017).
Which direction future regulation is going to take is still hard to determine, but it will undoubtedly further characterize the cryptocurrency market, and in particular the bitcoin market. For bitcoin investors it will be important to know how their bitcoins will be affected by potential future regulation and how their value will be influenced. Therefore, the main question of this paper is: “How does the market for bitcoin respond to tight regulatory measures?”
It can be easily hypothesized that tight regulation certainly will have an impact on the bitcoin market as it does on any market. Though the bitcoin market may prove to be more resilient than other traditional markets. It’s quite easy for investors to simply move their digital wallets to another country when cryptocurrencies are no longer in favor in their home
countries.
To answer this question, I’ll be conducting an event study on the announcement of the Chinese government to ban its domestic cryptocurrency exchanges. In addition, I will also analyze the announcement of the Japanese government to acknowledge bitcoin as a currency, which supposedly influenced the bitcoin price positively. An event study, as described by MacKinlay (1997), is used to investigate the significance of a price swing as a result of a certain event or announcement. A simultaneous question I try to answer with this paper is whether the
event study methodology proves fruitful when applied to the bitcoin market. To perform an event study, we need a reliable estimate for the expected return. For a market as volatile and unpredictable as that of bitcoin this is exceedingly hard.
I find that the event studies described in this paper show highly significant price swings as a result of their respective events. Thereby implying that tight regulatory measures negatively affect the bitcoin price. While positive announcements, such as the Japanese government acknowledging bitcoin, have a clear positive effect on the bitcoin price. I used both the mean return model and the market return model for the analysis. For the market return model, I used three separate market indexes. First the S&P500, for when bitcoin is viewed as an asset class (Glaser, Zimmermann, Haferkorn, Weber & Siering, 2014). Second, the PowerShares DBC index, for when bitcoin is viewed as a commodity (Swartz, 2014). And lastly the PowerShares DB G10 currency harvest fund, for when bitcoin is viewed as a currency as intended by its creator (Nakamoto, 2008). I find the event study model to be quite valuable in calculating the significance of these price swings, though some prudency with respect to time windows and confidence levels is required.
The paper is organized as follows. In section 2 I present a literature review on the virtues and shortcoming of cryptocurrencies, and more specifically bitcoin. In section 3 I present a literature review on the current regulatory environments of three major regions: The United States, Japan and China. I conclude this section with a remark on the future of bitcoin
regulation. In section 4 I present my methodology and data analysis. This section will be concluded by a discussion on the results found in this research. Ultimately, Section 5 will be used for some concluding remarks.
2. Virtues and Shortcomings of Bitcoin
The reason why bitcoin and cryptocurrency regulation is such a challenge for legislators, is because while bitcoin has some downsides that need to be dealt with, it also provides a lot of upsides that might be lost when regulation is too tight or poorly executed (Kaplanov, 2012). Therefore, to better understand the challenge faced by legislators, I will use this part to discuss the pros and cons of a world where bitcoin prevails as a relevant currency.
2a. Advantages of Bitcoin
One of bitcoin’s most notable features is its potential to lower transaction costs. When no banks or financial intermediaries are involved, less money is wasted on transaction fees and administrative costs. Even though bitcoin transaction fees have risen as usage went up, the transaction costs, especially with larger sums of money, remain lower than with many
traditional ways of sending money. Where traditional payment methods often require fees or commissions to be paid to financial intermediaries, bitcoin transaction typically only require a fixed amount to cover the computing costs of the transaction. This may be particularly attractive for small businesses, to whom transaction costs are a significant part of their total expenses. While credit cards have simplified transactions a lot for consumers, merchants often face a variety of fees (Tsukerman, 2015)
Bitcoin also allows businesses protection against chargeback fraud, where a customer falsely claims a product has not been delivered or a service not provided. Due to the
irreversibility of a blockchain transaction, you can assure the payment is made the moment the service is provided or good delivered, without giving dishonest customers the chance to falsely reclaim their money (Tsukerman, 2015).
But private users might also profit from bitcoin’s lowered transaction costs. Imagine holding a bitcoin wallet in The Netherlands and going on vacation in the United States. With bitcoin as a globally accepted method of payment, you won’t have to visit an exchange office and trade your Euros for Dollars. This saves you an exchange fee as well as time and effort (Dwyer, 2015).
Bitcoin might also proof to be a particularly efficient tool in emerging and developing countries. Partly by lowering transaction costs for sending remittances back to family in the home country (Tsukerman, 2015). Perhaps more importantly though, bitcoin might be able to allow people from developing countries to access the global market. This proves difficult for a lot of people in said countries since they are often unable to get bank accounts. Even those with credit cards see their international payments often declined due to high fraud rates in these payments. A digital wallet with an international currency might help solving this issue and create more sophisticated financial markets in these countries (Tsukerman, 2015).
Something that might also be considered an advantage is the lack of government influence. This would make bitcoin a currency that is not directly affected by the situation in a
country. The effect of exchange rate swings due to government failure, or conflict within or between countries could be largely offset because of this. This could make digital currencies, such a bitcoin, a superior store of value if it’s able to stabilize (Plassaras, 2013).
Finally, an advantage of bitcoins lies in its anonymity, though this is clearly a double-edged sword. A bitcoin transaction roughly consists of a user’s online location and the amount of bitcoin involved in the transaction. A person’s identity remains unknown unless the user voluntarily publishes it. This is a comforting idea for people who are concerned with things like phishing and identity theft (Kaplanov, 2012).
2b. Disadvantages of Bitcoin
Digital currency markets, and bitcoin especially, have been a target for much discussion and controversy since its first introduction. The most important reason for this lies within the last mentioned “advantage”: Anonymity. While anonymous trading might be an advantage in some occasions, it also opens up a world of possibilities for more sinister company. Doubtlessly, a regulator’s main concern regarding decentralized cryptocurrencies is tackling this part of the problem.
Nowadays black-market activity is intertwined with public imagination regarding bitcoin. The biggest case in bitcoin history that helped strengthen this perception was that of the
infamous “Silk Road” website. Combining bitcoin’s anonymous transactions with the
anonymizing network browser “TOR” (short for “the onion router”), the Silk Road facilitated the trade of approximately 1.2 billion dollars worth of illegal goods. These goods varied from several types of illegal drugs, to fire arms and hacking tools (Kleiman, 2013). The website operated for over two years before a multi-agency task force managed to seize its assets and catch its owner with his laptop open.
Though the Silk Road has been closed down, a lot of similar black-market sites are doubtlessly still in operation. And it’s not just black-market sites that pose a threat, cryptocurrencies are also used for other illegal activities such as terrorism funding. In the current regulatory landscape these channels remain hard to control, causing a threat to national- and international security that is not to be underestimated (Kleiman, 2013).
Another way the cryptocurrency technology can be misused is as a vehicle for tax evasion. Since many government anti-tax evasion strategies are currently aimed at financial
intermediaries, as opposed to the tax havens that host them, cryptocurrencies become a more important tool for tax evasion. Cryptocurrencies also intrinsically possess some advantages over traditional tax havens. First of all, as said, they are not dependent on financial intermediaries (Marian, 2013). Second, since they’re held in online wallets they do not operate in a particular jurisdiction which means they are not subject to taxation at the source (Marian, 2013). Lastly, users can hold as many wallets as they want, without providing any identifying information. With this, bitcoin wallet owners would evade international anti-evasion laws, unless they for some reason self-reported.
Furthermore, the widespread adoption of digital currencies like bitcoin still faces a lot of uncertainty. Despite its rapid growth and increasing popularity, it remains hard to determine whether these currencies might become widely accepted as a manner of payment, rather than to hold as a security like asset (Plassaras, 2013). Another way in which digital currencies are subject to uncertainty derives from its value. Since it does not rely on any intrinsic value, such as gold, nor derives its value from a commodity expressing some sort of purchasing power, the true value of digital currencies is uncertain and mostly determined by the perception of the public. Some critics even compare these currencies with Ponzi schemes (Plassaras, 2013). The comparison is easily made since someone else is needed to buy bitcoins if another investor wishes to retrieve his funds. Though such a comment may be a little short sighted since it implies the bitcoin itself, holds no value as a tradable currency.
A form of uncertainty which is perhaps the scariest for the general public is the potential damage hackers might inflict by hacking bitcoin exchanges. Bitcoin based companies and
exchanges are still relatively young due to the fact that bitcoin itself is still a new concept. Due to this, these companies may not be well equipped and lack the resources to fend off more sophisticated hackers (Tsukerman, 2015). It is important to note that the blockchain technology itself is nearly impossible to crack, due to the high levels of encryption needed to create a cryptocurrency (Nakamoto, 2008). The intermediaries that provide these online wallets can however turn out to be vulnerable. Especially since these intermediaries are in most cases not subject to the same capital holdings requirements as regular banks and stock exchanges (Tsukerman, 2015). A virtual attack on these exchanges can have disastrous consequences for private online wallet owners, who may see its entire contents going up in smoke. The most notable example of a case where the fear of hackers became reality, was in early 2014. Hackers managed to break into the systems of Mt. Gox. Mt. Gox was the most dominant bitcoin
exchange platform at the time, accounting for almost 80% of bitcoin transactions in the previous year. The hackers stole an approximated 850.000 bitcoins (at the time valued at about 450 million dollar). Eventually, roughly 200.000 bitcoins where recovered but the damage was done and Mt. Gox soon suspended trading (Trautman, 2014).
So as now is determined multiple concerns are involved with bitcoin. At the center of the concerns seems to be the variety of criminal activity that is enabled and amplified by the potential of the blockchain technology. Regulators in different regions are taking measures to battle these issues, though regulation is still in its preliminary stages. In the next part I will be discussing the current regulatory landscape, followed by proposed directions for future regulation
3. Reviewing Regulation
In this part I will be examining the current frameworks that are in place that can be used for bitcoin regulation. These include old measures in which bitcoin might be included, as well as new regulatory approaches implemented specifically for cryptocurrency regulation. As of yet, the global regulatory response has been varied. I will be looking at the different viewpoints of three of the biggest players in the bitcoin market. That being the United States, Japan and China.
3a. Regulatory State in the U.S.
The United States has made significant effort to get a solid grip on the unique properties and potential risks of virtual currencies. However, despite multiple reports and government ordered research, this hasn’t led to any formal recommendations or central guidelines with regards to the regulation of these currencies. As a result, the current regulatory environment is a chaotic mix of different regulatory bodies, courts and state legislators that all act
independently (Tu & Meredith, 2015).
An interesting example in this is that the US federal government does not officially consider bitcoin a currency, while the US department of treasury does (Ponsford, 2015). Even though specific regulatory directions and outlines are yet to be determined, some existing laws are in place under which virtual currencies might fall. These laws are already used by
governmental agencies to issue their own guidance in the treatment of virtual currencies (Tu & Meredith, 2015).
The Financial Crimes Enforcement Network (FinCEN) announced that sellers or exchangers of bitcoin may be regulated as money transmitters. This means that as a result, a comprehensive anti-money laundering and counter terrorism financing statute is enforced on the use of virtual currencies. FinCEN requires money transmitters to register and report to the federal government. Therefore, FinCEN’s decision solidifies the legal responsibilities for those trading in virtual currencies and provides some form of recordkeeping for companies involved with these currencies (Tsukerman, 2015). A possible additional burden for bitcoin firms in this may be the potential for further state-based regulation. In order to act according to federal guidelines, money transmitters need to obtain state money licenses for the state in which they operate. Moreover, the future of state-based regulation is, much like federal regulation, still unpredictable (Tu & Meredith, 2015).
Another important announcement with regards to regulating virtual currencies was issued by the IRS. They stated that for federal tax purposes, virtual currencies would be treated as property rather than as currency. This means that bitcoin investors are able profit from lower capital gains taxes and some tax write-offs (Tsukerman, 2015). This announcement does have the potential to bear some persistent consequences with regards to the widespread adoption of bitcoin as a currency. Using bitcoin as a currency would now mean you have to take capital gains into account, every time you want to buy something with bitcoin.
Some states have also worked on creating or adapting certain regulatory frameworks in order to accommodate virtual currencies. California for example had an existing section in its Corporations Code that prohibited the use of currencies not issued by a government entity. However, governor Jerry Brown signed an alternative currency law, to replace this section, clearly stating that virtual currency is not banned under Californian law (Tsukerman, 2015).
On the other side of the country, New York is actively working to develop bitcoin-specific regulation. The New York State Department of Financial Services issued the proposed use of a so called “BitLicense” for sellers and exchangers. This would keep exchanges within the visor of regulators, both for law enforcement as for consumer protection (Tu & Meredith, 2015). The implementation of the BitLicense caused for a lot of bitcoin companies to cease New York based operations. Currently only three bitcoin companies are provided with a New York BitLicense.
The United States government seems to be open to the idea of a society where bitcoin plays a role. And since it has invested quite a bit in getting better acquainted with the potential of bitcoin, it seems unlikely that harsher forms of regulation, such as China has demonstrated, can be expected in the near future.
In conclusion, the federal government has implemented some regulation with regards to taxation and recordkeeping. However, the regulatory environment in the U.S. is still an uncoordinated mix of different agencies trying to implement existing legal frameworks, to tackle a narrow part of the regulatory problem. Though some different states and regulatory bodies have tried to clarify their stance with regards to virtual currency regulation, a clear-cut direction is still missing.
3b. Regulatory State in Japan
After the Mt. Gox fiasco happened, confidence in bitcoin was severely damaged in Mt. Gox’ home country of Japan. The association with fraud and theft lingered and continued to have a strong impact on the virtual currency. But quickly after this period, as new bitcoin exchanges opened up and got quiet following, bitcoin slowly regained the public trust (Yagami, 2017).
Where China has decided to actively take measure against bitcoin, the Japanese
government has largely refrained from taking stance against the cryptocurrency. Though back in 2014, bitcoin was still considered to be neither a currency nor a bond under the current law, prohibiting banks and financial intermediaries from dealing in bitcoins.
However, as China and Korea increased efforts to oppress bitcoin usage in their respective countries, Chinese and Korean cryptocurrency traders sought refuge in Japan. This caused for an explosive growth in crypto trading in Japan. In May 2016 Japan announced its first steps towards accepting bitcoin, by recognizing it as a method of payment that is not a legal currency, whilst requiring exchanges to be registered. Roughly a year later, on April 1st of 2017, The Japanese government amended part of its Banking Act to allow virtual currencies, such as bitcoin, as a legal form of payment (Yagami, 2017). Japan’s financial regulators also officially approved the operation of 11 cryptocurrency exchanges in September, right after all Chinese domestic exchanges were forced to shut down (Yagami, 2017).
These measures taken by the Japanese government have significant implications on a global scale. These announcements helped increasing confidence in the bitcoin market as a whole and were doubtlessly part of the reason why the bitcoin price resurged to new all-time highs after the Chinese cryptocurrency crackdown. As of November 2017, daily bitcoin turnover that originated in Japan at times hit highs of 60% of global bitcoin volumes (Yagami, 2017).
Whether Japan proves to be a trailblazer for the rest of the world is yet to be seen. It’s still too early to make comments about the impact this decision will have on the Japanese economy, as well as the position of the Japanese Yen. Evident is that Japan has proven itself as a frontrunner when it comes to its stance in technological advancement and the digital age.
3c. Regulatory State in China
Compared to the United States and Japan, the Chinese government has been more active in regulating bitcoin and attempting to deal with its downsides. The People’s Bank of China refuses to give bitcoin any legal states and doesn’t recognize it as a currency. They also remain concerned that it is not protected and regulated by a central authority (Hsu, 2017).
In December 2013, the PBOC made its first steps in engaging in bitcoin regulation. They released a statement in which financial institutions are being prohibited from handling bitcoin transactions. Meaning these institutions are not allowed to buy, sell or insure any products linked to the cryptocurrency (Ponsford, 2015). These announcements created a turbulent relationship between the central bank and China’s financial institutions, who continued dealing with bitcoin despite the governments warnings. This back and forth continued until the PBOC decided to officially order commercial banks and payment companies to close their bitcoin trading accounts in April 2014 (Ponsford, 2015).
Because of these setbacks, bitcoin exchanges had to come up with creative ways to bypass the governments strict measures. At this time China’s largest exchange site, BTC China, came up with a bitcoin ATM and launched an online buy-and-sell app (Ponsford, 2015). These ATM’s are not connected to commercial banks. Similar bitcoin ATM’s started sprouting up in other regions as well, though lately they have been subject to debate due to their significant money laundering potential.
In 2017 China took another big step in their bitcoin regulation. In September Chinese regulators forced all Chinese domestic bitcoin exchanges to shut down, which ceased all bitcoin
trade in the Chinese domestic currency, the Yuan. This was right after the PBOC announced a ban on Initial Coin Offerings. These ICO’s were an important way for new blockchain based technology companies to gain funding, often in exchange for some new “crypto coin”. ICO’s have been controversial for some time, due to their high risk and fraud rate (Binham, 2017). Individual traders are still allowed to trade in bitcoins and holding a bitcoin wallet is still legal in China. Though due to the closing of their exchanges, the Chinese bitcoin market is severely crippled. At the time of the announcement, the Chinese market accounted for roughly 10% of the total bitcoin trading activity, most of which has shifted to the more forgiving
Japanese market.
Experts believe the ban on domestic bitcoin exchanges is only temporarily, though the future of Chinese bitcoin exchanges remains unclear. Hsu (2017) argues that it might just be a way for Chinese regulators to deal with their overfull agenda’s. China’s regulators have been working hard to battle money laundering and fraud over the years and allowing bitcoin to be exchanged freely is counterproductive in this. Also, keeping open a market used for speculation could plague the Chinese financial markets. Looking back at the stock market crash in the summer of 2015, due to speculation, this might be a real concern. Churchouse (2017) expects Chinese regulators to allow back selected exchanges in time, putting together a framework to ensure transparency and oversight on individual traders.
The future of bitcoin in China remains uncertain. And though China’s motives behind their bitcoin crackdown are clear, outlawing bitcoin entirely, which seems to be the direction, may not be the best approach. First, there are significant economic benefits to be gained by regulating bitcoin for financial markets as well as consumers (e.g. investment opportunities and new business practices and retail services). Second, the decentralized bitcoin system is very difficult to target, and a single government does not have the resources to contain it. Outlawing it would only drive it underground and make it less understood, contributing to the problems that caused for the ban in the first place (Ponsford, 2015).
While the bitcoin market appears to have shaken off the setback of losing the Chinese exchanges, it will be interesting to test the significance of the blow the market took after the announcement. This will give us some insight on the potential effect of tight regulatory policies and the potential effect of similar measures in other regions.
3d. The Future of Regulation
The regulatory problems cryptocurrencies pose is a hot topic in the agendas of financial regulators around the world. Though little explicit directions have been taken to date, some has been written on expected and suggested future measures.
Multiple financial regulators around the world are currently exploring the possibility of a new central bank issued cryptocurrency including China, Sweden and Japan. According to Peter Smith, CEO of Blockchain, a major government issuing a sovereign digital currency is probable within the next two years (Chen, 2017). Research on this subject has also been done by decree of the Canadian government, dubbed “Project Jasper” (Chapman, Garratt, Hendry, McCormack & McMahon, 2017), as well as by the United States government, dubbed “Fedcoin” (Koning, 2016). Both acknowledge the promise bitcoin has, but believe a central authority is necessary to control its volatility to make it accessible for most consumers. Replacing physical cash with the proposed digital form would also mean the central bank would have the capability to implement negative interest rates, which could be useful when recession hits when nominal rates are already low (Koning, 2016).
Though a central bank regulated cryptocurrency goes against the basic principles of bitcoin, in the fact that it loses its decentralized nature, it may very well replace physical money in the future. However, Bitcoin and other decentralized cryptocurrencies remain a different product entirely, since a lot of its appeal comes from the fact that it is not control by a central authority. The effect central bank issued cryptocurrencies will have on the bitcoin market remains uncertain, but it’s obvious that it cannot replace bitcoin entirely and therefore will not be a solution against the negative connotations of bitcoin.
The direction most governments intend to take on the path of bitcoin regulation Remains unclear. Most will agree that some form of regulation will be necessary, though the “wait-and-see” approach appears to be most popular for now. Sonderegger (2015) argues that some elements of the Chinese approach should be emulated, mainly the prohibition of financial institutions in dealing with bitcoin. However, she states goods and services should be priced in bitcoins, and it should be permitted as a payment system. She advocates a “vague framework that both legally defines Bitcoin while ultimately allowing it the regulatory freedom it requires to fully develop”. She further believes governments should take a hands-off approach while
limiting regulation to fighting criminal activity. Regulation that is too tight might chase bitcoin away leaving a country out of the regulatory discussion (Sonderegger, 2015)
When looking at the countries we’ve mentioned in this paper, it can be stated that China and Japan have taken almost opposite regulatory approaches. The United States on the other hand is still clouded in a mishmash of individual regulators trying to target narrow parts of the problem. Tu & Meredith (2015) stress the importance of greater interagency communication on the direction of regulation and pursuit more cohesive action. To do so policy goals of existing laws need to be evaluated and applied to the unique nature of virtual currencies. In doing so policymakers can better develop appropriate regulatory requirements to tackle the risks involved with virtual currencies (Tu & Meredith, 2015).
Tsukerman (2015) adds that in order for bitcoin to gain a wider adoption, regulators should focus on two tasks. First, regulators must create a system where bitcoin is clearly defined as a currency. While it’s currently defined differently by different agencies, some sort of
unanimity needs to exist on bitcoin’s nature in order for consumers to get more comfortable with relying on bitcoin as a medium of exchange. This implies that, according to Tsukerman (2015), the IRS’s current policy of treating Bitcoin as property must be reconsidered. Secondly, Tsukerman (2015) states that in order for bitcoin to lose its entanglement with criminal activity and deep web denizens, bitcoin users should have to register their public key addresses to their real identities.
This last measure may partly solve the problem anonymity brings with it, but it also denies bitcoin users the advantages anonymity can provide. When a measure like this is not internationally coordinated, it might simply drive bitcoin users away from the United States, as mention by Sonderegger (2015). This would hurt the domestic bitcoin market and may simply only shift the problem to another region.
Though the future of cryptocurrency regulation remains shrouded in mystery, it will most likely be of great importance to keep in track for individual investors in bitcoin and other cryptocurrencies. The tightness of the inevitable upcoming regulations will certainly directly affect the bitcoin market and its price. To investigate some of the effect these regulatory actions might have, I’ll be taking a look at the significance of the bitcoin price swing right after China ordered their domestic bitcoin exchanges to shut down. This will tell us something about the potential hit the bitcoin market might take, when future policies are too tight. In order to do this, I’ll be performing an event study around the time of the announcement, which I will
describe in the next part. Event studies are a great and easy way to measure the effect of an economic event, in this case the Chinese bitcoin exchange ban, on the value of a “firm”, in this case the bitcoin market (MacKinlay, 1997).
4. Methodology and Empirical analysis
To try and measure the effect strict regulation has on the market for bitcoin I’ll be looking at the example of early September 2017, when the Chinese government banned domestic bitcoin exchanges. Additionally, I’ll be looking at a second event around May 25th of 2016, when Japan officially acknowledged bitcoin as a currency and method of payment. To do this I will be conducting an event study as described by MacKinlay (1997). An event study looks at an event at a specific point or period in time and measures the effect this event had on the value of, in this case, bitcoin. We look at the actual returns around this event and compare them with the expected returns, to find the abnormal returns. For this I needed historical data the price of bitcoin, which I found via http://www.coinmarketcap.com. I also needed historical data on the closing price of a couple market indexes, as a means of calculating the expected return. The market indexes used are the S&P 500, the PowerShares DBC index and the PowerShares DB G10 currency harvest fund. I found these via http://www.nasdaq.com.
4a. Timeframe
To conduct an event study, it is important to start off by deciding on an appropriate timeframe. First reports of the Chinese exchange ban came out around the seventh of
September 2017. The announcement of Japan accepting bitcoin as a method of payment was on the 25th of May 2016. The ten days prior and after these events is called the event window. To calculate the expected return, we need historic data on bitcoin returns that is not influenced by the event. This estimation window is a longer period directly prior to the start of the event window (MacKinlay, 1997). Deciding the length of the estimation window can be a tricky thing in an ever-evolving market such as bitcoin. Enough data points are needed to make a sensible estimate of the expected return (see next section) but other shocks should be excluded as much as possible to avoid a bias in the estimation window.
MacKinlay (1997) states, that when using daily data an estimation window of 120 days typically suffices. When looking at the historical market price for bitcoin during the 120 days before both event windows, no obvious events can be found that would skew the estimation results. Since the historical data of the market indexes miss a few days (weekends and holidays), I have slightly extended the proposed estimation window by McKinlay to 150 days prior to the event window.
Image 1
Timeline of the event study on the Chinese crypto-exchange ban
4b. Calculating Abnormal Return
To determine the significance of the bitcoin price drop as a result of our event we need to find the abnormal return during the event window. Once the daily abnormal return is found, the cumulative abnormal return (CAR) can be calculated to analyze the total cumulative effect of the event.
First of all, to calculate the expected return, I’ll be looking at both the constant mean
return model, as well as the market return model. The constant mean return model assumes
normal returns to be a constant 𝑋𝑡. This constant is defined as the average (daily) return in the
estimation window. The constant mean return model can be defined as: 𝑅𝑖𝑡 = 𝜇𝑖+ 𝜀𝑖𝑡. Where
𝜀 is an error term with expected value of zero and a variance of 𝜎2𝜀. The market return model is
slightly more sophisticated and compares a relevant market index to the actual returns in the estimation window, to find the expected return. The market model can be defined as: 𝑅𝑖𝑡 = 𝛼𝑖+ 𝛽𝑖𝑅𝑚𝑡 + 𝜀𝑖𝑡. Where 𝜀 is an error term with expected value of zero and a variance of
𝜎2𝜀 , and 𝑅𝑚𝑡 represents the return of the selected market index. Now it should be noted that
categorize. Though it is a currency in its essence (Nakamoto, 2008), it does not behave as such. According to Glaser et al. (2014), most bitcoin owners view bitcoin as an asset, rather than as a method of payment. However, it can also be argued that bitcoin has comparable properties to that of a commodity. Intuitively, bitcoin is controlled by the owners and not by a government like regular currencies. The ‘mining process’ can be compared to that of certain commodities (e.g. gold, silver and other precious metals). And bitcoin is often used for speculation. These characteristics imply a closer relation to commodities then to other currencies (Swartz, 2014).
Due to the fact that the nature of bitcoin is debatable, I settled on three different market indexes to conduct the analysis on the afore mentioned events. First, the S&P500 index, second a diversified commodity index (PowerShares DBC index) and lastly a diversified currency index (PowerShares DB G10 currency harvest fund). Considering bitcoin does not compare well with any individual commodity, I decided to use a broadly diversified commodity index, to best analyze the bitcoin price swing if it is considered a commodity. The PowerShares DBC index is a fitting example of such a diversified commodity index. Furthermore, I decided on the
PowerShares DB G10 currency harvest fund to analyze bitcoin price swings if bitcoin is
considered a currency. The fund is designed to track a broad number of currencies and tries to exploit the trend that high interest rate currencies tend to rise more in value, relative to low interest currencies. It is of a speculative nature and the fund’s trading takes place in very volatile markets. Therefore, it seems a good fit to apply to the bitcoin market.
Once the expected return is computed, the abnormal returns can be calculated by comparing the expected returns with the actual returns in the event window: 𝐴𝑅 = 𝑅 − 𝐸[𝑅] . Once we have these abnormal returns, we can take the cumulative abnormal returns, 𝐶𝐴𝑅 = ∑ 𝐴𝑅𝑡1𝑡0 , of the event window. Here the summation starts at the beginning of the event period
and goes on till the end of the event period. It may also be extended into a post-event window if so desired. In the graphs below the CAR is plotted and compared to the expected return. This is done for the mean return model and the market return model with the diversified commodity index. The event in both graphs is the Chinese government banning domestic cryptocurrency exchanges.
Image 2
Graphical representation of the Cumulative Abnormal Return, compared with the mean return. Around the time the Chinese government banned domestic cryptocurrency exchanges. Using the Mean Return Model.
Image 3
Graphical representation of the Cumulative Abnormal Return, compared with the expected return. Around the time the Chinese government banned domestic cryptocurrency exchanges. Using the Market Return Model and the Powershares’ DBC index.
-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 8 /2 3 /2 0 1 7 8 /2 5 /2 0 1 7 8 /2 7 /2 0 1 7 8 /2 9 /2 0 1 7 8 /3 1 /2 0 1 7 9 /2 /2 0 1 7 9 /4 /2 0 1 7 9 /6 /2 0 1 7 9 /8 /2 0 1 7 9 /1 0 /2 0 1 7 9 /1 2 /2 0 1 7 9 /1 4 /2 0 1 7 9 /1 6 /2 0 1 7 9 /1 8 /2 0 1 7 9 /2 0 /2 0 1 7 9 /2 2 /2 0 1 7 9 /2 4 /2 0 1 7 9 /2 6 /2 0 1 7 9 /2 8 /2 0 1 7 9 /3 0 /2 0 1 7 C AR Date
DBC return E[R] DBC return CAR -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 8 /2 8 /2 0 1 7 8 /2 9 /2 0 1 7 8 /3 0 /2 0 1 7 8 /3 1 /2 0 1 7 9 /1 /2 0 1 7 9 /2 /2 0 1 7 9 /3 /2 0 1 7 9 /4 /2 0 1 7 9 /5 /2 0 1 7 9 /6 /2 0 1 7 9/ 7/ 20 17 9 /8 /2 0 1 7 9 /9 /2 0 1 7 9 /1 0 /2 0 1 7 9 /1 1 /2 0 1 7 9 /1 2 /2 0 1 7 9 /1 3 /2 0 1 7 9 /1 4 /2 0 1 7 9 /1 5 /2 0 1 7 9/ 16 /20 17 9 /1 7 /2 0 1 7 C AR Date
4c. Significance testing and results
To test the significance of the price fluctuation I use a simple t-test. For this we calculate the standard deviation of the expected return and divide the CAR with this standard deviation. The t-score we get out of this we can translate to a probability level using the t-distribution, which allows us to test its significance. For example, a score of two, meaning a value that lies two standard deviations away from the expected return, scores a significant result with a confidence level of 95%. Knowing this, when we look at Table 1 we see a t-score that goes to an extreme of minus eleven on September 14th, seven days after the event date. When looking at the market return models for this event, comparable results can be found, strengthening the reliability of the outcome.
This result is rather extreme in its significance and we can state with near certainty that the Chinese cryptocurrency exchange ban did significantly hurt the bitcoin market. It should be noted that the t-scores are not quite stable outside the event window either. This is no surprise given the high instability in the bitcoin market. Because of this, results should always be
interpreted carefully.
When looking at the event study on the Japanese government announcing that bitcoin will be acknowledged as a method of payment we find results in line with the ones mentioned above. In this example however, the market seems to respond a little slower and we find our highest abnormal return about three weeks after the announcement.
Table 1
The abnormal return, the CAR and the
corresponding t-test for the event window of the
Chinese crypto-exchange ban, using the mean
Return model (
𝜎 = 0.0435)
Date Abnormal Return CAR CAR T-test 8/28/2017 -0.01054 -0.01054 -0.24197 8/29/2017 0.034319 0.023784 0.546259 8/30/2017 -0.01348 0.010302 0.236623 8/31/2017 0.019763 0.030065 0.690531 9/1/2017 0.029618 0.059683 1.370793 9/2/2017 -0.07452 -0.01483 -0.34068 9/3/2017 -0.00957 -0.0244 -0.56048Table 2
The abnormal return, the CAR and the
corresponding t-test for the event window of the Chinese crypto-exchange ban, using the DBC
Market index to calculate expected returns. (σ = 0.041).
Date Abnormal Return CAR CAR t-test 8/23/2017 0.003514 0.003514 0.085157 8/24/2017 0.037093 0.040608 0.983968 8/25/2017 0.001761 0.042369 1.026641 8/28/2017 -0.00476 0.037604 0.911186 8/29/2017 0.037644 0.075248 1.82334 8/30/2017 -0.00867 0.066574 1.613153 8/31/2017 0.017523 0.084097 2.037758 9/1/2017 0.032679 0.116776 2.829605 9/5/2017 -0.11293 0.003849 0.093257 9/6/2017 0.041269 0.045118 1.093249 9/7/2017 -0.00643 0.038688 0.937463 9/8/2017 -0.08563 -0.04694 -1.13734 9/11/2017 -0.02351 -0.07045 -1.70708 9/12/2017 -0.01501 -0.08546 -2.0707 9/13/2017 -0.06817 -0.15362 -3.72249 9/14/2017 -0.1947 -0.34833 -8.44032 9/15/2017 0.144883 -0.20344 -4.92966 9/18/2017 0.110675 -0.09277 -2.24789 9/19/2017 -0.04126 -0.13403 -3.24778 9/20/2017 -0.0137 -0.14773 -3.57977 9/21/2017 -0.07677 -0.2245 -5.4399 9/4/2017 -0.08612 -0.11053 -2.53856 9/5/2017 0.022615 -0.08791 -2.01915 9/6/2017 0.039918 -0.04799 -1.10232 9/7/2017 -0.00988 -0.05788 -1.32935 9/8/2017 -0.09117 -0.14905 -3.42327 9/9/2017 -0.01112 -0.16017 -3.6787 9/10/2017 -0.03489 -0.19505 -4.47996 9/11/2017 -0.00119 -0.19624 -4.50725 9/12/2017 -0.0178 -0.21405 -4.91619 9/13/2017 -0.07057 -0.28462 -6.53714 9/14/2017 -0.1979 -0.48252 -11.0824 9/15/2017 0.142471 -0.34005 -7.81013 9/16/2017 -0.01392 -0.35396 -8.12975 9/17/2017 -0.02212 -0.37608 -8.63768
Table 3
The abnormal return, the CAR and the
corresponding t-test for the event window of the Chinese crypto-exchange ban, using the DB G10 Currency harvest fund to calculate expected Returns. (σ = 0.041). Date Abnormal Return CAR CAR t-test 8/23/2017 0.001474 0.001474 0.035722 8/24/2017 0.036587 0.03806 0.922465 8/25/2017 -0.00042 0.037645 0.912392 8/28/2017 -0.00621 0.031436 0.761906 8/29/2017 0.036849 0.068285 1.655018 8/30/2017 -0.0089 0.059382 1.439225 8/31/2017 0.022713 0.082094 1.98971 9/1/2017 0.035189 0.117283 2.842582 9/5/2017 -0.1121 0.00518 0.125557 9/6/2017 0.041673 0.046854 1.135588 9/7/2017 -0.00865 0.038204 0.925956 9/8/2017 -0.08792 -0.04971 -1.20485 9/11/2017 -0.01986 -0.06957 -1.68614 9/12/2017 -0.01285 -0.08242 -1.99751 9/13/2017 -0.066 -0.14842 -3.5972 9/14/2017 -0.19516 -0.34358 -8.3273 9/15/2017 0.144626 -0.19895 -4.82203 9/18/2017 0.108334 -0.09062 -2.19636 9/19/2017 -0.03931 -0.12993 -3.14908 9/20/2017 -0.00089 -0.13082 -3.17065 9/21/2017 -0.08884 -0.21966 -5.32375 Table 4
The abnormal return, the CAR and the
corresponding t-test for the event window of the Chinese crypto-exchange ban, using S&P500 Index to calculate expected Returns. (σ = 0.041).
Date Abnormal Return CAR CAR t-test 8/23/2017 0.006275 0.006275 0.151961 8/24/2017 0.037663 0.043938 1.064111 8/25/2017 0.000973 0.044911 1.087685 8/28/2017 -0.00448 0.040432 0.979217 8/29/2017 0.037538 0.07797 1.888328 8/30/2017 -0.0111 0.06687 1.619493 8/31/2017 0.021846 0.088715 2.148561 9/1/2017 0.032774 0.121489 2.942302 9/5/2017 -0.1107 0.010794 0.261404
9/6/2017 0.042608 0.053401 1.293301 9/7/2017 -0.00637 0.047031 1.139019 9/8/2017 -0.08741 -0.04038 -0.97787 9/11/2017 -0.02544 -0.06582 -1.59409 9/12/2017 -0.01511 -0.08093 -1.95992 9/13/2017 -0.0672 -0.14813 -3.58743 9/14/2017 -0.19434 -0.34246 -8.29396 9/15/2017 0.146797 -0.19567 -4.73875 9/18/2017 0.110079 -0.08559 -2.07278 9/19/2017 -0.04173 -0.12731 -3.08337 9/20/2017 -0.01193 -0.13924 -3.37228 9/21/2017 -0.07676 -0.216 -5.23121 Table 5
The abnormal return, the CAR and the
corresponding t-test for the event window of the
Japanese acknowledging bitcoin, using the
Mean Return model
(𝜎 = 0.0244)Date Abnormal Return CAR CAR t-test 5/15/2016 0.004004 0.004004 0.163837 5/16/2016 -0.00762 -0.00361 -0.14786 5/17/2016 -0.001 -0.00462 -0.18887 5/18/2016 0.001686 -0.00293 -0.1199 5/19/2016 -0.03516 -0.03809 -1.55859 5/20/2016 0.008884 -0.02921 -1.1951 5/21/2016 0.000987 -0.02822 -1.15473 5/22/2016 -0.0089 -0.03712 -1.51879 5/23/2016 0.010829 -0.02629 -1.07571 5/24/2016 0.003955 -0.02234 -0.9139 5/25/2016 0.007951 -0.01438 -0.58856 5/26/2016 0.008242 -0.00614 -0.25132 5/27/2016 0.044124 0.037982 1.55407 5/28/2016 0.119338 0.157319 6.436933 5/29/2016 -0.00735 0.149966 6.13605 5/30/2016 0.014334 0.1643 6.722539 5/31/2016 -0.00479 0.159507 6.526461 6/1/2016 0.010241 0.169749 6.945492 6/2/2016 0.00179 0.171539 7.018736 6/3/2016 0.057867 0.229406 9.386462 6/4/2016 0.006054 0.23546 9.634164
Table 6
The abnormal return, the CAR and the
corresponding t-test for the event window of the Japanese acknowledging bitcoin, using the DBC market index to calculate expected returns
(𝜎 = 0.029) Date Abnormal Return CAR CAR t-test 5/11/2016 0.0000806 0.0000806 -0.00276 5/12/2016 0.0039544 0.0038738 0.132384 5/13/2016 0.0018816 0.0057554 0.196688 5/16/2016 -0.0054561 0.0002993 0.01023 5/17/2016 -0.0023972 -0.0020978 -0.07169 5/18/2016 0.0027538 0.0006559 0.022416 5/19/2016 -0.0348710 -0.0342150 -1.16927 5/20/2016 0.0080488 -0.0261662 -0.89421 5/23/2016 0.0038942 -0.0222720 -0.76112 5/24/2016 0.0037959 -0.0184760 -0.6314 5/25/2016 0.0058742 -0.0126018 -0.43066 5/26/2016 0.0081947 -0.0044071 -0.15061 5/27/2016 0.0432955 0.0388883 1.328971 5/31/2016 0.1230308 0.1619192 5.533424 6/1/2016 0.0084046 0.1703238 5.820643 6/2/2016 0.0012991 0.1716229 5.865039 6/3/2016 0.0566036 0.2282266 7.799413 6/6/2016 0.0264250 0.2546516 8.702461 6/7/2016 -0.0172191 0.2374325 8.114015 6/8/2016 0.0056242 0.2430568 8.306219 6/9/2016 -0.0120596 0.2309972 7.894094 Table 7
The abnormal return, the CAR and the
corresponding t-test for the event window of the Japanese acknowledging bitcoin, using the DB G10 currency harvest fund to calculate expected returns (𝜎 = 0.034) Date Abnormal Return CAR CAR t-test 5/11/2016 -0.00053 -0.00053 -0.77146 5/12/2016 0.001144 0.000613 -0.73757 5/13/2016 -0.00123 -0.00061 -0.77388
5/16/2016 -0.00723 -0.00784 -0.98794 5/17/2016 -0.00455 -0.01239 -1.12267 5/18/2016 -0.00101 -0.01339 -1.15251 5/19/2016 -0.03812 -0.05152 -2.28185 5/20/2016 0.005532 -0.04599 -2.11798 5/23/2016 0.000348 -0.04564 -2.10766 5/24/2016 0.000837 -0.0448 -2.08287 5/25/2016 0.004169 -0.04063 -1.95936 5/26/2016 0.005163 -0.03547 -1.80642 5/27/2016 0.040774 0.005305 -0.59857 5/31/2016 0.119418 0.124723 2.938904 6/1/2016 0.006543 0.131265 3.132711 6/2/2016 -0.00144 0.129823 3.089974 6/3/2016 0.054367 0.18419 4.700477 6/6/2016 0.024758 0.208948 5.433868 6/7/2016 -0.01911 0.189833 4.867645 6/8/2016 0.004491 0.194324 5.000677 6/9/2016 -0.01523 0.179098 4.549649 Table 8
The abnormal return, the CAR and the
corresponding t-test for the event window of the Japanese acknowledging bitcoin, using the S&P500 index to calculate expected returns (𝜎 = 0.034).
Date Abnormal Return CAR CAR t-test 5/11/2016 0.000774 0.000774 0.022928 5/12/2016 0.001273 0.002047 0.060619 5/13/2016 -0.00132 0.000724 0.021443 5/16/2016 -0.00647 -0.00575 -0.17031 5/17/2016 -0.00414 -0.00989 -0.29302 5/18/2016 -0.00138 -0.01128 -0.33395 5/19/2016 -0.03826 -0.04953 -1.46704 5/20/2016 0.005862 -0.04367 -1.29343 5/23/2016 7.48E-05 -0.0436 -1.29122 5/24/2016 0.000983 -0.04261 -1.2621 5/25/2016 0.004933 -0.03768 -1.116 5/26/2016 0.005174 -0.03251 -0.96277 5/27/2016 0.041087 0.008579 0.254089 5/31/2016 0.119105 0.127684 3.781582 6/1/2016 0.007187 0.13487 3.994428 6/2/2016 -0.00126 0.133614 3.957204 6/3/2016 0.054775 0.188388 5.579454 6/6/2016 0.025528 0.213916 6.335515 6/7/2016 -0.01849 0.195425 5.787851
6/8/2016 0.00553 0.200955 5.951635 6/9/2016 -0.0153 0.185657 5.498555
4d. Discussion
Some additional remarks must be made. First of all, the fact that the bitcoin market price shrinks as regulation tightens does not necessarily mean ill for bitcoin itself and its potential as a currency. For bitcoin to truly evolve as a currency, it seems to be more important that it gains more stability and regulation may even help with that. The negative implications involved with results of this study are to be suffered by those that use bitcoin as a form of investment and a tool for speculation. The value of their investment will most likely deteriorate as regulation will tighten in the near future.
Further, as already mentioned before, the results from the different t-tests proved to be rather unstable outside the event window. Due to the high volatility of the market we find relatively high t-scores, during the estimation windows as well, that would be classified as significant when using common confidence levels (say 95 percent). This can be solved by using a particularly high confidence level before deciding on the significance of the results (say 99% or even higher). Results in this research have been significant enough even for extremely high significance levels.
Setting a confidence level this high however, means results need indeed be rather large before a price change due to an event is deemed significant. This would mean that price changes due to regulatory action may be overlooked due to too low significance, while they are indeed existing. This would make it hard to apply an event study to subtler regulatory action that may have a less strong effect on the bitcoin market. How this will behave with these subtler regulations will have to be proven in the future.
Another thing that should be noted is that the setting of a timeframe should be done very carefully. The bitcoin market is still very young and constantly changing and evolving. Because of this, data from only a couple of months ago may already be useless when
determining the expected return for the event window. Expect return is significantly different in different periods in the bitcoin market. Aside from this, when choosing an estimation window, the expected return will be influenced by other events in this period, which you don’t want to research. This may give a skewed look on the expected return. So before setting an estimation window it is important to properly research the intended estimation window.
Lastly, though abnormal returns can be reviewed and quantified in hindsight, it will be difficult to properly predict the effect of future regulation. Though it is clear that regulation will have a negative impact on the bitcoin, a lot of other factors will determine the magnitude of this impact. The exact manner of regulation, the region in which regulation takes place, as well as the timing of the regulation and the current state of the bitcoin market will determine how the bitcoin market will exactly react. It is also imaginable that certain forms of regulation may actually increase confidence in the bitcoin market and positively influence the price.
So, while a lot is still unclear on the future of regulation, as well as its effect on the bitcoin market, a lot more will most likely be clear in the future. Will regulation chase people away from bitcoin or will bitcoin prove to be resilient? And will the bitcoin market be able to stabilize in the face of regulation and become and be accepted as a commonly used currency? These questions will be interesting to track in the coming years.
5. Concluding Remarks
The main question of this paper was: “How does the market for bitcoin respond to tight regulatory measures?” The results clearly indicate a significant drop in bitcoin price around the time of the announcement that the Chinese government would shut down its domestic
cryptocurrency exchanges. They also indicated a significant rise in bitcoin price around the time of the announcement that the Japanese government acknowledged bitcoin as a currency and an acceptable method of payment. Based on these results I can confidently state that government regulation directly impacts the market for bitcoin and will be important to bitcoin investors to keep track of. Furthermore, I set out to test the event study model and test its applicability on the bitcoin market. Despite presenting some volatile outcomes outside the event windows as well, the event study proved capable of clearly showing an effect between our event dates and the course of bitcoin market prices. Thus, the event study does prove able to present us fruitful results regarding tight regulatory policies. This goes for both the market return model as the simpler mean return model. It may however prove unqualified when dealing with more subtle forms of regulation, since t-scores may prove less significant and more difficult to decisively judge within the nature of this highly volatile market. Whether this is the case will have to be researched once regulation takes shape in the upcoming years.
Bibliography
Binham, C. (2017, December 15). FCA to examine ‘high risk’ initial coin offering market. Retrieved from http://www.ft.com.
Buergin, R., Jennen, B., & Follain, J. (2017, December 18 ). Germany Joins French-led Moves to Regulate Bitcoin at G-20 Level. Retrieved from https://www.Bloomberg.com
Chapman, J., Garratt, R., Hendry, S., McCormack, A., & McMahon, W. (2017). Project Jasper: Are Distributed Wholesale Payment Systems Feasible Yet? Financial System, 59. Chen, Q. (2017, November 30). Next stop in the cryptocurrency craze: A government-backed
coin. Retrieved from http://www.cnbc.com.
Churchouse, T. (2017, September 16). China's crackdown won't kill cryptocurrencies — but it will have an impact. Retrieved from http://www.businessinsider.com.
Glaser, F., Zimmermann, K., Haferkorn, M., Weber, M., & Siering, M. (2014). Bitcoin-asset or currency? revealing users' hidden intentions. Twenty Second European Conference on
Information Systems, Tel Aviv 2014
Hsu, S. (2017, September 15) China's Cryptocurrency Crackdown: Is Bitcoin A Threat? Retrieved from https://www.forbes.com.
Kaplanov, N. (2012). Nerdy money: Bitcoin, the private digital currency, and the case against its regulation. Loyola Consumer Law Review, 25, 111.
Kleiman, J. A. (2013). Beyond the Silk Road: Unregulated Decentralized Virtual Currencies Continue to Endanger US National Security and Welfare. American University
National Security Law Brief, 4(1), 5.
Koning, J. (2016). Fedcoin: A Central Bank-issued Cryptocurrency. R3 Report, 15. MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of economic
Marian, O. (2013). Are Cryptocurrencies Super Tax Havens?. Michigan Law Review First
Impressions, 112(1), 38-48
Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system.
Plassaras, N. A. (2013). Regulating digital currencies: bringing Bitcoin within the reach of IMF. Chicago Journal of International Law, 14, 377.
Ponsford, M. P. (2015). A Comparative Analysis of Bitcoin and Other Decentralised Virtual Currencies: Legal Regulation in the People's Republic of China, Canada, and the United States. Hong Kong Journal of Legal Studies, 9, 29.
Sonderegger, D. (2015). A regulatory and economic perplexity: Bitcoin needs just a bit of regulation. Washington University Journal of Law & Policy, 47, 175.
Swartz, N. D. (2014). Bursting the Bitcoin Bubble: The Case To Regulate Digital Currency as a
Security or Commodity.Tulane Journal of Technology and Intellectual Property, 17, 319.
Titcomb, J. (2018, Januari 23). Bitcoin drops as South Korea bans anonymous cryptocurrency trading. Retrieved from https://www.telegraph.co.uk.
Trautman, L. (2014). Virtual Currencies Bitcoin & What Now After Liberty Reserve, Silk Road, and Mt. Gox? Richmond Journal of Law & Technology, 20(4), 13.
Tsukerman, M. (2015). The block is hot: A survey of the state of Bitcoin regulation and suggestions for the future. Berkeley Technology Law Journal, 30, 1127.
Tu, K. V., & Meredith, M. W. (2015). Rethinking virtual currency regulation in the Bitcoin age. Washington Law Review, 90, 271.
Yagami, K. (2017, November 2). Japan: A Forward Thinking Bitcoin Nation. Retrieved from https://www.forbes.com.
