An Econometric Approach to the Key Drivers of Bitcoin Price

An Econometric Approach to the Key Drivers of Bitcoin Price

Sang B. Do

University of California, Irvine

Acknowledgements

I would like to extend my greatest thanks to my research paper advisor, professor Fabio Milani, for his guidance and advice. Lastly, I would like to thank the other students in the Economics 137W - Macroeconomics and Financial Market Writing class for offering invaluable feedback.

Abstract

During November 2013, the price of a bitcoin skyrocketed from less than $200 to nearly $1,000 in late 2013 before falling to less than half that level over the course of the next two years, and then rebounding to more than $1,200 in 2017. What exactly caused the high volatility in Bitcoin price? It could be the supply and demand of the Bitcoin network due to the quantity theory of money. It could be the growing popularity of the new digital currency on news coverage. Or it could be the illegal activities using Bitcoin that created a negative impact in public opinion. In order to understand the key drivers that influence the price of Bitcoin, this research paper will examine different drivers from different categories whether it is economic or technical aspect to gain insights of the main drivers of Bitcoin price. We will study key drivers that influence the price of a traditional currency and, then, study the technology that runs Bitcoin network called Blockchain to determine potential drivers of Bitcoin price. Then, we will put those drivers in different categories and collect time-series data for each driver. A regression model will be built based on these drivers and a t-test will be conducted to determine whether these drivers are statistically significant or not to indicate their impact on Bitcoin price.

Keywords: digital currency, Bitcoin, key drivers  

Introduction

Bitcoin is a new open-source digital currency that has caught public attention for both good and bad reasons in the last couple of years. Bitcoin is created as a combination of verifying and recording electronic payments in a public ledger which is operated by an extensive network of computers. The process is called mining and the more computing power that is put in, the more Bitcoins that are created in return. This process will continue until 21 million Bitcoins have been mined at a rate that will take the process well into the next century. Even though Bitcoin is used digitally, some aspects of the digital current could be trace back to the conventional currency. They can be used to purchase goods, services, and can also be exchanged for conventional currency such as US dollars. But one intrigued feature of this digital market is that the price of Bitcoin has soared from about $5 each in 2011 to more than $1,200 as the time this research paper is written. It is easy to think that Bitcoin price’s movement is simply the result of pure speculation; however, there might be other traditional economic drivers that influenced its volatility, such as demand and supply. So exactly what drivers have determined the price of Bitcoin? This is the question that this research paper is trying to answer. There are many type of potential drivers that could determine Bitcoin price. To order by categories, they are economic drivers, technical drivers, investors’ interest driver, historical Bitcoin-related events driver, and safe heaven drivers.

Literature review

The empirical literature this paper contributes to can be summarized as follows: Baek and Elbeck (2014) use the method of detrended ratios in order to study relative volatility as well as drivers of Bitcoin returns using S&P 500 daily return data. They find that Bitcoin volatility is internally driven and conclude that the Bitcoin market is currently highly speculative. Kristoufek (2013) studies the relationship between Bitcoin price and related searched terms on Google Trends and Wikipedia and finds a positive correlation between a price level of Bitcoin and the searched terms as well as a dynamic relationship which is bidirectional. Moreover, the research shows a significant asymmetry effect of search queries related to Bitcoin price above and below a short-term trend. Kjartan and Jens (2015) examine the relationship between Bitcoin price and a series of year-to-year event studies and come to the conclusion that the stability in Bitcoin price’s movement has not yet been achieved. A more mature user base that is not characterized by an investment purpose has to take the lead in order to reduce the volatility. Badev and Chen (2014) perform their empirical analysis on transaction-level Bitcoin data and found that Bitcoin Price Index (BPI) in US dollars exhibits somewhat complicated dynamics. For the period from 2013 to 2014, the BPI increased more than 50-fold. Unnoticed by the public, however, the daily variance of the BPI remained significantly stable for the same period. While there has been remarkable research done to analyze the Bitcoin network, limited research has been conducted to analyze the network’s influence on Bitcoin price. Greaves and Au (2015) investigate the predictive power of Blockchain network-based features on the future price of Bitcoin. As a result of using machine learning algorithms, they obtain Bitcoin price movement classification accuracy of roughly 55% and gain some meaningful insights into the behavior of Bitcoin users. In particular, when the demand is neither high nor low, new users are willing to immediately use their Bitcoin. However, when demand is increasing or decreasing, users are more likely to hoard their Bitcoin and use it less.

Key drivers of Bitcoin price

Economic drivers

According to Buchholz et al., one of the key drivers of Bitcoin price is the interaction between Bitcoin supply and demand on the Bitcoin market (2012). In this paper, we will examine the quantity theory of money to quantify the economic drivers’ impact on the Bitcoin price. In its simplest form, the quantity theory of money is expressed as: MV = PT (the Fisher Equation). Each variable denotes the following:
M = Money Supply 
V = Velocity of Circulation (the number of times money changes hands)
P = Average Price Level
T = Volume of Transactions of Goods and Services

According to the quantity theory of money, the Bitcoin supply is determined by the total amount of Bitcoin in circulation. The Bitcoin demand is represented by the size of Bitcoin economy such as its usage in exchanges for other currencies and the velocity of Bitcoin circulation which measures the frequency at which one unit of Bitcoin is used for purchase of goods and services. We want to examine two economic drivers of Bitcoin price from supply-side and demand-side.

For the supply-side, we use the total number of Bitcoin in circulation which given by a known algorithm and asymptotically until it reaches 21 million Bitcoins. From the theory, everything else being equal, an increase in Bitcoin supply will cause inflation which depreciates the value of one unit of Bitcoin. According to Marc Gronwald, the total number of Bitcoins, the number of Bitcoins in circulation and the growth rate are known with certainty (2015). In other words, there is no uncertainty on the supply-side of Bitcoin price. On the other hand, as a measure of the transactions use, demand for the currency, we use the ratio between trade volume and exchange transaction volume or trade-exchange ratio. It shows the ratio between volume in trade such as purchases or services and volume on the currency exchange markets. Therefore, the lower the trade-exchange ratio is, the more frequently Bitcoin is used for real transactions. From the theory, Bitcoin price should be positively correlated with its usage for real transactions because this increases the utility of holding the digital currency. Overall, the hypothesis is that the observed price fluctuations would be well-explained by the trade-exchange ratio from the demand-side than the number of Bitcoin in circulation from the supply-side.