The digital currency in the eyes of economists – price articles: ups and downs in the bubble

[Editor's note] Faced with the new challenges of digital currency, economists are divided into two categories, one is scornful, and it is not worth mentioning that digital currency is a speculative bubble; the other is cautiously accepted and begins cutting-edge exploration. As the digital currency grows, more and more economists join the second camp. For this reason, Kay unveiled a series of articles on “Digital Currency in the Eyes of Economists”, which is the most comprehensive review of digital currency literature. In the six aspects of classification , market , exchange , price, risk, and supervision, the global economists' research on the frontier theory of digital currency is summarized, which provides a useful reference for interested researchers. This article is the fourth price article in the series, the next one is the risk article, so stay tuned.

The rise of Bitcoin and other digital currencies around the world has puzzled many economists who are familiar with the history of currency development. For the first time in human history, there is an unexplained phenomenon–thousands of institutions and individuals trade through the Internet for a string of virtual numbers that seem to have no value. Numerous platform exchanges that have sprung up have fueled the speculative wave of wild digital currencies. Researchers believe that the speculation of digital currencies is similar to the 19th century gold fever and the 17th century tulip bubble. There are two kinds of attitudes in the business news: optimism and singing. Those who sing the losers think that bitcoin and other digital currencies will eventually disappear. The supporters are looking forward to turning this carnival into a dazzling star through the efforts of unity . Different circles have different views on the attributes of digital currency, so they have different expectations, which leads to misjudgments of their values.

In fact, the digital currency attribute is very unique . In the classification of the previous article “Digital Currency in the Eyes of Economists”, we elaborated on the huge difference between digital currency and other types of assets, so we cannot simply classify it.

• Industry Weekly | Facebook currency is yellow? – Waiting for the official hearing in July
• Mid-term inventory in 2019 (next part): What happened to the encryption industry in the past six months?
• Facebook Libra Supervisor: Users don't have to trust us, they will attend the hearing this month.

The uniqueness of digital currency is also that its production and supply are different : it is generated by the computing power of the computer running CPU, commonly known as "mining", and the quantity and time of supply are fixed. In comparison, the supply and demand of goods can be changed. The traditional money supply can be expanded or contracted according to government policy objectives, and stocks can be added or repurchased according to conditions. But in the digital currency mechanism, supply is generally fixed.

Moreover, commodities, stocks and currencies are based on certain basic values, but the basic value of digital currencies is still difficult to confirm . Some people say that payment is the value of digital currency, which can't help but scrutinize. Cash, credit cards, debit cards, and electronic and mobile payment can all be used for digital currency payment, and even better. If the digital currency can't prove any value base in the real economy, then the price of the digital currency is purely determined by the participants' psychological expectations and emotions. Such a mechanism is destined to have large price fluctuations and is prone to structural bubbles.

In fact, as we will explain in the next section, the only condition for the price of Bitcoin or other digital currencies to remain relatively stable is that as monetary scarcity increases, demand growth slows down – this economic phenomenon is clearly Violation of observations and any economic theory known to man. Because the total amount of Bitcoin is constant, the currency is constantly being dig until it is dug by 21 million, which means that the bit more bitcoin is scarce. From the most basic economic theory, supply growth is less, scarcity is increasing, and demand is unchanged. Prices will rise sharply. If prices are to remain stable, demand will be reduced. The scarcity of goods, in other conditions remain unchanged, generally stimulate the demand psychologically, and rarely see the demand to reduce.

Therefore, the price stability of Bitcoin is theoretically unfounded .

Views of Bitcoin prices from all walks of life

In the financial sector, pricing is a core issue. The representative of digital currency – bitcoin pricing is particularly interesting, especially the price of bitcoin has been staged in the roller coaster market: after bitcoin skyrocketed 20 times in 2017, it reached nearly 20,000 dollars, and fell to 3,000 dollars in 2018 In 2019, it rebounded to more than $10,000 in just three months. This kind of rare huge increase and decrease, it is inevitable that people can not help but suspect that it is a huge bubble.

There is a unique phenomenon. The more prominent the authoritative figure in the financial world, the less optimistic about Bitcoin, the more it is considered to be a speculative bubble, or a scam.

Many well-known economists, including several Nobel laureates in economics, have publicly expressed doubts about Bitcoin. For example, Robert Shiller called Bitcoin a bubble. In an interview with CNBC in April 2018, he said: "Bitcoin The phenomenon seems to me to be a manifestation of human fashion. It looks gorgeous, like the Dutch bubble in the 1740s.” Another Nobel economist, Joseph Stiglitz, said Bitcoin, in addition to avoiding regulation and law. Besides, there are no useful features, so it is illegal. And Nobel economist Paul Krugman said "bitcoin seems to be a pure bubble."

Compared with economists, business and regulatory authorities have more diverse views on Bitcoin. Most people in the scientific and technological circles welcome the bitcoin. Microsoft founder Bill Gates said "bitcoin is the result of a strong technical advancement." Google Chairman Eric Schmidt said that "bitcoin is one of the achievements in cryptography, it is not speculative in itself, but many people do regard it as a speculative product." Antivirus software McAfee founder John McAfee is optimistic about Bitcoin and predicts that "Bitcoin will rise to $500,000 in three years."

However, there are many business leaders in the financial world who are not optimistic about Bitcoin. Jamie Dimon, CEO of JPMorgan Chase, believes that "bitcoin is a 'scam' and is more serious than the 'tulip bubble'." Buffett repeatedly said in the annual meeting and other public interviews that "bitcoin is a 'gambling squad' and will not be hospice." Jim Rogers, a former partner of hedge fund Soros, said: "Bitcoin is a bubble from any angle." Lloyd Blankfein, former CEO of Goldman Sachs, expressed his neutral view of Bitcoin: "I have been thinking about Bitcoin, no In conclusion, there is no support or no objection. "Of course, in the financial world, there are also support for Bitcoin, such as Morgan Stanley CEO James Gorman said: "Bitcoin is definitely not just a short-lived fashion phenomenon."

The regulators have different views on Bitcoin. In 2014, Janet Yellen, then chairman of the Federal Reserve, said: "The Fed does not actually have any regulatory rights over Bitcoin." In June 2015, but Ravi Menon, head of the Singapore Financial Management Association, said: "Bitcoin is likely to be Revolutionary.” Ben Bernanke, the former chairman of the Federal Reserve, did not think so. He said: “Bitcoin has serious problems.” Bank of Japan Governor Haruhiko Kuroda said: “Bitcoin has the potential to change the entire financial services system.” Christine Lagarde, president of the International Monetary Fund, said in 2017: "I think it is unwise to exclude virtual currency."

There has also been an increasing amount of research on the pricing and bubbles of digital currencies in the academic community. There is relatively little controversy about the existence of a bubble. However, the extent of the bubble is related to the intrinsic value of Bitcoin. This is a very controversial issue. Some researchers try to start with the traditional supply and demand relationship, while others analyze the price of bitcoin from the perspective of the cost of producing bitcoin, as well as public sentiment and sensation in behavioral finance. Analysis to study the price of bitcoin.

Bitcoin pricing: supply demand method and cost method

The earliest study of bitcoin prices from the supply-demand relationship was Buchholz et al. (2012), who found that, to some extent, bitcoin prices were ultimately determined by supply and demand. However, Bitcoin has a unique fixed supply: not only the total number is fixed, but it is all sent out in 2040, and the frequency and interval are also preset. Since there is no dynamic adjustment of supply, the change in demand is directly reflected in the price change, which explains the high volatility of Bitcoin's unique supply demand relationship.

The demand for bitcoin is mainly concentrated on trading and speculation. Koutmos (2018)'s research focused on the trading needs of Bitcoin and used the bivariate vector autoregression (VAR) model to prove that Bitcoin's return has a strong correlation with the number of transactions. He believes that the price of Bitcoin is not determined by the economic value of Bitcoin. If Bitcoin is really valuable, it can only be the value of an exchange intermediary. The value of this exchange intermediary is through trading. produced. The activity of the transaction can be derived by calculating the number of bitcoin transactions and the address of the bitcoin. The value of the exchange intermediary has a network effect. The more people use bitcoin, the higher its value. In his research paper, Koutmos used an econometric model to prove that fluctuations in bitcoin prices and fluctuations in bitcoin trading volume are synchronized and linked.

Ciaian el at (2016) combines all previous methods of studying the formation of bitcoin prices, using the daily price data of Bitcoin from 2009 to 2015 to study three aspects of the bitcoin price formation mechanism: 1) supply and demand; 2) Attraction and speculative demand; 3) Development of the global macroeconomic and financial system. The results show that supply and demand are the main factors determining the price. However, Ciaian's findings cannot deny the short- and long-term effects of speculative factors on prices. The impact of macro factors on bitcoin prices cannot be supported by sufficient data.

Kristoufek (2015) uses Continuous Wavelet Analysis, especially Wavelet Coherence, to find out the factors that determine bitcoin price changes, including basic economic factors, speculative factors, and technical factors. According to his research, in the long-term price, Bitcoin still conforms to the theory of demand supply balance, but in the short-term price, it is in line with the fluctuation pattern of bubble expansion and bubble burst. However, the so-called "bitcoin transaction brings value" is not significant in the long run.

Hayes (2019) did not study the demand supply direction like others. Hayes believes that Bitcoin does have real and quantifiable intrinsic value. He studied the bitcoin-specific price model by studying the mining and market demand of Bitcoin. He derives the bitcoin price by calculating the marginal cost of producing bitcoin, and in another research report, uses this model to back-test the historical price of bitcoin. His model shows that the estimated price does not differ much from the price displayed by the market. Moreover, the estimated price of the model can explain most of the changes in market prices within the statistical standards. This conclusion challenges many scholars who believe that bitcoin is worthless. Even if Bitcoin skyrocketed to a price of more than $10,000, his proposed production cost model remained valid. The bitcoin price hike that began in the second half of 2017 formed a price bubble that deviated from the price measured by his model, but from the beginning of 2018, the market price began to return to the marginal production cost calculated by the model.

Wheatley et al (2018)'s research focuses on the bitcoin price bubble, price plunging and bitcoin pricing, using the Metcalfe's Law and a Log-periodic Power Law Singularity (LPPLS) model. Research shows that the LPPLS model can provide early warning of market volatility, the probability of energy collapse and the approximate time range of plunging, and the calculation results of these models are consistent with the market reality. However, when or what caused the bitcoin price to plummet, it became the last straw to crush the camel, which was an external factor and could not be estimated from the model.

Bitcoin pricing: emotional sentiment analysis

There are still many people who think that bitcoin is more of a gimmick and an irrational speculation, so it forms a huge bubble and can last for a long time . Therefore, many researchers have studied from the direction of behavioral finance and adopted it. A large number of methods for emotional sentiment analysis.

Kristoufek (2013) should be the first scholar to study the relationship between bitcoin prices and public sentiment. She believes that Bitcoin has no fundamental value and is purely for speculative trading. The statistics of Bitcoin searched by Google or Wikipedia can be regarded as the public's sentiment index on Bitcoin. Her research shows that bitcoin prices are positively correlated with sentiment indices. Georgoula et al (2015) went one step further on this basis, using time series and public opinion analysis to discover the determinants of bitcoin prices. Instead of using Google or Wikipedia, Georgoule uses Twitter and Wikipedia to search and optimize it using machine learning algorithms. His research shows that public sentiment does have a positive relationship with the price of bitcoin. Another interesting finding is that the mining difficulty coefficient Hash Rate and price are also positively correlated.

Dastgir et al (2018) used the same Google search as other scholars as a public interest index for Bitcoin. What is different is that Dastgir uses the Copula-based Granger Causality in Distribution (CGCD) test model to test the public's interest in Bitcoin. Concerned about the causal relationship between its price and the entire year of 2013. This research article shows that there is a two-way causal relationship between the two, the only exception being the central distribution of 40% to 80%. That is to say, the two-way causal relationship only has extreme situations, or the market is very poor or excellent.

Karalevicius et al (2017) developed a method of emotional sentiment analysis for Bitcoin using natural language processing techniques. He uses a linguistic analysis based on a specific dictionary (a general and financial-related social psychology dictionary) to measure the impact of media bitcoin-related news on investor sentiment to predict bitcoin prices. The main finding of this article is that through professional media, we can predict the price of bitcoin in the semi-short term. At the same time, the Bitcoin market overreacted to the beginning of the news, causing the price to fluctuate back and forth several times to correct the previous overreaction.

How big is the bubble?

Since the birth of Bitcoin, society and academia have been questioning the true value of Bitcoin. Most people think there is a bubble in the bitcoin market. In order to test how much bubbles exist in the bitcoin market, academic research in this area continues to emerge, using a variety of different approaches from different perspectives. Academic conclusions vary and are not invulnerable. Some scholars use the mature measurement model to directly test the bitcoin market; some scholars think that the bitcoin itself is worthless after studying the value of the bitcoin itself, so the bubble will burst and the bitcoin price will inevitably collapse; however, some scholars believe that the bit The currency itself does have a certain value, its value has yet to be discovered, and the bitcoin price will not all be zero.

Cheung et al (2015), one of the first scholars to study the bitcoin market bubble, tested the bitcoin price of the Mt.Gox exchange with a very mature and mature bubble detection model Phillips-Shi-Yu (2013), trying to find out Assess the bubble in the bitcoin market. He found that there were many short-term bubbles in 2010-2014, and in the late 2011-2013, there were three major bubbles with the greatest impact, and lasted 66-106 days, the last big bubble happened just in Mt. When Gox went bankrupt. His research has aroused the curiosity of the academic world. If there is a structural break in the bitcoin price, what is the cause of the bitcoin bubble? What are the factors that can trigger a bubble burst?

Thiesa and Molnár (2018) applied Bayesian Change Point analysis to study the average return and volatility of bitcoin prices. They found that the average return and volatility of bitcoin prices changed very frequently. Based on the time when structural changes occur, the study divides the time series of bitcoin into segments and classifies the characteristics of each time structure change into a large class. After this treatment, there are several major types of structural changes accompanied by a positive average return, while only one large category of structural changes has a negative average return. All classes exhibit a commonality of high volatility with high returns, except that the most volatile class returns are negative, which is the only negative return category of all major classes. The research by Thiesa and Molnár provides a good starting point for future academics to use the metrological detection model to study the variation law of bitcoin price time series in more depth.

In addition to using the bubble and structural change test models to directly detect bitcoin price time series, it is worth exploring the fundamental value of bitcoin. Because if Bitcoin has real intrinsic value, then whether bitcoin has a bubble and how large its bubble is is easily identified.

Scholars have tried to study from the perspective of Long Memory. Mensi et al. (2018) used four general econometric models (GARCH, FIGARCH, FIAPARCH, and HYGARCH) to study the effects of structural changes in Bitcoin and Ethereum on the time-delayed memory of two currencies. They found that considering the factors of time-delayed memory, the level of persistence of returns and volatility was reduced. Moreover, the FIGARCH model with structurally variable variables can best predict market returns. These findings allow investors to better understand the behavior patterns of the market and use the model's ability to predict returns and time-lapse memories to pursue higher return on investment. Market concerns about such risks are obvious due to the risk of price volatility due to automatic adjustments or influencing events. At the same time, it can be seen that institutional change means that the bubble originates from a place where the price does not match the basic value. Through this research we can also find that when the market is afraid of risk, automatically corrected or affected by some events, the structural changes that occur under these circumstances indicate that the bitcoin bubble is more than the basic value of Bitcoin. caused.

In the history of bitcoin price development, there are two major bitcoin booming periods, 2013 and 2017. The bitcoin price in 2017 has skyrocketed even more. In 2013, many scholars have already determined that bitcoin is a bubble. If they are right, then in 2017, the bubble will become more serious. MacDonell (2014) is an article that studies the period of bitcoin boom in 2013. For the first time, the study used the ARMA (Autoregressive Moving Average) model to analyze the bitcoin transaction price and found that the price of bitcoin is related to the CBOE volatility index. The findings suggest that the main driver of bitcoin price increases during this period is that speculators can't find speculative opportunities in traditional financial markets and turn to bitcoin speculation. They don't value the underlying value of Bitcoin. MacDonell's research also used the Log-Periodic Power Law (LPPL) model to try to predict the timing of the market crash, and found that the model could predict the December 2013 bitcoin plunge. Therefore, the study claims that the LPPL model is a promising and valuable tool for studying the behavior of digital currency bubbles. His research ignited academic interest in the model and used the latest data to test the validity of the model. Later, some scholars used a similar model, Log-Periodic Power Law Singularity (LPPLS), to study the bubble and price of Bitcoin.

Fry and Cheah (2016) are very interested in digital currency, and have published several related articles, one of which talks about the reverse bubble and impact of the bitcoin market. In this article, they say that economic physics plays a vital role in the global digital money market, and they use research to prove the speculative bubbles of Bitcoin and Ripple, and the mutual spillover between the two markets. . The research report also pointed out that the impact of negative events is complex and multifaceted, and sometimes it does not necessarily have a negative impact on prices. The event impact mechanism proposed in the report is worth studying because everyone should be interested in predicting the price of bitcoin. In another research report, Cheah and Fry (2015) argue that the price of bitcoin presents a significant speculative component, and that bitcoin is a speculative bubble. They used experimental evidence to prove that the value of Bitcoin is zero. However, Corbet et al. (2017) believe that bitcoin prices above $1,000 are the bubble stage.

From the above discussion of Bitcoin pricing and bubbles, we can see the relationship between the two: the bubble is generated because the price in the market is much higher than the true value of Bitcoin . In other words, in order to detect and evaluate bubbles, we must know the true value of Bitcoin . However, the commodity can calculate the basic value according to its thoroughfare. The securities can be used to calculate the value according to the products and services provided by the issuing company. The basic value of the digital currency is not easy to calculate. In addition to the use of blockchain technology, digital currency has much intrinsic value and has been controversial. According to John Maynard Keynes (1936), the academic scholar who first studied bubbles, he believes that the bubble in financial markets stems from irrational investors. However, Brunnermeier & Abreu (2003) believes that the reason that the bubble can last for a long time is that rational investors do not go short, but instead participate in the process of pushing up the bubble. As Richard Thaler, a Nobel laureate in economics and behavioral finance expert, says, many human behaviors are irrational. If we all attribute the bubble to irrational behavior, then the bitcoin bubble phenomenon is even more difficult to explain. The analysis of the emotional market analysis of the digital money market discussed above is based on the rational and irrational behavioral research of investors, thus helping us to better understand the formation mechanism of the digital money market bubble.

Although there have been many research attempts and models for the Bitcoin bubble, these studies are only the beginning. In fact, there are many mature and tested financial markets and real estate bubbles that can be used for reference. An academic article by Weites et al. (2010) that discusses the measurement model for detecting foam provides a lot of insights that are worth learning and applying in the study of digital currency bubbles. This article evaluates the strengths and weaknesses of current popular bubble detection models, including Variance Bounds, West's (1987), Diba & Grossman's (1988a), Froot & Obstfeld's (1991), Wu's (1997), Van Norden's (1996) and Hall & Solas' ( 1993), Phillips, Wu & Yu ́s (2007) These eight models. Weites et al. believe that these models are more or less disadvantageous, and that these models cannot effectively distinguish the price changes due to the change of the underlying value or the existence of the bubble. The measurement model relies too much on the assumptions set by the person using the model, and the results of the model run cannot lead to factors that truly reflect the foam structure. So they proposed three new bubble test models: the Dividends' Growth Expectation Test, the Price/Earnings-Ratio Tests, and the Option Based Tests. These three models are relatively more capable of detecting the presence of foam, but still cannot quantify the existence of the expressed bubble.

There are many academic discussions on qualitative and quantitative research bubbles . White (1990, p. 67) questions the rationality of quantitative research bubbles. He said that the measurement model cannot distinguish whether the price increase is due to the value increase or because the bubble is more serious. He is very far-sighted at this point. We find that in a large number of academic articles on the study of digital currency bubbles, researchers are more likely to determine whether there is a bubble in the digital currency market, but it is more difficult to determine which part of the price is a bubble. As for the relationship between the underlying value of the digital currency and the bubble That is more challenging.

In short, how big a bubble is in the digital currency is the most concerned and controversial topic in the academic world, and it will continue to be debated. Maybe until one day, everyone will look back and see if Bitcoin is in a bubble debate. It has been tenacious for 10 years, 20 years, or even longer. What is the point of discussing whether it has a bubble?

After the price is finished, click to check the first few articles in this series . The next risk article – the wind is rising , so stay tuned .

=== References: ====

· Buchholz, M., J. Delaney, J. Warren, and J. Parker. 2012. “Bits and Bets, Information, Price Volatility, and Demand for BitCoin, Economics 312.” www.bitcointrading.com/pdf/bitsandbets. Pdf.

· Koutmos, Dimitrios (2018). Bitcoin returns and transaction activity. Economics Letters. Volume 167, June 2018, Pages 81-85

· Ciaian, P., Rajcaniova, M., and Kancs d'A. (2016). The economics of BitCoin price formation. Applied Economics, 2016 – Taylor & Francis. Volume 48, 2016 – Issue 19

· Kristoufek, Ladislav (2015). What Are the Main Drivers of the Bitcoin Price? Evidence from Wavelet Coherence Analysis. PLoS ONE 10(4): e0123923

· Hayes, Adam S. (2019) Bitcoin price and its marginal cost of production: support for a fundamental value, Applied Economics Letters, 26:7, 554-560, DOI: 10.1080/13504851.2018.1488040

· Spencer Wheatley, Didier Sornette, Tobias Huber, Max Reppen, and Robert N. Gantner, Are Bitcoin Bubbles Predictable? Combining a Generalized Metcalfe's Law and the LPPLS Model, Swiss Finance Institute, c/o University of Geneva, Switzerland. 2018

· Kristoufek, L., (2013). Bitcoin meets google trends and wikipedia: quantifying the relationship between phenomena of the internet era. Sci. Rep. 3, 3415.

Georgoula, Ifigeneia; Pournarakis, Demitrios; Bilanakos, Christos; Sotiropoulos, Dionisios N.; and Giaglis, George M., (2015). "Using TimeSeries and Sentiment Analysis to Detect the Determinants of Bitcoin Prices". MCIS 2015 Proceedings. . http://aisel.aisnet.org/mcis2015/20

· Dastgir, Shabbir & Demir, Ender & Downing, Gareth & Gozgor, Giray & Lau, Chi Keung. (2018). The Causal Relationship between Bitcoin Attention and Bitcoin Returns: Evidence from the Copula-based Granger Causality Test. Finance Research Letters. 10.1016/j.frl.2018.04.019.

· Karalevicius, Vytautas & Degrande, Niels & De Weerdt, Jochen (2018). "Using sentiment analysis to predict interday Bitcoin price movements", The Journal of Risk Finance, Vol. 19 Issue: 1, pp.56-75, https: //doi.org/10.1108/JRF-06-2017-0092

· Cheung, A., Roca, E., Su, J.-J., (2015). Crypto-currency bubbles: An application of the Phillips-Shi-Yu (2013) methodology on Mt.Gox bitcoin prices. Applied Economics , 47(23), 2348–2358. doi:10.1080/00036846.2015.1005827

· Thiesa, Sven and Molnár, Peter (2018). Bayesian change point analysis of Bitcoin returns. Finance Research Letters. Volume 27, pp. 223-227.

· Mensi, W., Al Yahyaee, K., & Kang, SH (Accepted/In press). Structural breaks and double long memory of cryptocurrency prices: A comparative analysis from Bitcoin and Ethereum. Finance Research Letters. https://doi .org/10.1016/j.frl.2018.07.011

· MacDonell, A., (2014). Popping the Bitcoin bubble: an application of log-periodic power law modelling to crypto currency. Preprint.

· Fry, John and Cheah, Eng-Tuck (2016). Negative bubbles and shocks in cryptocurrency markets, International Review of Financial Analysis, Volume 47, 2016, Pages 343-352, ISSN 1057-5219, https://doi.org /10.1016/j.irfa.2016.02.008.

· Cheah, ET, Fry, J., (2015). Speculative bubbles in Bitcoin markets? An empirical investigation into the fundamental value of Bitcoin. Economics Letters, Volume 130, 2015, Pages 32–36. ISSN 0165-1765, https: //doi.org/10.1016/j.econlet.2015.02.029.

· Corbet S., Lucey B., and Yarovaya L. (2018). Datestamping the Bitcoin and Ethereum bubbles. Finance Research Letters Volume 26, September 2018, Pages 81-88, https://doi.org/10.1016/j. Frl.2017.12.006.

We will continue to update Blocking; if you have any questions or suggestions, please contact us!