» » » A REGRESSION ANALYSIS OF CRYPTOCURRENCY INFLUENCE ON THE US DOLLAR
New publications!

A REGRESSION ANALYSIS OF CRYPTOCURRENCY INFLUENCE ON THE US DOLLAR

ALINA KHVAN,
LOMONOSOV MOSCOW STATE UNIVERSITY, ECONOMICS FACULTY
KHVANALINA17@GMAIL.COM

Abstract: The study shows that the proliferation of cryptocurrencies, bitcoins in particular, weakens the US dollar and, therefore, may negatively affect the country’s economy. This raises the need to regulate the Bitcoin system.

Keywords: cryptocurrency, bitcoin, exchange rate, dollar, US, analysis.

JEL Classification: E42, G17, C5

 

KHVAN, ALINA (2016) "A REGRESSION ANALYSIS OF CRYPTOCURRENCY INFLUENCE ON THE US DOLLAR". Journal of Russian Review (ISSN 2313-1578), VOL. 1(4), 8-12.

 

1. Introduction

In recent years, the interest in cryptocurrencies has been growing, which includes bitcoins as the most popular type of cryptocurrency. The growing popularity of bitcoins comes along with the problems associated with them: the government cannot monitor all transactions with cryptocurrency and therefore, cannot control cryptocurrency as they do transactions in conventional currencies. As a result, a certain part of the monetary system operates outside government regulation, threatens the stability of financial systems and financial sovereignty of countries. Legislations of many countries have already banned or restricted the use of bitcons. However, we should not consider bitcoins only as a threat, as it seems impossible to eliminate them. Many experts understand this, and, according to the President of the largest commercial Bank in Russia, Sberbank, German Gref, virtual money and technology can make a revolution in the system of Bank payments [1].

This suggests that it is crucial to understand how the rise of bitcoins may affect the financial sector, so that governments could effectively respond to their spread and use them for purposes of their countries’ development.

The goal of this study is to assess the impact of the circulation of bitcoins on the US dollar. The American currency is selected because Bitcoins are exchanged only for USD.

To achieve the objective of the study we will look at a simplified model of money circulation. We assume that there is only one country with a well-developed currency circulation. Then, with the emergence of bitcoins the competition between the two currencies begins: the national currency competes with bitcoins. It is obvious that not only economic agents, but also ordinary citizens tend to give priority to the currency that is more profitable. Bitcoins may turn out to be such, because they may help to avoid paying taxes. In this case, the process of gradual replacement of the national currency by bitcoins may begin.

Let us complicate the situation and assume that only two countries exist in the world; one has a strong and stable currency, while the other has a weak one. If cryptocurrency comes along in such a world it will start to replace the money in the country with weak currency. On the contrary, in the country where the currency is strong, bitcoins will not replace the existing money supply, but most likely, they will begin to substitute its gain.

Based on this conclusion let us formulate the hypothesis of the study: since the dollar is the world reserve currency, bitcoins could replace the gain of the US money supply. Such a substitution may definitely weaken the national currency. To evaluate of the phenomenon of «weakening national currency» the exchange rate of the U.S. dollar against the Euro (EUR/USD) is used. The dependencies are assumed as follows:

  1. The more bitcoins there are in circulation, the less the content of the dollar in one Euro (i.e. the exchange rate EUR/USD is higher).
  2. As the number of bitcoins in circulation increases, their impact on the US economy becomes more significant.
  3. In addition, it is interesting to see the influence of the market price of bitcoins and their expected quantity on the EUR/USD exchange rate.

2. Literature review

231 works have been found upon the request by keyword «bitcoin» in the SSRN electronic library (www.ssrn.com), three of which are of interest for this study.

In the paper “Can Bitcoin Become a Major Currency?” [2]  (William J. Luther, Lawrence H. White, 2014) the authors describe the way of development of bitcoins and analyze whether the cryptocurrency is to become the primary means of payment for goods and services, replacing the more familiar world currency.

The authors of “Can We Stabilize the Price of a Cryptocurrency?: Understanding the Design of Bitcoin and Its Potential to Compete with Central Bank Money” [3]  (with M. Iwamura, Y. Kitamura, T. Matsumoto, K. Saito, 2014) also regard cryptocurrency as competing with other currencies. In their opinion cryptocurrency is not growing even faster because of high volatility of its prices.

In “An Analysis of Bitcoin Exchange Rates”[4]  (Jacob B. Smith, 2016), the author wonders whether the exchange rates of cryptocurrencies have an impact on the economy, or, on conversely, the economy affects the prices of bitcoins, causing its volatility.

In the Russian scientific electronic library E-library (www.elibrary.ru) 102 publications were found upon the request by keyword «bitcoin». They all mainly provide a legal analysis of the cryptocurrency. None of the works analyzes the impact of bitcoin on the economy.

We also found the paper «Analysis of Bitcoin Transaction Flows to Reveal Usage and Geographic Patterns» (Bissessar, 2013), which recognises the importance of the study of bitcoin transactions and their impact on the economy, and analyzes geographical patterns of cryptocurrency flows.

3. Methodology of research

Regression analysis is used to achieve the goal of this research. The research consists of the following stages: first, identification of key hypotheses, second, the choice of variables for the model; next, a specification of equation, then analysis of the given model and, finally, the main findings.

The sample used for the research comprises of 1832 observations, which cover the period from January 2009 to April 2016, that is, from the year when bitcoins appeared to the present time.

To develop the model and avoid biased estimates of the coefficients, the factors assumed to affect the money supply are selected.

The designated covariates and the dependent variable are identified as:

Dependent variable Description Sources
eur_usd The EUR/USD exchange rate
(The content of USD in one EUR) 
Federal Reserve Bank of St. Louis[5] 
Regressors Description Sources
btc_amount The total value of bitcoins in circulation, already mined, or the current number of bitcoins Blockchain Info [6] 
btc_predict The estimated value of transactions, USD mln. Blockchain Info [7] 
btc_price Average market price in USD in the leading bitcoin markets Blockchain Info [8] 
five_year_inflation_exp The US five-year estimated inflation Federal Reserve Bank of St. Louis[9] 
interest_rate US key interest rate U.S. Department of Treasury[10] 
unemp_rate US unemployment rate Bureau of Labor Statistics[11] 
public_debt США US public debt TreasuryDirect[12] 

Table 1

Source: Developed by the author

4. Calculations

Firstly, in the econometrics package Gretl we built the regression of l_eur_usd variable on the independent variables, for the period from 2009 to 2016. Let us define it a model (1). 1832 observations are used to build the model. We received the correlations as follows:

All independent variables in the model have a highly significant impact on the EUR/USD rate with a probability of 99%, except for the variable btc_predict, which is also significant with a probability of 90% (see Appendix 1). It can be seen that only the US key interest rate and the US public debt have inverse correlations with the USD rate among all other variables.

The coefficient of determination (R2) of the model is quite high, and is equal to 0.783, which means that 78.3% of the values of the dependent variable are explained by this model.

All the observations are divided into 3 parts, to see how the Bitcoin market has had an impact on the US economy as the Bitcoins spread. For each of the three groups of observations l_eur_usd regression variable to the independent variables are built - the model (2), (3) and (4) (Appendix1). The sample model (2) covers the period of 2009 and 2010, a period of a very small spread of Bitcoins. The sample model (3) covers the period from 2011 to 2013, when Bitcoins started to be actively used for speculation. The last period from the beginning of 2014 to April 2016 was assigned to the model (4). This period describes the impact that Bitcoins had on the US economy. During that period, the speculation in Bitcoins declined, and the Bitcoin payment function took a center stage.

It is clearly seen, that in 2009-2010 the correlation between the amount of Bitcoins in circulation and the EUR/USD rate was negative, and the variable had no significant effect on the USD. In model (2) the correlation changed to positive, but the variable still did not have a significant impact on the EUR/USD. In model (3) while the correlation remained positive, the spread of Bitcoins had a significant impact on the EUR/USD with a 90% probability.

5. Findings

  • The first hypothesis that the more Bitcoins there are in circulation, the less the content of the dollar in one euro is (ie. the higher the EUR/USD exchange rate), has been confirmed. Other things being equal, as the number of Bitcoins in circulation (emission) increases by 1 million, the EUR/USD rate rises by 6.1%, that is, the amount of dollars to buy 1 Euro increases. This suggests that the distribution of Bitcoins weakens the US dollar.
  • The second hypothesis that, with an increase in the number of Bitcoins in circulation their influence on the US economy grows and becomes more significant has also been confirmed. This is evidenced by the fact that cryptocurrency had no significant effect on the US dollar during the first reporting period 2009-2010, whereas in the last period of 2014-2016 cryptocurrency had significant influence with 90% probability. This finding suggests, that the impact of the cryptocurrency on the US economy will grow in the future.
  • The Bitcoin price has also a substantial impact on the dollar. Ceteris paribus, while increasing the Bitcoin price per its unit the US dollar rises by 0.7%.

6. References

  1. Shiva P. Bissessar « Analysis of Bitcoin Transaction Flows to Reveal Usage and Geographic Patterns » // MSc in Information Security at University College. London. 2013.
  2. William J. Luther, Lawrence H. White Can Bitcoin Become a Major Currency? // George Mason University - Department of Economics, 2014
  3. M. Iwamura, Y. Kitamura, T. Matsumoto, K. Saito Can We Stabilize the Price of a Cryptocurrency?: Understanding the Design of Bitcoin and Its Potential to Compete with Central Bank Money // Institute of Economic Research Hitotsubashi University Kunitachi, 2014
  4. Jacob B. Smith An Analysis of Bitcoin Exchange Rates // University of Houston, 2016
  5. Blockchain.info. URL: https://blockchain.info/ru/
  6. The database of Federal Reserve Bank of St. Louis. URL: https://research.stlouisfed.org/fred2/
  7. The database of United States Department of the Treasury. URL: https://www.treasury.gov/Pages/default.aspx
  8. The database of United States Department of Labor. URL: http://www.bls.gov/home.htm

7. Appendices

Appendix 1

Regressor Model (1)
2009-2016
Model (2)
2009-2010
Model (3)
2011-2013
Model (4)
2014-2016
intercept 0.811 (***)
(0.2)
-0.316
(0.349)
3.61 (***)
(0.369)
-1.713
(1.058)
btc_amount 0.061 (***)
(0.011)
-0.017
(0.029)
0.009
(0.041)
0.09 (*)
(0.048)
btc_predict 0.007 (** )
(0.003)
0.24 (* )
(0.136)
-0.001
(0.002)
-0.038
(0.026)
btc_price 0.0002 (***)
(0.023)
0.198 (** )
(0.085)
0.0001 (***)
(0.082)
0.0001 (** )
(0.00005)
five_year_inflation_exp 0.081 (***)
(0.012)
0.029 (** )
(0.014)
0.067 (***)
(0.013)
0.059 (* )
(0.033)
interest_rate -0.15 (***)
(0.05)
-0.29 (***)
(0.056)
  0.021
(0.136)
unemp_rate 0.054 (***)
(0.008)
0.062 (***)
(0.018)
-0.103 (***)
(0.024)
0.218 (***)
(0.067)
public_debt -0.11 (***)
(0.02)
0.008
(0.045)
-0.173(***)
(0.023)
-0.043
(0.04)
         
R2 0.783 0.721 0.583 0.883
N – number of observations 1832 504 753 575

Source: Developed by the author in econometric package Gretl

Links

[1] The prohibition of bitcoin in Russia questioned the technology blockchain // Vedomosti URL: https://www.vedomosti.ru/technology/articles/2016/02/29/631755-skovannie-odnoi-tsepyu (Assessed on 4th April 2016)

[2]William J. Luther, Lawrence H. White Can Bitcoin Become a Major Currency? // George Mason University - Department of Economics, 2014

[3] M. Iwamura, Y. Kitamura, T. Matsumoto, K. Saito Can We Stabilize the Price of a Cryptocurrency?: Understanding the Design of Bitcoin and Its Potential to Compete with Central Bank Money // Institute of Economic Research Hitotsubashi University Kunitachi, 2014

[4] Jacob B. Smith An Analysis of Bitcoin Exchange Rates // University of Houston, 2016

[5] FRED URL: https://research.stlouisfed.org/fred2/series/M2 (Assessed on 21th April 2016)

[6] URL: https://blockchain.info/ru/charts/total-bitcoins (Assessed on 15th April 2016)

[7] URL: https://blockchain.info/ru/charts/estimated-transaction-volume-usd (Assessed on 15th April 2016)

[8] URL: https://blockchain.info/ru/charts/market-price (Assessed on 15th April 2016)

[9] URL: https://research.stlouisfed.org/fred2/series/T5YIFR (Assessed on 21th April 2016)

[10] URL: http://1.usa.gov/1SUlB9u (Assessed on 29th April 2016)

[11] URL: http://data.bls.gov/pdq/SurveyOutputServlet (Assessed on 29th April 2016)

[12] URL: http://1.usa.gov/21HI0Jm (Assessed on 1st May 2016)

 

скачать dle 10.2 Авто Тюнинг кузова