Research Area

24 January 2020 1

Digital Finance and

Blockchain

Cryptocurrencies and

Asset Pricing

Cryptocurrencies and

Financial Risk

Blockchain Consensus

Protocols and Energy Efficiency

Crowdfunding (ICO, STO, IEO)

Sentiments, Scams, and Frauds

24 January 2020 1Vaasan yliopisto | Niranjan Sapkota

Volatility Spillovers (G10 Currencies and Bitcoin)

Predicting Cryptocurrency Defaults (Seminar( Aalto, Hanken, Vaasa, Jyväskylä))Media Coverage (Forbes, Vaasa Insider)

Conference (Finance, Property, Technology and the Economy- UniSA)

Cryptocurrencies and Momentum (EL)

Assets Market Equilibria(Privacy Vs. Non-privacy Coins)

Cryptocurrenciesand Liquidity

Cryptocurrencies and Technical Trading Strategies

(FRL)

Asset Market Equilibria in Cryptocurrency Markets:

Evidence from a Study of Privacy and Non-Privacy Coins

-Niranjan Sapkota and Klaus Grobys

FinTech Conference UniSA, Dec 2-3, 2019

Outline

Purpose of the study: to find whether asset market equilibria incryptocurrency market exist.

Via: Johansen’s (1991, 1992, 1994, 1995) multivariate cointegrationmethodology to explore whether or not asset market equilibria in linewith Engle and Granger’s (1987) cointegration theory exist.

Research Question: ” Do privacy coins form a distinct submarketwithin the cryptocurrency market?”

Result: Privacy coins and non-privacy coins exhibit two distinctmarket equilibria

24 January 2020 3Vaasan yliopisto | Niranjan Sapkota

24 January 2020 4

Bitcoin and Privacy

24 January 2020 5

Bitcoin and Privacy…..contd

24 January 2020 6

Bitcoin and Privacy…..contd

• How private are the cryptocurrencies like Bitcoin?

24 January 2020 7

(Source: Goldfeder et al., 2017, MIT Technology Review)

Vaasan yliopisto | Niranjan Sapkota

24 January 2020 8

The Fungibility Problem (Clean Vs. Dirty Coins)

=?

Traditional Currency(US Dollar)

Cryptocurrency(non-privacy coin)

Vaasan yliopisto | Niranjan Sapkota

24 January 2020 9

=?Cryptocurrency(Privacy coin)

=?

The Fungibility Problem (Clean Vs. Dirty Coins)

24 January 2020 10

Non-privacy Coin+

Dark Web

Privacy Coin+

World Wide Web

24 January 2020 11

Prior Research on Cryptocurrencies and Privacy

• Androulaki, E., Karame, G.O., Roeschlin, M., Scherer, T. and Capkun, S.,2013, April. Evaluating user privacy in bitcoin. In InternationalConference on Financial Cryptography and Data Security (pp. 34-51).Springer, Berlin, Heidelberg.

• Goldfeder, S., Kalodner, H., Reisman, D. and Narayanan, A., 2018. Whenthe cookie meets the blockchain: Privacy risks of web payments viacryptocurrencies. Proceedings on Privacy EnhancingTechnologies, 2018(4), pp.179-199.

• Khalilov, M.C.K. and Levi, A., 2018. A survey on anonymity and privacy inbitcoin-like digital cash systems. IEEE Communications Surveys &Tutorials, 20(3), pp.2543-2585.

• Kumar, A., Fischer, C., Tople, S. and Saxena, P., 2017, September. Atraceability analysis of monero’s blockchain. In European Symposium onResearch in Computer Security (pp. 153-173). Springer, Cham.

• Foley, S., Karlsen, J.R. and Putniņš, T.J., 2019. Sex, drugs, andbitcoin: How much illegal activity is financed throughcryptocurrencies?. The Review of Financial Studies, 32(5),pp.1798-1853.

• Brenig, C., Accorsi, R. and Müller, G., 2015, May. EconomicAnalysis of Cryptocurrency Backed Money Laundering. In ECIS.

• Kethineni, S., and Y. Cao, 2019, The Rise in Popularity ofCryptocurrency and Associate Criminal Activity. InternationalCriminal Justice Review, forthcoming.

24 January 2020 12

Prior Research on Cryptocurrencies and Illegal Activities

24 January 2020 13

Table 1 Top 10 Privacy and Non-privacy Coins.

Panel A : Top 10 Non-Privacy Coins

S.No Non-Privacy Coin Symbol January 3, 2016 Coin Rank /572 Coins Capitalization ($) December 30, 2018

Coin Rank/2073 Coins Capitalization ($) 3 Years’ Market Capitalization Growth%

Bitcoin BTC 1 6467437080 1 67475512827 943.31 1 Ripple XRP 2 201799631 2 15076740856 7371.14 2 Litecoin LTC 3 152873521 7 1912263648 1150.90 3 Ethereum ETH 4 73843278 3 14560066114 19617.52 4 Dogecoin DOGE 6 14940681 23 278470304 1763.84 5 Peercoin PPC 7 9756959 181 14254314 46.10 6 BitShares BTS 8 8591688 44 106520276 1139.81 7 Stellar XLM 9 8436465 6 2250048215 26570.51 8 Nxt NXT 10 6863998 113 29426627 328.71 9 MaidSafeCoin MAID 11 6789470 65 61335019 803.38

10 NameCoin NMC 12 6073338 220 10480336 72.56 Average (Excluding Bitcoin) 5886.45

Panel B: Top 10 Privacy Coins S.No Privacy Coin Name

1 Dash DASH 5 19794713 15 698091183 3426.65 2 Bytecoin BCN 14 5582979 38 134517567 2309.42 3 Monero XMR 15 5295952 13 806939516 15136.91 4 DigitalNote XDN 51 447057 252 8565009 1815.87 5 CloakCoin CLOAK 125 201995 318 6249733 2994.00 6 Aeon AEON 137 137088 393 4475790 3164.90 7 NavCoin NAV 142 121805 213 11127542 9035.54 8 Verge XVG 149 109968 43 110307808 100209.00 9 Stealth XST 161 8352 515 2539074 30300.79

10 Prime-XI PXI 322 8889 1701 4236 -52.35 Average 16834.07

Note: This table reports the top 11 non-privacy coins (including Bitcoin) and top ten privacy coins based on their market capitalization as of January 3, 2016. There were 572 cryptocurrencies available (including both privacy and non-privacy coins) as of January 3, 2016, and 2073 coins as of December 30, 2018. Coin Rank shows the position based on a coin’s market capitalization. Three Years’ Market Capitalization Growth shows the percentage growth in market capitalization from January 3, 2016 until December 30, 2018. Panel A shows the top 11 non-privacy coins and Panel B shows the top ten privacy coins in terms of market capitalization (Source: coinmarketcap.com/historical/).

24 January 2020 14

To test the order of cointegration, we employ the trace test for each submarket of

cryptocurrencies, given by:

𝐿𝐿𝐿𝐿(𝑟𝑟0) = −𝑇𝑇∑ 𝑙𝑙𝑙𝑙𝑙𝑙�1 − 𝜆𝜆𝑗𝑗 �𝐾𝐾𝑗𝑗=𝑟𝑟0+1 , (2)

where 𝐾𝐾 = 10, and 𝜆𝜆𝑗𝑗 are the eigenvalues obtained by applying RR regression techniques to

the fully unrestricted Vector-Error Correction model (VECM) (Johansen 1991, 1992a, 1992b,

1994, 1995) given by:

Δ𝒚𝒚𝑙𝑙 ,𝑡𝑡 − 𝜇𝜇𝑙𝑙 ,1 = 𝚷𝚷𝑙𝑙 �𝒚𝒚𝑙𝑙 ,𝑡𝑡−1 − 𝜇𝜇𝑙𝑙 ,0 − 𝜇𝜇𝑙𝑙 ,1(𝑡𝑡 − 1)� + ∑ 𝚪𝚪𝑙𝑙 ,𝑗𝑗 �Δ𝒚𝒚𝑙𝑙 ,𝑡𝑡−1 − 𝜇𝜇𝑙𝑙 ,1� + 𝒖𝒖𝑙𝑙 ,𝑡𝑡𝑝𝑝−1𝑗𝑗=1 , (3)

where the 10x1 vector Δ𝒚𝒚𝑙𝑙,𝑡𝑡 contains the log-returns of privacy coins if 𝑙𝑙 = 1 or non-privacy

coins if 𝑙𝑙 = 2.1 Moreover, 𝜇𝜇𝑙𝑙 ,0 and 𝜇𝜇𝑙𝑙 ,1 denote constant and trend, 𝚷𝚷𝑙𝑙 and 𝚪𝚪𝑙𝑙 ,1, 𝚪𝚪𝑙𝑙 ,2, …, 𝚪𝚪𝑙𝑙 ,𝑝𝑝−1

are KxK parameter matrices. The trace test tests the sequence of hypotheses given by:

𝐻𝐻0(0): 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝚷𝚷𝑙𝑙) = 0 versus 𝐻𝐻1(0): 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝚷𝚷𝑙𝑙) > 0,

𝐻𝐻0(1): 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝚷𝚷𝑙𝑙) = 1 versus 𝐻𝐻1(0): 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝚷𝚷𝑙𝑙) > 1,

⋮

𝐻𝐻0(𝐾𝐾 − 1): 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝚷𝚷𝑙𝑙) = 𝐾𝐾 − 1 versus 𝐻𝐻1(𝐾𝐾 − 1): 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝚷𝚷𝑙𝑙) = 𝐾𝐾.

Methodologies

24 January 2020 15

Table 2 ADF tests for privacy and non-privacy coins.

Privacy coins Non-privacy coins

Model 1 Model 2 Model 1 Model 2 Coin Interceptª Lagsᵈ Intercept

and trendᵇ

Lagsᵈ Coin Interceptª Lagsᵈ Intercept and

trendᵇ

Lagsᵈ

DASH -1.79 0 0.25 0 XRP -0.99 2 -1.19 2 BCN -1.34 1 -0.90 1 LTC -1.13 0 -0.23 0 XDN -1.39 0 -0.91 0 ETH -2.93** 0 -0.63 0 XMR -2.16 0 -0.15 0 DOGE -1.30 0 -1.41 0 CLOAK -1.51 2 -0.72 2 PPC -1.43 0 -1.18 0 AEON -1.46 1 -0.57 1 BTS -1.24 0 -0.41 0 XST -1.49 1 -1.45 1 XLM -0.93 1 -1.45 1 PXI -1.44 2 -1.16 2 NXT -1.37 0 -0.62 0 NAV -1.86 4 -0.60 4 MAID -2.86** 0 -1.69 1 XVG -1.23 4 -1.27 4 NMC -1.65 1 -1.59 1 BTC -1.31 0 -0.06 0 Note: This table reports the results for Augmented Dickey Fuller tests of the daily price series in logs for privacy and non-privacy coins. Model 1 accounts for an intercept in the test regression, whereas model 2 accounts for both an intercept and trend term. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations. **Statistically significant on a 5% level. ª Critical values for 10%, 5% and 1% significance levels are -2.57, -2.86 and -3.44. ᵇ Critical values for 10%, 5% and 1% significance levels are -3.13, -3.41 and -3.97. ᵈ Lag-order is chosen by using the Schwarz info criterion. The maximum lag length is chosen by default is 21.

24 January 2020 16

Table 3 Trace test for cointegration employing privacy coins.

Hypothesized Trace 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.** None * 0.0627 283.1797 273.1889 0.0174

At most 1 0.0502 212.5937 228.2979 0.2095 At most 2 0.0381 156.4145 187.4701 0.5996 At most 3 0.0249 114.0723 150.5585 0.8223 At most 4 0.0238 86.63982 117.7082 0.7982 At most 5 0.0211 60.38928 88.80380 0.8456 At most 6 0.0153 37.17442 63.87610 0.9241 At most 7 0.0098 20.37189 42.91525 0.9535 At most 8 0.0050 9.687908 25.87211 0.9376 At most 9 0.0039 4.214442 12.51798 0.7109

Note: This table reports the results for the trace test for cointegration applied to a set of ten privacy coins exhibiting the highest market capitalization as of Jan 3, 2016. The test statistic allows for linear deterministic trend in data, that is, an intercept and trend in cointegration equation but no intercept in the Vector-Autoregression. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations. Trace test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

24 January 2020 17

Table 4 Trace test for cointegration employing non-privacy coins excluding Bitcoin.

Hypothesized Trace 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.** None * 0.0636 289.7541 273.1889 0.0079

At most 1 0.0493 218.0979 228.2979 0.1342 At most 2 0.0334 163.0434 187.4701 0.4297 At most 3 0.0328 125.9784 150.5585 0.5074 At most 4 0.0231 89.62467 117.7082 0.7146 At most 5 0.0166 64.13613 88.80380 0.7299 At most 6 0.0163 45.94574 63.87610 0.6029 At most 7 0.0122 28.07154 42.91525 0.6178 At most 8 0.0092 14.69995 25.87211 0.5995 At most 9 0.0042 4.641440 12.51798 0.6484

Note: This table reports the results for the trace test for cointegration applied to a set of ten non-privacy coins (excluding Bitcoin) exhibiting the highest market capitalization as of Jan 3, 2016. The test statistic allows for a linear deterministic trend in data, that is, an intercept and trend in the cointegration equation but no intercept in the Vector-Autoregression. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations. Trace test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

24 January 2020 18

Table 5 Trace test for cointegration employing non-privacy coins including Bitcoin.

Hypothesized Trace 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.** None * 0.0724 345.4557 322.0692 0.0042

At most 1 0.0540 263.4921 273.1889 0.1196 At most 2 0.0371 202.9953 228.2979 0.3928 At most 3 0.0330 161.7328 187.4701 0.4628 At most 4 0.0288 125.1233 150.5585 0.5323 At most 5 0.0257 93.25557 117.7082 0.5990 At most 6 0.0165 64.90058 88.80380 0.7028 At most 7 0.0158 46.75482 63.87610 0.5645 At most 8 0.0127 29.41653 42.91525 0.5370 At most 9 0.0099 15.51272 25.87211 0.5321 At most 10 0.0043 4.671033 12.51798 0.6441

Note: This table reports the results for the trace test for cointegration applied to our set of non-privacy coins

including Bitcoin. The test statistic allows for linear deterministic trend in data, that is, an intercept and trend in the cointegration equation but no intercept in the Vector-Autoregression. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations.

Trace test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values

24 January 2020 19

Results

Table 6 Vector-Error Correction model estimates using privacy coins.

i Coin 𝜷𝜷�𝒊𝒊 𝜶𝜶�𝒊𝒊 1 DASH 1

( ̶ ) 2.8E-03 (1.42)

2 BCN 0.82** (2.28)

-1.5E-03 (-0.40)

3 XDN -0.46 (-1.01)

4.9E-03 (1.42)

4 XMR 3.75*** (5.74)

2.7E-03 (1.18)

5 CLOAK 0.04 (0.15)

2.0E-03 (0.45)

6 AEON -0.63 (-1.50)

5.2E-03 (1.39)

7 XST 1.36*** (4.03)

-1.0E-03 (-0.24)

8 PXI -0.49* (-1.78)

5.1E-03 (0.85)

9 NAV -3.24*** (7.52)

3.0E-02*** (7.42)

10 XVG -0.69** (-2.59)

1.7E-02*** (3.52)

𝑡𝑡 -3E-03*** (-3.88)

𝜇𝜇 -8.88 ( ̶ )

Note: This table reports the estimates for a fully specified Vector-Error-Correction Model using our set of privacy coins. The model accounts for an intercept 𝜇𝜇 and a time trend 𝑡𝑡 in the cointegration equilibrium relationship. Our model uses daily data of log prices. The model has a lag-order of 𝑝𝑝 = 5. We report the estimates for the cointegration vector 𝜷𝜷 and the estimates for the adjustment parameter vector 𝜶𝜶. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations. ** Statistically significant on a 5% level. *** Statistically significant on a 1% level.

Results

24 January 2020 20

Table 7 Vector-Error Correction model estimates using non-privacy coins.

i Coin 𝜷𝜷�𝒊𝒊 𝜶𝜶�𝒊𝒊 1 XRP 1

( ̶ ) 3.2E-03 (0.86)

2 LTC 0.35 (1.02)

-6.4E-04 (-0.22)

3 ETH -1.53*** (-5.29)

1.5E-02*** (4.84)

4 DOGE -1.59*** (4.16)

1.4E-02*** (4.13)

5 PPC -0.57 (-1.00)

-2.6E-03 (-0.76)

6 BTS -0.91*** (-2.85)

8.7E-03** (2.24)

7 XLM 0.37* (1.67)

9.1E-03** (2.23)

8 NXT 0.49** (2.10)

1.8E-03 (0.47)

9 MAID 0.68** (2.16)

-1.4E-03 (-0.42)

10 NMC 2.18 (0.44)

-5.4E-03 (-1.16)

𝑡𝑡 9E-04** (1.98)

𝜇𝜇 -1.11 ( ̶ )

Note: This table reports the estimates for a fully specified Vector-Error-Correction Model using our set of non-privacy coins. The model accounts for an intercept 𝜇𝜇 and a time trend 𝑡𝑡 in the cointegration equilibrium relationship. Our model uses daily data of log prices. The model has a lag-order of 𝑝𝑝 = 5. We report the estimates for the cointegration vector 𝜷𝜷 and the estimates for the adjustment parameter vector 𝜶𝜶. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations. * Statistically significant on a 10% level. ** Statistically significant on a 5% level. *** Statistically significant on a 1% level.

24 January 2020 21

Results

Fig. 1. Cointegration relationship of privacy and non-privacy coins. This figure plots the cointegration relationships for privacy and non-privacy coins over time. The sample period is from January 1, 2016 until December 31, 2018 corresponding to 1096 observations. The correlation is estimated at -0.10 implying that those two market equilibria are two distinct phenomena.

-4

-3

-2

-1

0

1

2

3

Jan 07, 2016 Jul 25, 2016 Feb 10, 2017 Aug 29, 2017 Mar 17, 2018 Oct 03, 2018

Privacy coins Non-privacy coins

Conclusion

• Majority of cryptocurrencies are a part of that market equilibrium for both the sub markets.

• Our findings provide evidence for market inefficiency in both submarkets of privacy and non-privacy coins.

• Underlying forces that cause privacy coins equilibrium are unrelated to those at work in the non-privacy coins market.

• It could be that the market actors in the privacy coin market are different from those that trade in the non-privacy coin market.

• Moreover, potential factors that might have caused the cointegration relationships should be the subject of future research.

24 January 2020 22

Thank You!

24 January 2020 23Vaasan yliopisto | Niranjan Sapkota