Memoire Online - Global portfolio diversification with cryptocurrencies

Master thesis submitted to the Faculty of Economics and Business
Institute of Financial Analysis

Faculté des sciences économiques Avenue du 1er-Mars 26 CH-2000 Neuchâtel www.unine.ch/seco

Abstract

This study raises questions about the potential of cryptocurrencies as a new alternative investment. I explore the ability to which major cryptocurrencies endow diversification and hedging benefits to a global investor. Using the dynamic conditional correlation model developed by Engle (2002), I find evidence of effective diversification and weak hedging effects against global traditional assets. Furthermore, I investigate the issue using risk based portfolio optimization frameworks. I find enhanced performance of the investor's portfolio when including Bitcoin to a well-diversified portfolio. Likewise, later generation of cryptocurrencies Ripple, Dash and Litecoin provide better improvement on a risk adjusted basis. Nonetheless, their very high volatility worsen off the portfolio's downside risk.1

1 ACKNOWLEDGEMENTS: I would like to express my gratitude to my supervisor Prof. Frédéric Sonney for his guidance and valuable advice. I would like to thank my family for their love and unconditional support. I am also grateful to my partner and my best friend for brightening my life throughout the thesis process.

1. Introduction

The 2008 financial crisis led to growing skepticism around traditional financial systems.

As a response to these backdrops, a programmer under the pseudonym of Nakamoto introduced Bitcoin, a decentralized medium of exchange in 2009. The peer-to-peer electronic currency shows unique features. It is not backed by any central authority, has a fixed supply set in advance of 21 million Bitcoins and is created via an innovative technology, Blockchain that registers transactions and stows them into transparent blocks. These blocks are completed once these transactions are verified and secured into a distributed network. Therefore, this creation process is similar to gold mining, which led to Bitcoin being called the «digital gold».

Following the inception of Bitcoin, many alternative cryptocurrencies, using the same blockchain technology emerged. Whether addressing same purposes as Bitcoin or providing innovative decentralized solutions, they attracted investors' attention and gained growing market shares with the most popular among them being Ethereum, Ripple, Dash, Stellar and Litecoin.

Driven by high capital inflows, Cryptocurrencies witnessed price rise in tandem. Market capitalization and volume traded continued growing exponentially until the end of 2017, when Bitcoin realized its meteoric rise before topping out in the following months.

This rapid surge in prices and the high volatility displayed by the cryptocurrency market has attracted mostly speculators seeing Cryptocurrencies as a speculative asset rather than a currency store of value. Consequently, many investors question whether cryptocurrencies are just fictitious currencies forming speculation bubbles or a valuable opportunity investment.

The main purpose of this study is to explore Cryptocurrencies as a new alternative investment.

The research is conducted from the perspective of a global investor who considers diversification as of paramount importance. Considering the excessive volatility encompassing

cryptocurrencies, I restrain cryptocurrencies to hedging and diversification uses only. According to Baur and Lucey (2010), a hedge is an asset that shows adverse correlation to another asset, whereas a diversifier exhibits marginal positive price co-movements with the other asset. Furthermore, this study sheds light on the use of the cryptocurrencies as performance enhancers in a global portfolio while previous research focused only on Bitcoin. I benchmark the cryptocurrency market with Bitcoin and three major alternative cryptocurrencies: Ripple, Dash and Litecoin. Whilst among traditional assets I consider stocks, bonds, real estate and gold.

In order to investigate the diversification and hedging traits of the aforementioned cryptocurrencies and capture the co-movement between each cryptocurrency and the traditional assets, I consider the multivariate dynamic conditional correlation GARCH model of Engle (2002). The results offer compelling evidence of cryptocurrencies as effective diversifiers, yet they exhibit weak hedging properties. I find that Bitcoin acts as a strong hedge only against price movements of Chinese equities, global real estate and corporate bonds. Additionally, Litecoin qualifies as a hedge for global real estate and Japanese equities. Whereas Dash is only a good hedge against global bonds. Ripple, on the other side, does not possess any hedging properties since it exhibits moderately positive correlation with all traditional assets.

I further examine the diversification perquisites of cryptocurrencies in a portfolio comprising all traditional asset classes. Bruder and Roncalli (2012) argue that risk aversion of institutional and individual investors increased significantly after the 2008 financial crisis. Thus, they prompt the use of investment strategies based on risk budgeting and diversification instead of return forecasting ones.

Regarding this matter and the high volatility nature of Cryptocurrencies, this study adopts four optimization frameworks that focus on measuring risk: traditional minimum variance, minimum conditional value at risk, inverse volatility and maximum diversification.

The out-of-sample performance of the optimal portfolios is studied on a risk adjusted return basis. The back-testing results confirm the evidence of cryptocurrencies being outstanding diversifiers. Regardless of the strategy, I find that adding cryptocurrencies to the basic portfolio, the risk return ratio increases significantly albeit at different magnitudes. On the other side, the downside risk of the portfolio increases especially when alternative tokens are included. Interestingly, Bitcoin succeeds to reduce the downside risk under minimum variance and inverse volatility.

The research paper is structured as follow: Section 2 presents the literature review, section 3 undergoes the methodology details, section 4 introduces the data, section 5 reports empirical results and robustness check and section 6 provides conclusion.