Bitcoin and portfolio optimization

ListedReserve Pty Limited
ACN 624 133 133
AFSL 513 745

Executive summary 

The addition of a small proportion of bitcoin to investment portfolios has significantly enhanced performance over the period under review, 2013-2019, without significant increases in risk. The base portfolio for comparison consists of 60% S&P 500 and 40% S&P bond index.

The power of diversification

Diversification is one of the most powerful and readily available risk management tools. But it’s not simply a matter of spreading investments around.A key factor in lowering overall portfolio risk is finding assets that will behave differently to the existing portfolio, under different circumstances.For share and property investors, this could mean gold, volatility and cryptocurrencies.

Research (Liu and Tsyvinski, 2018)shows that cryptocurrencies have no equivalent exposure to most common stock market and macroeconomic factors. They also have limited correlation to the returns of other currencies and commodities.

We have picked the ten most actively tradedcryptocurrencies and run correlations with different asset classes. Strong correlation is implied by figures approaching 1.0, weak correlation below0.5, negative figures imply an inverse correlation, so -1.0 would result in the exact opposite thing happening to one asset versus the other. The result ranges from -0.19 to 0.30 showing very little to no correlation with traditional asset class such as major currencies, broad market equity indices, bonds and gold. We consider then thatcryptocurrencies might be a diversifying component within multi-asset class investment portfolios.

Bitcoin and Portfolio Optimization

We have constructed a few hypothetical simulated portfolios: base portfolio, base portfolio with 1%/3%/5% bitcoin (component asset weights are held constant over the period). Then, mean-variance analysis is used to assess risk and reward attributes of the hypothetical simulated portfolios, and Sharpe ratio is used to measure risk adjusted excess return (portfolio performance).

Base portfolio consists of a 60% allocation to equity investment via the S&P 500 and a 40% allocation to bond investment via the S&P US Aggregate Bond Index.

The full matrix of correlations can be found in the appendix where both currencies, major stock indices and commodities are compared, with strikingly minimal correlation to cryptocurrency across all classes.

Table 1: Hypothetical simulated portfolio performance

29 April 2013 to 29 November 2019

* Annualized figures are based on 252 trading days.

** The Sharpe ratio is calculated as the annualized excess return of the portfolio over the 3-month US T-Bill divided by the standard deviation of excess returns.

*** Ratio improvement is calculated by taking the Sharpe ratio of the base portfolio + 1%/3%/5% BTC Portfolios and dividing each by the Sharpe ratio of the base portfolio.

Graph 1: Hypothetical Simulated Portfolio Performance

CRYPTOCURRENCIES HAVE HISTORICALLY EXPERIENCED SIGNFICANT INTRADAY AND LONG-TERM PRICE SWINGS. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN. PAST PERFORMANCE IS NOT INDICATIVE OF FUTURE RESULTS.

In conclusion, the analysis reveals that

• A small allocation to bitcoin significantly enhanced the cumulative return of the base portfolio during the period under review, while only minimally impacting its volatility.

• The Sharpe ratio improves while adding allocation of BTC to the base portfolio from 1%-5%, proving that it enhanced the performance of investment portfolios to more favorable risk/return profiles.

Reference

Liu, Y. and Tsyvinski, A. 2018, Risks and Returns of Cryptocurrency. Web access 19 Dec 2019 through

https://economics.yale.edu/sites/default/files/files/Faculty/Tsyvinski/cryptoreturns%208-7-2018.pdf

Wu, C.Y. and Pandey, V.K 2017, The Value of Bitcoin in Enhancing the Efficiency of an Investor’s Portfolio. Web access 19 Dec 2019 through

https://www.researchgate.net/publication/280883667_The_value_of_Bitcoin_in_enhancing_the_efficiency_of_an_investor’s_portfolio

APPENDIX

Matrix 1: Cryptocurrencies vs majorworld currencies

29 April 2013 to 29 November 2019

Matrix 2: Cryptocurrencies vs major asset classes and indexes

29 April 2013 to 29 November 2019

The digital assets shown above have historically experienced significant intraday and long-term price swings. As the period during which digital assets have been available for trading is limited, the correlations may not be meaningful when considering longer periods. Past performance is not indicative of future results.

*Interestingly to note that most cryptocurrencies have negative correlation with Australian dollar. We will run some further analysis later between cryptocurrencies and Australian based asset classes.