Yale Research Highlights the Risks and Returns of Cryptocurrencies
Yale Economists Weigh In
A recently released study authored by a pair of Yale economists, conducts a deep dive into the financial profile of cryptocurrencies, comparing attributes to traditional asset classes. Examining Bitcoin, Ethereum and Ripple, the three largest cryptocurrencies by market capitalization, the study arrives at several interesting conclusions. After assessing several factors, it is established that the risk-return profile of cryptocurrencies is distinct from those of stocks, currencies and precious metals. Cryptocurrencies occupy their own unique niche as an investment asset, where returns are significantly affected by momentum and investor attention. The findings have implications for portfolio construction within the traditional investor community and evaluating market behaviour of cryptocurrencies.
The Crypto Risk-Return Profile
Cryptocurrencies are displaying similar and often more favourable financial profiles to traditional asset classes. A measure that is often employed to assess the risk-return profile of a given asset is known as the Sharpe ratio. The Sharpe ratio indicates how much excess return you are receiving for the extra volatility that you endure for holding a riskier asset. The general rule of thumb is that the higher the Sharpe ratio, the better an asset’s returns relative to the amount of risk (measured as volatility) associated with that asset.
The paper finds that the Sharpe ratios of Bitcoin, Ethereum and Ripple at the daily levels are around 50-75% higher than traditional asset classes. At the monthly frequency, the Sharpe ratios are comparable to stocks for the same time period although higher than the historical Sharpe ratios for stocks. Essentially, cryptocurrencies are very high risk, experiencing frequent and wide fluctuations in price, but investors have historically been well-compensated for this risk with high returns.
Investors must have a strong stomach for volatility - the study found that cryptocurrency returns have high probabilities of exceptional negative and positive daily returns. For example, in Bitcoin, the probability of a -20% daily return “disaster” is almost 0.5% and the probability of a 20% daily return “miracle” is almost 1%.
Furthermore, the study concludes that cryptocurrency returns have low exposures to traditional asset classes- stocks, currencies, and commodities. Liu and Tsyvinski examined 155 potential risk factors in finance literature and found that almost none of them account for the returns of cryptocurrencies.
Driving Factors Behind Crypto Returns
The authors found that unique factors, distinct from traditional asset classes, are driving cryptocurrency return profiles: momentum and investors attention.
In capital markets, momentum is the rate of acceleration of a security's price or volume. The idea is simple: a rising asset’s price will rise further and falling asset’s price will fall further. The study found compelling evidence of a strong momentum effect at various time horizons. For Bitcoin daily returns, the current return positively and statistically significantly predicts 1-day, 3-day, 5-day, and 6-day ahead returns. Specifically, a one standard deviation increase in today’s return leads to increases in daily returns by 0.33%, 0.17%, 0.39% and 0.50% increases at the 1-day, 3-day, 5-day, and 6-day ahead returns, respectively. Ripple and Ethereum also showed similar momentum profiles, although Ethereum has a slightly weakened momentum effect compared to Bitcoin and Ripple.
Investor attention was the second significant predictor of cryptocurrency returns. The proxies for investor attention used by the authors included analysis of Google search trends and Twitter post counts. They found that high investor attention predicts high future returns over 1-2 week horizons for Bitcoin, a 1-week horizon for Ripple, and 1, 3, and 6-week horizons for Ethereum. The significance of momentum and investor attention in consistently explaining the variations in cryptocurrency returns is unsurprising. In its current state, the market is highly speculative and retail-driven.
Lastly, the authors examine alternate potential predictors of returns: mining costs, price-dividend ratio (because cryptocurrencies do not pay dividends, the authors proxy the fundamental value by using the number of Bitcoin Wallet users), and realized volatility. Contrary to prior suggestions, each of these independent measures do not predict cryptocurrency returns in any significance.
Necessary Exposure for the Diversified Investor?
Overall, cryptocurrencies represent an asset class that can be assessed using simple finance tools. However, they also occupy their own investment niche which is radically different from traditional asset classes. As such, a typical investor might consider a small allocation of their portfolio to cryptocurrencies. In fact, Liu and Tsyvinski recommend it. Using the Black-Litterman model, they suggest that a conservative investor might allocate 1% of their portfolio to Bitcoin. If the investor believes Bitcoin will continue to do as well as the past seven years, they should hold a 6% share of Bitcoin in their portfolio.
As every prudent investor with a spare dollar is, at a minimum, trying to keep pace with inflation and take advantage of the powers of compounding, analysis of the risk-reward profile of this nascent asset class is extremely relevant. Such studies make informative contributions to the growing body of crypto-literature and support the continued definition and legitimization of cryptocurrency as its own asset class.