Risks and Returns of Cryptocurrency
Read the paperopens doi.org in a new tab
What they found
The first systematic asset-pricing study of crypto. Using Bitcoin, Ripple, and Ethereum returns from 2011 to 2018, the authors found that crypto returns have essentially no exposure to stock, currency, commodity, or macro factors, so they are a distinct asset class. Crypto returns were predicted by crypto-specific momentum (last week's return positively predicts this week's) and by investor attention (Google searches and Twitter posts), with high attention followed by higher returns. Production cost proxies for Bitcoin mining did not explain returns. Crypto has a high average return and enormous volatility with a strong right tail.
What you can use
- Crypto returns have been uncorrelated with traditional risk factors, which makes it a diversifier but also means standard valuation tools do not apply.
- Short-term momentum (one to four weeks) has been strong in crypto; last week's winners tended to keep going.
- Attention drives returns: spikes in searches and social media have preceded price rises, which is both an edge and a bubble warning.
Caveats
Sample covers only three coins through 2018, a period dominated by one bubble; the momentum and attention effects may not persist. A free NBER version exists.
Tags: crypto, asset-pricing, momentum, attention
Summaries are our own reading of the paper, not the authors' words. Educational only, not advice. Discuss it in Book Club.