The New Asset Class Opportunity


It’s no secret that returns on the majority of asset classes have been frothy for the past several years. This year has been especially fraught with tech IPOs performing about as badly as they did in 2009, resulting in the worst returns of this past decade. Global macros are not positive, either; venture capital funding in China has taken a first downturn since its inception twenty years ago, and political tensions certainly aren’t helping.  Subsequently, many investors agree that the longest economic expansion in U.S. history will have to come to an end, and soon.

While it could be argued that equities will retain some upside (at least for the near-term), markets will likely regress to the mean as the correction occurs. If bonds are out (believe us, they may be), and we can expect to continue to live in a world where negative interest rates prevail, then alternative investments should take precedence as the opportunity du jour.

The New Asset Class Opportunity

By alternative investments we mean private equity, hedge funds, REITs, and commodities (especially the newest ones). But of course, there is one opportunity that we think reigns supreme: cryptocurrency. And while much has been said with regards to bitcoin being a hedge akin to gold, a new study titled Risk and Returns of Cryptocurrency by Yukun Liu and Aleh Tsyvinski explains how cryptocurrencies are, in fact, not correlated with any stock, currency, or precious metal. That said, they aren’t totally indecipherable as an asset class and it is possible to assess them using simple finance tools.

We’ve stated this before, but cryptocurrency as an investment opportunity is immense. As the study explains, the mean and the standard deviation of its returns are magnitudes higher than traditional asset classes. The weekly mean for bitcoin (or, rather its weekly average return) is 3.79% with a standard deviation of 16.64. Sharpe ratios at the monthly levels are comparable to stocks, but the daily and weekly levels are 50% and 75% higher, respectively. Ethereum to gold is the only exception for any covariance to precious metals. As for macroeconomic factors, for Ethereum there is some loading on the durable consumption growth factor, but otherwise all cryptocurrencies remain uncorrelated.

As for cryptocurrency-specific factors that indicate price movements, let’s take a look. The constituents of the study were: (1) cryptocurrency momentum, (2) proxies for average and negative investor attention, (3) a proxy for price-to-“dividend” ratio, and (4) proxies for the supply conditions.

First, there is a significant time-series cryptocurrency momentum at the daily and weekly

frequencies for Bitcoin, Ethereum, and Ripple. A one standard deviation increase in today’s return on Bitcoin, leads to increases in average daily returns by 0.33 percent. The current return predicts the daily return; the authors ran regressions to show that when the current return increases by a certain amount, a unit of its variability correlates to a certain percentage of the bitcoin daily return. For Ethereum, the momentum effect is statistically significant at the 1-day and 5-day period, but at the 5-day period it negatively covaries. For Ripple, it is significant on the 1-5 day horizon.

Second, for bitcoin, Google searches could accurately predict returns one and two weeks ahead. A one standard deviation increase in this week’s searches leads to increases in weekly returns of 1.84 percent and 2.30 percent at the 1-week and 2-week ahead returns. For Ripple, they could predict 1-week ahead returns: a one standard deviation increase in this week’s searches leads to increases in weekly returns of 10.86 percent at the 1-week ahead returns. For Ethereum, searches were significant for 1-week, 3-week, and 6-week ahead returns. A one standard deviation increase in this week’s searches leads to increases in weekly returns of 4.36 percent, 3.45 percent, and 3.65 percent at the 1-week, 3-week and 6-week ahead returns. For negative investor attention, a one standard deviation increase of the ratio leads to a 2.75 percent decrease of Bitcoin returns next week.

As for a proxy for price-to-dividend ratio, the authors proxied the fundamental value of bitcoin by measuring the number of bitcoin wallet users. The price-to-dividend ratio is a measure of the gap between the market value and the fundamental value of an asset; the market value of cryptocurrency is just the observed price. The authors found that overall, there is a very weak relation between the future Bitcoin returns and the current price-to-dividend ratio (for the others, the data was not made immediately available).

Lastly, the study measured supply factors for these cryptocurrencies to see if these affected the value. Electricity and computer power were deemed the most logical factors for computing mining power; electricity was proxied with: (1) the value-weighted stock returns of the U.S.-listed electricity industries; (2) the value-weighted stock returns of the China-listed electricity industries; and (3) the Sinopec (SNP) stock returns, because U.S. and Chinese miners dominate the industry. For proxies of computer power, stock returns of companies that are major manufacturers of either GPU mining chips (Nvidia Corporation and Advanced Micro Devices, Inc), or ASIC mining chips (Taiwan Semiconductor Manufacturing Company, Limited and Advanced Semiconductor Engineering, Inc) were analyzed. Bitcoin and Ripple returns were not statistically significantly exposed to any of the factors, while Ethereum was, but only to the AMD stock returns.

Study Limitations

There are, of course, limitations to this study. For example, to construct a proxy for negative investor attention, the authors of the study created a ratio between Google searches for the phrase “Bitcoin hack” and searches for the word “Bitcoin”. We can hardly claim that this is a comprehensive corollary, being purveyors of the bitcoin market ourselves. Whether or not it’s the best example for negative investor attention is up for debate, especially in a nascent industry full of other volatile happenings and controversies like regulatory uncertainty.

As for constructing a proxy for supply factors, as this study does – although the authors considered that supply factors may be more important in recent periods, as mining is becoming costlier with the rising popularity of cryptocurrencies, and found no support for this hypothesis -, we believe that supply conditions are much more complicated than simple measures of electricity and computing power, especially considering factors like proof of stake protocols and others.

A Bright Future

A common critique of crypto is that their market value as an asset class is highly speculative, due to their complexity to retail investors and the fact that they have yet to be adopted as a form of payment (plus, that there are high barriers before this will occur). There are naysayers who call bitcoin a throughput for tax evasion, a ponzi scheme, and a bubble. Yet, we still have a bright outlook on the future. Why? Because the biggest institutional players are already in crypto: Bitcoin futures trade on the CME every day, Fidelity Digital custodies bitcoin, and the ICE has invested heavily in Bakkt.



Share this Post