Churches & Bars, Bitcoin & Stocks

By admin_45 in Blog Churches & Bars, Bitcoin & Stocks

Many of you know I went to the University of Chicago for my MBA many years ago. At the time the school was best known for steely-eyed quantitative analysis. Occasionally, however, they essentially tortured the data until it gave up the answers the researchers wanted.

This approach sometimes led to hilarious conclusions. One grad student did a simple regression analysis on the number of churches and bars in each of Chicago’s wards (neighborhoods). He found the relationship to be essentially one-to-one. The data seemed to say, “Churchgoers are also really big drinkers”.

What he neglected to account for, of course, was population size. More densely populated wards had more churches, bars, supermarkets and pretty much any sort of storefront. Happy to recycle bad papers for didactic purposes, the school used the work as an example of how NOT to do data-centric modeling.

Thirty years later and I still approach statistical analysis with a good deal of caution, but today we want to address a frequently asked question: how correlated are bitcoin and US stock markets just now? We pulled the daily price data for each back to the start of 2016, and calculated daily returns. (Never do correlations or regressions on prices. Use percentage changes… Another life lesson from Chicago…)

Here is what we found:

#1. Bitcoin is now as much or more correlated to US stock prices as it has been since it first started to break into public awareness back in 2016.

We looked at 3 holding period ranges for bitcoin and the S&P 500: 10 days, 30 days, and 90 days. All back to January 1st 2016.

For the first two measures - 10 and 30 day holding periods - bitcoin’s daily price action is currently 79% and 52% (respectively) correlated to daily S&P returns. This is equivalent to other high points for short-term bitcoin/S&P 500 correlations during 2016 and 2017.

The 90 day return correlation for bitcoin/S&P 500 currently sits at 33%, its highest ever back to January 2016. This is 74% higher than the previous 90-day correlation high of 19%, set in mid December 2017.

The long run averages (2016 to 2018 YTD) of 10, 30, and 90 days correlations between bitcoin and S&P 500 daily returns are all close to zero: -4.6%, -1.4%, and -1.0%, respectively. Therefore, the current high-correlation environment is extremely unusual.

#2. There are some fundamental reasons why bitcoin is now more correlated to US equities.

Bitcoin has become a much more mainstream investment vehicle in the last 6 months. There are now 23 million bitcoin wallets, up 35% from 17 million in early October 2017. Crypto currencies have moved from the fringes of investor interest and it is no longer just libertarians and computer nerds trading these assets. That growth creates much more overlap between equity and bitcoin owners. Despite all the chatter about credit card and mortgage loans going to fund bitcoin purchases, we suspect that most of the growth in bitcoin ownership comes from individuals with savings to invest. And since investors only have one brain to process risk, they will make similar decisions about crypto currencies and stocks when they see price volatility in the latter.

The word “Bubble” gets thrown around pretty freely when discussing both US equities and bitcoin. Each has its true believers, to be sure. But each has its own base of fence-sitters that, as we have noted, own some of both assets.

#3. Given the size of the US equity market (+$20 trillion) versus bitcoin ($137 billion) and crypto currencies generally ($386 billion), we have no doubt that if these correlations persist it will be stocks dragging crypto around, and not vice versa.

Our bottom line on this analysis: over the near term (the rest of this month) a volatile and declining US equity market will be bad for bitcoin, but higher S&P levels may not help much. Correlations between the two assets seem to rise when US stocks falter, and disconnect when equities rise. We think that makes sense. The investor base where stocks and bitcoin overlap will shed risk in both when the S&P moves sharply lower, but allocate capital more selectively otherwise.

Even bitcoin fans will have to admit that it is in an investment purgatory at the moment, making lower lows and lower highs. The story for stocks is clear enough: will rising earnings expectations trump rising rates? For bitcoin, the narrative is cloudier just now: how will regulators around the world address their concerns? Will prices rise enough to sponsor new investment interest? It is a very different calculus from stocks.