Bitcoin Has an Unusual Relationship With Volatility

Too much hype could be the sign of a cryptocurrency bubble.

It will pop sooner or later.

Photographer: Ralf Hirschberger/AFP/Getty Images

Cryptoassets are showing the signs of a bubble. Market capitalization is $222 billion, 16 times as much as a year ago, with no significant change in tangible economic value. The virtual currencies elicit investment excitement and breathless news coverage; sober analysis treating both technical issues and economic fundamentals is rarer.

Bitcoin Ends Year at $10,000, Says Mike Novogratz

There is incessant cheerleading by people who own cryptoassets, manufacture them or sell associated services. A prominent example is the Amazon best-seller "Cryptoassets," 1  which is filled with classic financial hype and bubble logic. The book offers no clear evidence that investors will realize value from buying these assets today. There is no discussion of exit strategy, no speculation about who will pay to deliver profits to current investors.

The book begins with a graph showing rapid increase. It's just powers of 2, the numbers don't relate to anything. "Cryptoassets" is filled with upward graphs 2  of Google searches for blockchain, FANG stock performance, U.S. monetary base, world gold supply. These data are unrelated to cryptoasset fundamentals, and seek to create a mood of irresistible rapid growth. 3  If virtual money crashes, this book could be a monument to hype like "Dow 36,000," which was published six months before the Nasdaq crash in 1999. 4

All bubbles are accompanied by hype, but hype is not proof we are in a bubble. One quantitative bubble test considers the relation between asset price and volatility. The mathematics get intimidating, but the general idea is that if volatility increases too fast as price goes up, an asset price has to go to either infinity or zero. Because it can't go to infinity, 5  it has to go to zero. It will shoot upward and then crash. 6

The chart shows bitcoin annualized volatility versus price. 7  They have an unusual relationship. At prices from $0.06 to $0.25 there is the rapidly increasing volatility associated with a bubble. But then volatility falls, only to repeat the process three more times. The overall trend seems to be down. The straightforward interpretation is that there are people willing to fund a bitcoin bubble at any opportunity, but in the past the price has always stabilized at a non-bubble level. I have never seen this kind of pattern in other assets, so it's pure guesswork, but it fits my intuition that cryptoassets have solid economic fundamentals -- and also a lot of wannabe bubble profiteers impatient to push the price up beyond those fundamentals. (Full disclosure: I own bitcoins and other cryptocurrencies.)  

At prices beyond $256 you could argue that volatility is rising with price, or that volatility has stablilized around 80 percent to 100 percent a year. But bitcoin trading doesn't represent a lot of real money and liquidity is erratic, so it's unwise to put too much faith in the numbers.

We'll know more if bitcoin futures traded on Cboe Global Markets and CME Group attract significant institutional interest, or if the clearing house LedgerX has continued growth in trading volume. 8  If an influx of money causes price and volatility to rise, that suggests a bitcoin bubble. The price rise says that bitcoin price is determined by the ability to sell to new investors, not fundamentals. 9  The rise in volatility accompanying a price increase suggests bubble dynamics that lead to accelerating price gains followed by a crash.

If institutional money comes in without dramatic price and volatility increases, it's evidence against a bitcoin bubble. If there is scant institutional interest and bitcoin prices are stable, it's evidence of no bubble. But if bitcoin prices plunge, it suggests they were supported only by the hope of selling to greater fools.

There's no doubt cryptocurrencies are surrounded by the kind of hype you see in a bubble, and that many boosters are talking their book (saying what will make them money rather than what they have good reason to believe is true). We can see what appears to be the effect of hype in the history of bitcoin volatility, but so far at least, prices do not display the long-term dynamics associated with bubbles. The next few months should bring important evidence to help us decide.

Bloomberg Prophets Professionals offering actionable insights on markets, the economy and monetary policy. Contributors may have a stake in the areas they write about.

  1. By Chris Burniske and Jack Tatar.

  2. It also has lots of graphs showing many series in grayscale that are impossible to distinguish; the graphs are clearly there to make the book seem weighty rather than to communicate information, since if the reader can't tell which line represents which series, there is no way to learn anything from it.

  3. The authors are careful to cover themselves in the text with pious reminders of risk that are never clear or graphic enough to compete with the relentless drumbeat of enthusiasm.

  4. Instead of tripling from 12,000 to 36,000, the DJIA fell one third to 8,000 and didn't recover for seven years.

  5. Actually it can, if the units you are measuring price in go to zero. This is an important qualification with cryptocurrencies, but not one I plan to discuss here.

  6. A simple mathematical example is a price that either doubles or goes to $1 every period with equal probability. If it goes to $1, it stays there forever. The expected next period price is always $0.50 greater than the current price, so investing in the asset always has positive expected value. Yet the price is certain to crash to $1 eventually. It's true that the long-term future price of the asset is its current price plus $0.50 times the number of periods you look in the future, but the farther out you go, the more that expectation consists of microscopic probabilities of astronomical value. Those astronomical values quickly become unrealistic.

  7. Both scales are logarithmic.

  8. As of Nov. 15, LedgerX had almost $2 million of trading volume, including an institution-sized call option purchase for 100 bitcoin at $8,000 per coin, expiring Dec. 29. The implied volatility was 44 percent, less than the lowest observed volatility for bitcoin.

  9. It's ironic that many cryptocurrency boosters assume that bitcoin prices will rise rapidly if it becomes easier for institutional and retail investors to buy them. This means they assume bitcoin is a bubble that needs greater fools, rather than something with solid economic value that could be held long-term to realize return in real goods and services. Larger money flows into cryptoassets should make the price more accurate, not necessarily higher or lower, if investors are buying on fundamentals.

To contact the author of this story:
Aaron Brown at

To contact the editor responsible for this story:
Max Berley at

Before it's here, it's on the Bloomberg Terminal.