Bitcoin’s Risk and Return Explained in 7 Amazing Charts (That You’ve Never Seen Before)
A finance professor explains how Bitcoin has really performed relative to the stock market
What’s bitcoin really like as a potential investment, and how does it compare with traditional investments in terms of risks and recent returns? Will bitcoin soon be part of a new cryptocurrency asset class that all investors should consider as part of their investment portfolio? We now have over six-and-a-half years of price data since mid-September 2014, when U.S. regulators approved the first derivative product involving bitcoin. So, to answer those questions, I gathered and analyzed price data on bitcoin, the S&P 500, a bond index, gold, the dollar, and the FANG growth stocks — Facebook, Amazon, Netflix, and Google, now Alphabet. (If you want to know more about the data, details are presented in an Appendix at the end of this article.)
There are lots of lenses through which to view cryptocurrencies such as bitcoin, for example: evaluating the fundamentals of bitcoin as a technology; its use as a currency, particularly with newsworthy events such as Tesla’s purchase of $1.5 billion of bitcoin and with the initial public offering of Coinbase Global, Inc., a cryptocurrency exchange; and bitcoin’s environmental impact related to the mining of new coins. These are great discussions to be had, but here I’ll simply be investigating bitcoin as a potential investment, from a historical perspective relative to other investments. I’ll share my findings in seven simple charts, each of which tells an amazing story — you can also get a copy of my spreadsheet and the charts here.
Chart #1: Bitcoin’s Incredible — and Bumpy — Rise
Unlike most charts you’ll see, I chose to use a log scale. The beauty of the log scale is that it is easy to capture the rate of increase: going from $100 to $1,000 gives you the same rate of return as going from $1,000 to $10,000. If growth is steady then we would expect to see a relatively straight-line increase from bottom-left to upper-right.
What’s amazing about this chart is the incredible rise in the price of bitcoin in just over six-and-a-half years, from $457 on September 17, 2014, to over $63,000 in April 2021 before a recent pullback, or over 111 times. I can’t even fathom any traditional investment coming close. We certainly see an upward trendline, but it is far from straight. For example, bitcoin price went from around $19,500 on December 16, 2017 down to $3,242 roughly a year later, a drop of over 83 percent. That’s an amazing drop during such an incredible upward trend.
Chart #2: Bitcoin’s Path Compared with Traditional Investments
What’s amazing about this chart is the growth in a $1,000 investment in bitcoin September 2014 to almost $139,000 in April 2021, before the pullback. It shows me why there is so much excitement in bitcoin as an investment — where else could we dream of such incredible returns? By comparison, among the FANGs, only Amazon was able to cross the $10,000 threshold, which is still a fantastic return over the period, especially for established stocks. The S&P 500 index roughly doubled, which is an above-average return compared with average returns over the long-run. A $1,000 investment in gold grew to over $1,400, bonds to around $1,200, and the dollar investment to something slightly less than bonds.
Another amazing thing about this chart is that all of the assets increased over the time period, despite a major drop in many of them, particularly near the start of the Covid-19 pandemic in March 2020. We shouldn’t expect all assets to go up over any specific period, particularly one that is relatively short.
Chart #3: Bitcoin Return-and-Risk Compared with Traditional Investments
This chart captures both return and risk. I’m measuring return here as the compound annual growth rate or CAGR (statisticians also call this the geometric average) or the rate of growth between the starting value and the ending value (an alternative measure is the simple average, the mean, or arithmetic mean, which is the average returns each year over the number of years). I’m measuring risk as the standard deviation of returns. I’m using the daily price changes and converting them to annualized return and risk measures.
What’s amazing about this chart is that it shows how bitcoin is in a completely different risk-return universe. We tend to think of the traditional asset universe in the red box in the lower left-hand corner. A broad stock market index like the S&P 500 has had a long-term average annual return (including dividends) of around 10 percent with an annual standard deviation of just below 20 percent. The 2014–2021 period is consistent with these long-run averages. Bonds tend to have lower returns and risk compared with stocks, and that was the case here as well. Individual stocks tend to be riskier than the overall market, and that was the case with each of the FANG stocks, which also happened to have higher average returns than the S&P 500 index over this period (of course by definition, not all stocks can outperform the market average!). While bitcoin averaged an incredible triple-digit average annual return, it also displayed mega-risk, with an annualized standard deviation of over 73 percent.
Chart #4: Bitcoin’s Correlation with Traditional Investments
Correlation is a statistical measure of the extent to which two asset prices change in a similar or different matter. Correlation is scaled from -1 or perfect negative correlation, to +1 or perfect positive correlation. In perfect negative correlation, when asset A’s price increases, asset B’s price decreases by the same amount; with perfect positive correlation, asset A’s price and asset B’s price move in lock-step. Lower and even negative correlation among assets is a good thing from a portfolio diversification perspective, as individual asset ups and downs are smoothed to some extent. If you randomly choose any two stocks, they would most likely have a low and positive correlation.
What’s amazing about this chart is how low bitcoin’s correlation is relative to other asset classes. The correlation with the S&P 500 is 0.14, which suggests that having bitcoin in one’s portfolio is a positive thing to help smooth the ups and downs of the stock market. Bitcoin has a slight negative correlation with bonds; since bond prices move inversely with yields or interest rates, this suggests a slight positive relationship between bitcoin prices and interest rate movements.
What surprised me the most was the negative correlation between bitcoin and the dollar. Since bitcoin is a cryptocurrency, I thought there would be a positive correlation with the dollar, but that appears not to be the case. To the extent that bitcoin is positively correlated with increasing interest rates which tend to capture expected inflation, there may be a potential inflation hedge with bitcoin, but the correlation isn’t that strong.
Chart #5: Bitcoin Price Changes Compared with a Normal Distribution
You are probably familiar with the Normal distribution or bell curve — for example, the shape of a histogram of gender-specific adult heights, or exam scores, like the shape of the green line in the chart. As a first approximation, stock returns tend to exhibit the shape of a Normal distribution, but not quite — and the same goes for bitcoin price changes. What’s amazing about this chart is that not only are bitcoin price changes more bunched around the average compared with a Normal distribution, but the tails are much fatter. With a Normal distribution, about two-thirds of the observations tend to be plus/minus one standard deviation away from the mean, while 95 percent of the observations tend to be plus/minus two standard deviation away from the mean, and only 0.3 percent are more than three standard deviation away from the mean.
What we see in this chart is that tail events are much more prevalent than we would expect with a Normal distribution. For example, bitcoin’s worst one-day drop occurred on March 12, 2020, with a decrease of 37 percent, early in the Covid-19 pandemic and around the time stocks were declining substantially as well — an event more than 8 standard deviations below the average price change. By comparison, as I noted in a recent article, the worst one-day drop in stock market history was -22.6 percent, in 1987.
On the other hand, huge price increases are much more prevalent than with a Normal distribution as well. For example, the biggest one-day price increase, 25.3 percent, occurred on December 7, 2017, and followed another huge gain the previous day of 19.9 percent. Clearly, these huge price swings aren’t for the faint of heart.
Chart #6: Bitcoin Price Changes Compared with Nasdaq in the Dot-Com Bubble
Even if you didn’t live through the dot-com bubble, you have certainly heard of it. The World Wide Web dates back to the early 1990s. In 1995, web browser Netscape had an extremely successful initial public offering, which accelerated interest in internet-based companies and helped spawn the dot-com bubble. Most of these stocks were listed on the tech-dominated Nasdaq exchange, and so the Nasdaq composite index became a proxy for the dot-com craze. The index peaked on March 10, 2000 at a level of 5,048.6. It wouldn’t pass that level again until April 2015 (it’s now around 14,000). In retrospect, investors’ expectations were too optimistic.
Here’s a thought experiment: how does the 2014–2021 bitcoin price change compare with the rise in the Nasdaq during the dot-com craze? The way I created this chart was to superimpose the 2014–2021 bitcoin data to coincide with the peak of the dot-com bubble as captured by the Nasdaq index, from August 1993 to March 2000 (I also continued the Nasdaq chart through to the trough in October 2002).
What’s amazing about this chart is that if Nasdaq increased in price during the dot-com bubble at the same rate of increase as bitcoin between 2014 and 2021, then instead of peaking at a level of around 5,000, it would have peaked at close to 100,000, or almost 20 times greater. I think this really puts into perspective what’s been happening with bitcoin. In retrospect, if the mid-1990s to early-2000 Nasdaq was really a bubble, then what is bitcoin now? Something to ponder.
Chart #7: Bitcoin’s Volatility Over Time
In this last chart, I wanted to compare a moving-average (last 100 trading days, during weekdays when stock exchanges were open) of bitcoin volatility compared with that of the stock market index, the S&P 500.
What’s amazing about this chart is that bitcoin isn’t getting any less volatile over time. Historically, risk — as measured by standard deviation of returns — tends to be much more stable than returns. The riskiness of the stock market in the 2014 to 2021 period is similar to that of over the past century (the 1930s was the only decade of substantially higher-than-average volatility). As bitcoin becomes more established, I thought we might see a decrease in its volatility, but that hasn’t been the case. If you have a well-diversified portfolio of stocks, like the S&P 500 index, and you occasionally have trouble sleeping because of market volatility, I imagine you would have insomnia if your portfolio was dominated by bitcoin.
So, what do I conclude from the analysis and charts? Of course, the major caveat is that past price changes aren’t necessarily indicative of future performance, and this analysis is based on less than seven years of data.
● It really matters when you buy and sell, particularly in bitcoin given the huge swings in prices.
● Bitcoin offers over-sized risks for the potential returns, with volatility that is orders of magnitude greater than with traditional investments.
● Low correlations suggest bitcoin could potentially become a diversifier, as part of a broader portfolio of traditional investments.
● Compared with the rise in the Nasdaq index in the mid-to-late 1990s, during the so-called dot-com bubble, bitcoin’s rise has been about twenty times as great, suggesting we may be experiencing a bitcoin bubble that makes the dot-com era look like a mere blip.
If you like dabbling in cryptocurrencies, are aware of the risks, and can afford to lose, then go right ahead. But, in my opinion, until bitcoin experiences returns and risks more in-line with traditional assets — or even very risky stocks — I don’t see it, along with other cryptocurrencies, being a “must-have” investment as part of one’s portfolio, at least not yet. Perhaps some investors with very high-risk tolerances will consider bitcoin as at most a small component of an “alternative investments” basket in their portfolios. Whatever you decide, it’s important to understand the stories the data are telling.
Appendix: Data Details
The major data source I used was Yahoo!Finance. Bitcoin expressed in U.S. dollars is available by searching for BTC-USD. The first price available is September 17, 2014. Other sources such as Marketwatch.com have bitcoin prices back to 2011. Prior to February 2011, bitcoin’s price was less than a dollar.
Unlike stocks and bonds, bitcoin is traded every day. Since I wanted to compare bitcoin with stocks that don’t trade on weekends or holidays, I scrubbed the bitcoin data by deleting the weekend and holiday prices. While not impacting on returns over the period I studied, that actually slightly increased the volatility measure since Monday’s price changes were now being measured relative to Friday’s instead of Sunday’s. That increased the daily volatility (standard deviation) measure from 4.13 percent to 4.62 percent.
To get a stock market index, and I chose the most common benchmark, the S&P 500 index. While the index itself isn’t an investment product, there are numerous ETFs and funds that replicate the index and at very negligible cost. For bonds I chose an actual traded product, iShares 20+ Year Treasury Bond ETF (ticker TLT). Different bond funds that weren’t investing exclusively in long-term Treasuries would have different return-to-risk profiles.
I chose gold as another investment class. And since bitcoin is a type of cryptocurrency, I chose another currency benchmark, the “Trade Weighted U.S. Dollar Index: Broad, Goods and Services” which is available from Federal Reserve Economic Data (FRED). While this index isn’t an investable product per se, it is meant to represent a currency strategy, for example, betting on the strength of the U.S. dollar against a basket of currencies of major trading partners.
In addition to asset classes, I wanted to compare bitcoin to individual “growth stocks” that aggressive investors might also have considered in September 2014. To mitigate a look-ahead bias and just choose stocks that I later knew had done well, I found out that in February 2013 Jim Cramer, host of CNBC’s Mad Money, and RealMoney.com analyst Bob Lang introduced the phrase FANG stocks — Facebook, Amazon, Netflix, and Google (now Alphabet) — to collectively represent “hot” growth stocks (Apple was later added and the acronym was changed to FAANG), so I gathered data on them. For simplicity, I didn’t include dividends on any of these stocks, nor for the S&P 500 (the current dividend yield is around 1.5 percent), and so that slightly underestimated actual returns.
Stephen Foerster is a co-author, with Andrew Lo, of In Pursuit of the Perfect Portfolio: The Stories, Voices, and Key Insights of the Pioneers Who Shaped the Way We Invest, Princeton University Press (August 17, 2021).