This is a summary of “Decentralized Efficiency? Arbitrage in Bitcoin Markets,” by Sinan Krückeberg and Peter Scholz, published in the Third Quarter 2020 issue of the Financial Analysts Journal.
Listen to an audio version of this summary.
Frequent and sizable arbitrage opportunities exist to exploit price differences for bitcoin across cryptocurrency exchanges. The authors find profit opportunities in excess of $377 million in 2017 and $384 million in the first quarter of 2018.
What Is the Investment Issue?
Do professional arbitrage traders have opportunities to profit in the bitcoin markets? The market for bitcoin has grown in size from negligible a little over a decade ago to USD0.8 trillion: approximately one-ninth the size of the gold market.
The authors suspected that because the bitcoin market is relatively new and is dominated primarily by individual investors, inefficiencies in trading might exist that institutional traders could exploit. Other researchers had found that opportunities to profit from this type of inefficiency decreased over time.
Such inefficiencies would typically manifest in price discrepancies between identical assets in different bitcoin exchanges. If such price differences were present for long enough periods, then professional investors would be able to profit by purchasing the cheaper-priced bitcoin and selling those priced higher in an arbitrage.
How Do the Authors Tackle the Issue?
The authors analyzed price data from many of the major bitcoin exchanges around the world over the period 2013–2018. The data included volume, price, and timestamp information for each trade or tick. The authors took a three-pronged approach:
- Do bitcoin bid–ask spreads across the various bitcoin exchanges provide an opportunity for arbitrage?
- Do any predictable patterns exist? At which exchanges did the opportunities occur, when did the arbitrage opportunities occur, and did any particular events offer a greater chance of profits?
- Did the arbitrage opportunities last long enough and were they large enough to cover transaction costs with sufficient volume to simultaneously buy and sell? Wide spreads lasting a reasonable length of time suggest that unexploited opportunities exist in the market.
What Are the Findings?
In the 16 months from January 2017 through the end of the first quarter of 2018, arbitrage opportunities totaled $761 million at a minimum in potential risk-free profits. The authors calculated such potential profits by identifying more than 2 million 10-second intervals where the spreads averaged more than 4% of the value of a bitcoin.
Two exchanges stood out as consistently offering lower prices for bitcoin for quotes in US dollars: Bitfinex and Bitstamp.
Certain times during the day offered wider spreads for arbitragers to exploit: Spreads were wider between 08:00 UTC (coordinated universal time) and 16:00 UTC and narrower between 16:00 UTC and 24:00 UTC. The spread data show that arbitragers would benefit from operating primarily outside of US working hours.
The authors found that when a new bitcoin exchange opens, the spreads across the industry increase. However, this difference tended to get smaller over time.
When bitcoin exchanges are hacked or robbed, the market spreads tend to increase for a period of approximately five weeks following the malfeasance.
While many new markets that increase in size and popularity see arbitrage opportunities decrease as they mature, the authors noted that traders found “significant and increasing” spreads from which financiers could profit over the 2013–18 period.
What Are the Implications for Investors and Investment Managers?
...specialized institutional actors, who have both sufficient capital and the necessary human resources, can find uncaptured opportunities to generate arbitrage profits in bitcoin markets. Our findings also imply that a first-mover advantage would accrue to institutional money that acts to capture profits before arbitrage trading in bitcoin markets gains further mainstream attention.
— Krückeberg and Scholz, page 151