How Long Will Cryptocurrency Arbitrage Last? - Mitchell Dong
This is the most commonly asked question by our investors in our cryptocurrency market funds.
Not only is this question asked today, but also when I ran a hedge fund that did mutual fund market timing in the 1990’s.
There was an institutional flaw in the way that US mutual funds were calculating their NAVs. At the close of the NYSE every day, funds valued each of their equity positions based on the last traded price. But, if the funds held positions in Hong Kong, for example, the last traded price was 12 hours stale.
So, you could buy Hong Kong equities at yesterday’s price! How could you possibly go wrong? 61% of our trades were profitable with a 2:1 profit loss ratio.
When I explained this to our previous investors, they would ask, “the hole is so gaping, how long is this inefficiency going to last?” The flaw was prevalent long before I entered the business in 1995. And it lasted 9 years while I ran Chronos until 2004. Sometimes it takes a long time before certain inefficiencies get “arbbed out.”
A classmate of mine started trading FX in the mid 1970’s when the British pound (symbol: GBP) first started trading in Chicago. He called long distance to London every day from Chicago, and was able to arbitrage the GBP traded in the USA versus London, just for the cost of a long distance call. He told me that he did this arb for over a decade. He also did this when the Japanese Yen first started trading in the USA and that lasted for many years too.
I remember when I traded ADRs in the US against underlying stocks globally, in HK, South Africa, Europe, etc.There were two separate investor types and cultures which resulted in different prices. That arb lasted for a decade.
When we first entered the crypto arbitrage business in 2014, there were large arbs between bitcoin traded in the USA and China. The arbs were as large as 5–10% at times.
In 2016–2017, the arbs shifted to bitcoin traded in USD versus the Korean Won and the Japanese Yen. The arbs were typically 10–20% and at times, as large as 50–100%. But those arbs are long gone, and now, only appear occasionally for a few minutes, or a few hours and are on the order of 1–2%. The arbs we did a few years ago were “arbbed out” but we are always on the lookout for the new emerging arbs. That’s our raison d’etre.
In 2018- 19 we saw the arbs move from the spot market to the derivatives market. When retail investors were excited about the market, they placed their bets on future contracts rather than the spot market because they could get more leverage and the trading fees were much lower. We saw pretty significant arbs in certain altcoins too, such as EOS, LTC and BSV, on the order of 5- 20% between their spot prices and their futures prices which expired in a month or quarter ahead. We took full advantage of this.
Along with the growth of the futures market, we saw the growth of daily swaps or perpetual contracts, which are futures without an expiration date. First created by the Bitmex exchange in Asia but then copied by the other large Asian exchanges, they were a product designed to mimic the spot price, but in the form of a derivative. They have a mechanism which keeps the price in line with the spot price, e.g when the swap price gets higher than the spot price, the owner of the swap pays a premium funding rate and vice versa. This created a “funding rate arbitrage” in these derivatives.
What is the future of crypto arbitrage? Our mission at Pythagoras is to conduct research to keep looking out for the next arbitrage. Given the fact that crypto is a new and evolving market, it seems to constantly have new arbitrages. For example, here a few new arbs we are monitoring:
We believe that the crypto options market could be a big market in 2020–21.
- CME launched its crypto options in Dec 2019 and Bakkt, a sister company of the NYSE and subsidiary of ICE, announced that it will launch crypto options shortly
- OKEX, one of the largest exchanges in Asia just launched its crypto options in 2020
- The other large Asian crypto exchanges such as Binance, Bitmex, Huobi and FTX are all working on launching their options products
- In the West, Deribit and LedgerX have had options on their exchanges for the past few years; but volumes are trivial. But, we expect that volumes on all the options exchanges will grow exponentially over the next year or two or three.
We want to be early as we were for the spot and futures trading.
Other countries are also jumping on the bandwagon. We just are particularly keeping our eyes peeled on Southeast Asia. Given that retail investors in China, Japan and Korea are already active, why wouldn’t the rest of Asia, particularly SEA, join the party? Favorable regulations on crypto are afoot in Vietnam, Philippines, Thailand, Malaysia and Indonesia, just to name a few.
Africa and South America are also on our radar. If arbs appear in these local markets, Pythagoras is ready with the appropriate infrastructure to jump in. Setting up in foreign lands is not easy. You need local subsidiaries which can pass the local AML KYC tests of the local banks and exchanges. They must also be locally tax efficient. This requires good advice from local tax and legal experts. If it were easy, the arbs would not exist.
DeFi is another area of interest for future arbs. DeFi includes fixed income markets in crypto. There is a growing and robust market for lending crypto or USD for interest, e.g. people rich in crypto can lend their crypto and collect interest. Given the dozens of lenders and borrowers and platforms which facilitate this business, it’s not surprising that the interest rates could vary significantly between persons, countries and platforms. For example, we see opportunities to borrow at 5% and relend at 7% or higher.
Decentralized exchanges present another opportunity. We currently trade mostly on centralized exchanges where we deposit our crypto and fiat on the exchanges. Decentralized exchanges are peer to peer matching platforms where users don’t deposit any assets on the exchange, but rather just exchange assets between users, without any custody or deposits to an exchange. This reduces the risk stemming from a single central custodian getting hacked. The counterparty risk is shifted to the other users on the platform. We see arbitrage opportunities between and among decentralized exchanges and centralized exchanges. As of this writing, the price of some cryptos are 1% higher on certain decentralized exchanges versus the price on centralized exchanges.
Our role at Pythagoras is to constantly seek the next big arb opportunity. Over the past 5 + years, we have moved from arbs in China to Japan and Korea, from spot to futures arb, to arb in perpetual contracts and many more specialized arbs. We are optimistic that with a young and emerging market in crypto, there will be arbs for those who are prepared.
There are constantly new exchanges, new coins, new regulations, new money, new people in this rapidly growing industry. When all is “arbbed out,” then we will move onto the next emerging industry. This is in our DNA. Please join us in this exciting area.
