Martin Saps

Decentralised Exchanges: What’s the Point? Part One: Technological Revolutions and the products that nobody uses

April 23, 2018


It’s Deja Vu All Over Again: Technological Revolutions

When technological revolutions occur, why do so many projects end up failing? Perhaps our inability to figure out how and where to effectively deploy new technology results in what Greenspan termed ‘irrational exuberance’ back in 1996. To put things bluntly, humans can be pretty terrible at speculating. Overzealous investors are often too optimistic and too quick with funding, which results in less attention on process, and more attention on the company valuations that shoot up as a result.


In effect, we see a massive misallocation of capital while everyone figures out the uses for this new technology. This is not dissimilar from experiences in today’s token economy. Both VCs and retail ‘ICO’ investors are guilty in this quest to find tomorrow’s winners. Immature and nascent technology, complex cryptography combined with distributed systems all supercharged by poorly-understood crypto-economics makes this industry both fascinating and dangerous. In fact, this ‘frenzy’ characteristic of technological revolutions has been so historically consistent that Carlota Perez modelled it:

[image id=’2008′]

Fig.1: The Perez Technological Surge Cycle. ‘Frenzy’ is Stage 2 of the ‘Installation’ Phase

The ‘Perez Technology Surge Cycle’ has been a useful framework in establishing some parameters around how long technological revolutions take, and how they unfold. Going by her framework, the decade-old lifespan of the token economy would suggest we are still in the installation phase, while a consistent quarter-by-quarter increase in funding would suggest we are in the frenzy stage. To put things into perspective, token-based projects raised almost $3.3 billion (add a further $1.7 billion if we consider the money raised by Telegram Open Network) in Q1 2018, compared with $20 million in Q1 2017, according to our internal research. Roughly speaking, that’s a 165x increase. If history is telling, much of that capital will be sunk into failed and forgotten projects.

[image id=’2011′]

Fig.2. A bar chart comparing quarter-by-quarter funding raises in the token economy

Yet, atop this graveyard of misallocated funds, how do we build the success stories of tomorrow? We can’t know for sure what will and won’t succeed, but we can hedge our bets. In the current climate, we can do this by taking advantage of the open-source environment, which will increasingly allow companies to combine their innovations with those of their competitors. Let’s explore this idea in decentralised exchanges. Why would anyone use them?


Why would anyone use decentralised exchanges?


  1. The ‘honeypot’ problem

Today, most financial exchanges are centralised. This means that money deposited by users is often held in one location. For hackers, that location is known as a ‘honey pot’ — the sweet spot they aim to attack. Centralised exchanges have a history of poor security, and they haven’t performed particularly well in recent years. In the past year alone, the token economy has seen over $1 billion in funds stolen. This problem isn’t unique to tokens. In 2016, the Central Bank of Bangladesh lost approx. $81 million to a malware attack, their holdings being withdrawn from the New York Federal Reserve. It is estimated that throughout 2016, Ukraine and Russia similarly lost hundreds of millions of dollars in related hacks. Increasingly open-source infrastructure online will provide more data for the development of artificial intelligence (AI) systems, including malicious ones. As a result, centralised exchanges are becoming more of a structural risk to the global financial infrastructure. A decentralised exchange would disintermediate funds, leaving no honey pot to be targeted by hackers.


  1. Who needs middlemen?

Having a middleman involves a host of issues: they can be slow, unavailable, and expensive. True, there are centralised exchanges like GDAX which are able to remove exchanges fees. However, not every exchange has the benefit of also being a major market entry point (they also run Coinbase) which is able to charge 1.49% on transactions. The vertical integration between Coinbase and GDAX allows them to rebalance their fee structure, but this is a topic for another post. The point is that most centralised exchanges cannot negate the costs incurred by having middlemen. On that same note, downtime can also be an issue. This was demonstrated earlier this year by the #7 crypto exchange, Kraken. On January 10th, Kraken was scheduled to go down for 2–3 hours of maintenance work. In relentless and volatile crypto markets, 2–3 hours would have been troublesome for many users. The exchange was instead down for 40 hours. In centralised cryptocurrency exchanges, the private key to all of your funds is held by that exchange, so if the exchange goes down, you cannot touch them. This is roughly the same amount of time it took for ripple’s price to half on the same exchange. Though decentralised exchanges can also experience downtime, funds are not locked up and can be transferred at any time. Herdius run through these issues in detail. In short, because decentralised exchange operations occur primarily on smart contracts, there are no middlemen involved. As a result, in decentralised exchanges, any ‘middle-man’-associated issues in speed, unavailability, and fees disappear.

For the most part, centralised exchanges are great. Yet, they have fundamental flaws which result from having everything in one place. These issues are addressed by disintermediating and automating the process. This must surely make decentralised exchanges an idea worth pursuing, right?


If decentralised exchanges are so great, why does nobody use them?

Despite their benefits, decentralised exchanges remain underutilised. At the time of writing, every $1000 traded in token exchanges is matched with only $2 in decentralised exchanges. To put this into perspective, the combined trading volume of the top 10 decentralised exchanges (see appendix) roughly equates to 1/15th of the #10 centralised exchange. This is because despite seeing decentralised exchanges introduce useful new features, they are forgetting to combine with what competitors are offering (that is the spirit of open-source afterall) and what incumbents already offer (how else do you expect people to move from their trusted service providers?). With this in mind, decentralised exchanges currently face three fundamental issues preventing institutional adoption: insufficient infrastructure, a lack of custodians, and the lack of a user-focused interface (UX/UI).

[image id=’2012′]

Fig 3. A pie chart comparing trade volume in top 10 centralised exchanges vs the top 10 decentralised exchanges by 24h trading volume


  1. A lack of infrastructure resulting in too little liquidity (crypto and fiat) and too much slippage

This problem is composed of three core limitations: liquidity issues for crypto and fiat, and a lack of slippage-prevention for large investors.


Problem 1(a): Too little crypto liquidity

Within the token economy, even common words like ‘liquidity’ can sometimes be jumbled up with a crypto-world equivalent. I use the term ‘crypto liquidity’ here to refer to the level of ease (liquid) or difficulty (illiquid) in exchanging tokens for a common intermediary token pair, like BTC, ETH, or USDT. This is important because many exchanges do not actually offer fiat liquidity, less-so for decentralised exchanges. In terms of crypto liquidity, the #1 decentralised exchange commands under half of the demand of the #10 centralised exchange, and less than 1/5th the volume held for any pair on the #1 centralised exchange.

[image id=’2013′]

Table 1. Comparing liquidity depth of the #1 (by market cap) decentralised exchange with the #1 and #10 centralised exchanges based on their highest volume mutual tokens (BTC@$6850)

To those within the token space, this lack of crypto liquidity can be off-putting. This is because on low liquidity exchanges, it is often harder to fill orders, and prices can deviate significantly from the market average. On Lykke exchange (see appendix) for example, EOS is valued at $10.96 compared with the market average of $9.76. Great opportunity for arbitrage? Probably not, considering it’ll be difficult to fill your order. By putting off those within the space, it becomes difficult to attract more liquidity. A lack of liquidity continues to disincentivise users from adopting those exchanges. A negative ripple effect.


Problem 1(b): Too little fiat liquidity

Now this is what the institutional investors are interested in. Real, cold, hard, fiat liquidity. At the time of writing, the crypto market has a capitalisation of $266 billion, roughly 1/24th the size of the $6 trillion handled by pension funds. This demonstrates a stark contrast between the amount held by institutional investors and crypto markets. Given that the lack of liquidity in the entire crypto market has previously been listed as a barrier for institutional investors considering entering the market, how would a decentralised exchange which cannot even match 1/5th (a generous overestimation) of the top centralised crypto exchange handle institutional capital?

“Cryptocurrencies do not have a size and liquidity that is appropriate for institutional asset allocation and the environmental, social and governance concerns of Bitcoin probably rule them out for many pension funds”

Yet, even these low levels of ‘liquidity’ are not quite… ‘liquidity’. They are crypto liquidity. Despite all the advantages offered by tokens like BTC, USDT, ETH, they are not real world trading pairs such as USD, GBP, or EUR. When dealing with anywhere between millions and trillions of other people’s money, as funds tend to do, this can be off-putting. The irony is that institutional investors seek liquidity as one of many criteria to enter the market. Yet, the liquidity provided by institutional investors has the potential to attract many small investors, whose liquidity attracts more investors. A positive ripple effect.


Problem 1(c): Too much slippage

Relatively speaking, when ‘big’ money moves into ‘small’ markets, a lot of price fluctuation can occur. These waves in market price are known as ‘slippage’. For example, an institutional investor wanting to sell one million Ethereum tokens might have a hard time if they are hoping to achieve a decent average on their returns. In wildly volatile crypto markets, they would have to sell their tokens in a short enough time period if they receive a similar price for them. Whether this order is broken into a few moderately large trades, traded in small amounts over a couple days, or traded with another whale, the effects would still be noticeable. Ishtiaq Rahman investigates the specifics deeper, and it turns out we may not need much capital to cause large price shifts. Arguably, this was demonstrated by the recent ‘Mt. Gox dump’, which occurred during Bitcoin’s steep price decline. For those selling, the impact would be a dip in the market. For those buying, there would be a surge in price (making it more expensive to buy). Regardless of the option chosen by the whale, current infrastructure would result in slippage since it was not possible to keep the identity or intention of the investor secret during the transaction(s).


Problem 2: A lack of custodianship

Currently, a handful of whales exist in the crypto space — early adopters, crypto funds, successful traders. However, many of our whales do not compare to real world whales — such as those pension funds we mentioned (the one’s controlling over $6 trillion in assets). For these mega-whales, who are dealing with trillions of dollars of other people’s money, they must be extremely cautious. Because of this, it is imperative that they have a reputable system to operate on. Most crypto exchanges aren’t insured, their security breaches do not inspire trust, and there hasn’t been sufficient legislation in place to protect them in one of these events. This lack of custodianship (for the most part) has contributed to the lack of institutional engagement in crypto markets.


Problem 3: Poor user-friendly interface

Working closely with the blockchain industry, it can be easy to lose track of what is ‘technical’ and what is ‘normal’ speak when it comes to crypto. I read an article on UX in crypto earlier this month, which reminded me of that. Here, Flavio Lamenza mulls over his qualms with the space, and quite plainly, how difficult it is to navigate around some of the most user-friendly applications the token economy has to offer. Websites like Coinbase. What would he say if he tried to use a decentralised exchange? The truth, plain and blunt, is that UX always takes priority. Security is great. Privacy is great. Decentralisation is great. But no company ever got successful by asking their customers to sacrifice what they enjoy. In this case, that would be accessibility, and ease-of-use. If you are going to replace the titans, you better come prepared to match their offer and then some. Jeff Bezos outlines this principle well in Amazon’s mission statement:

“There are many advantages to a customer-centric approach, but here’s the big one: customers are always beautifully, wonderfully dissatisfied, even when they report being happy and business is great. Even when they don’t yet know it, customers want something better, and your desire to delight customers will drive you to invent on their behalf […] Staying in Day 1 requires you to experiment patiently, accept failures, plant seeds, protect saplings, and double down when you see customer delight. A customer-obsessed culture best creates the conditions where all of that can happen.”


Please note: This does not constitute investment advice. Crypto-currencies can be extremely volatile and subject to rapid fluctuations in price, positively or negatively. The content provided in this article is provided for information purposes only.

Stay in the loop

Subscribe to our weekly overview of the crypto market.