Paradigm Labs, Contextualized

Liam Kovatch

Published in

Paradigm

6 min readNov 6, 2019

Paradigm Labs is building liquidity solutions for decentralized finance.

TL;DR

Core pieces of market infrastructure, including exchanges, are being redesigned and rebuilt from the ground up to endow individuals with true financial sovereignty.
Resulting from technological limitations and diverse trader demands, a heterogenous marketplace of non-custodial trading venues will likely emerge.
Paradigm Labs is developing two trading solutions: Kosu and Zaidan.
Zaidan is a non-custodial request-for-quotation trading system for price-prioritizing traders.
Kosu is an experimental liquidity aggregation network aimed at traders that prioritize accessibility and decentralization.

Background

In the Information Age, individuals will be able to use cybercurrencies and thus declare their monetary independence. — The Sovereign Individual, pg. 24 (1997)

During the late 1990s and early 2000s financial markets witnessed a profound structural evolution as exchanges, traders and regulators alike swiftly adapted to the world’s sweeping adoption of modern communication technology. Overall market efficiency saw dramatic improvements as transaction costs fell and liquidity increased.

Unfortunately, these advancements in efficiency have been available to only a privileged few and, based on present incentives, it’s hard to imagine this changing under the current framework of control. Today, the diversity and competition that once drove fervorous evolution has consolidated. Three exchange groups now control 12 of the 13 registered US equity exchanges. This consolidation ultimately represented the conclusion of the first great electronic market evolution. A catalyst for substantial advancements to these markets is now largely absent. As such, further improvements in the structural efficiency of these “first-generation” electronic markets will likely be incremental.

As with many stepwise evolutions, electronification was driven largely from the periphery, in this case by talented programmers with shrewd foresight of communication technology’s effects on market function. At the time internet usage was growing exponentially and stock ownership had become almost entirely digital. Instinet had established itself as a popular electronic alternative to traditional trading venues. Even Nasdaq, a relative incumbent, was moving towards electronification having responded to Instinet’s success with buffed electronic trading capabilities of their own in 1984. The pieces were in place for an evolution and the few that recognized it would shape the future structure of markets for the decades that followed.

In many ways the climate of today feels similar to how I imagine that of the late 90s must have felt. The aim of the next great financial evolution will of course be different, but subtle signs of impending change are apparent. Across the globe, more and more citizens of dominative nation states are demanding financial sovereignty. At the same time, technology that will make the demands of these individuals possible is maturing. Entirely sovereign currency and trustless computing platforms are steadily making both technological and social progress. Exciting experiments in entirely trustless financial infrastructure have been developing over the past few years and have spawned a sovereign financial ecosystem that is becoming increasingly approachable. All of this and more suggests that the seeds of the next financial market evolution have already been sowed.

Paramount to a mature financial system is well functioning exchange infrastructure. In an attempt to better predict the future market structure, it makes sense to closely consider some notable qualities of the most mature financial markets of today. One quality that particularly stands out is the relative heterogeneity of trading venues that together compose the landscape of the US equity marketplace. Today there exists 13 major US equity exchanges and 57 registered alternative trading systems. Without reasoning about the motivation behind the current structure, it at least seems reasonable to conjecture that the future marketplace will be similarly heterogeneous. Diversity in trader demands combined with fundamental limitations in distributed computing provide a basis for this expectation. Early developments in the non-custodial trading landscape seem to validate this thesis where decentralized exchange solutions such as Uniswap, 0x relayers and Oasis coexist by providing distinct advantages to specific demographics of traders.

Our Theses

Trading will largely become non-custodial.

This thesis can be motivated from a few different angles. Perhaps most readily, non-custodial trading reduces the number of roles that can be filled by rent-seeking intermediaries. By reducing an arbitrator’s necessary responsibility in the process of trading, we expect to observe a reduction in the cost of trading. In addition, we anticipate a reduction in indirect costs as the custodial risk for traders and trade facilitators is largely eliminated. More formally, as the transaction costs of non-custodial trading systems approach those of centralized solutions, exchange entities will adopt the former to reduce liability costs. Finally, for users of token-based dapps, on demand, physically delivered trading is a requisite. As these applications gain traction, so will non-custodial trading solutions.

We expect there to be profound impacts from the transition towards non-custodial trading. Generally, we suspect market structures will undergo a fundamental transformation, not too dissimilar in magnitude to the ECN revolution of the late 90s/early 00s. Completely new trading systems will be developed in the context of this evolution. Trading systems that offer experiences competitive with those of current centralized exchanges will require time and an iterative approach to develop, but once matured will dominate the trading landscape.

Access and price are the two most fundamental characteristics of a competitive non-custodial trading UX today.

Current non-custodial exchanges can generally be sorted into one of two categories. The first are price-prioritizing which includes the likes of dYdX and Paradex, among others. These exchanges include a centralized operator at some point in the trading process often to coordinate and match trades. Exchanges of this category, while providing competitive pricing, make tradeoffs ranging from strictly limiting the tradable universe to excluding traders in specific jurisdictions to even requiring KYC’d account registration.

The second category of non-custodial exchanges are access-prioritizing systems. Generally, exchanges of this category are “fully-decentralized” meaning there is no single operator. In addition to making them more robust to censorship, full-decentralization generally allows this class of systems to facilitate contract-fillable-liquidity and trading for long-tail assets not listed on exchanges of the previous category. To provide these unique qualities access-prioritizing exchanges generally make a performance tradeoff by relying on blockchains as the sole arbiter in the exchange process. Some examples of systems that can be classified as access-prioritizing include Uniswap, Radar Relay (specifically with 0x-mesh as a liquidity backbone), Kyber and Oasis.

Even today, there is significant diversity among non-custodial trading solutions. While some will be more successful than others, there seems to be very little reason to suspect that this diversity will be completely eliminated in the long run. At Paradigm Labs we thus imagine a mature heterogeneous non-custodial market landscape where distinct trading networks/solutions catering to specific demographics of traders will coexist. As non-custodial trading grows, traders’ preferences will inevitably become more demanding resulting in additional DEX solutions that prioritize other characteristics (responsiveness etc.).

Note: The categories defined in this section are motivated by trader demographics and neglect subtleties between liquidity models, and other exchange elements within each category.

Our Solutions

Our solutions are thesis-driven, focusing on non-custodial trading in both price and access critical domains.

Zaidan

Zaidan is a non-custodial trading system built on 0x that uses a unique dealer model to provide traders with a simple, price-prioritizing experience. Zaidan’s primary innovation is its application of ZEIP-18 to a unique interaction model wherein the maker, which we designate the dealer, is responsible for execution in addition to order creation. Dealers and takers communicate directly via a request for quotation (RFQ) scheme. Upon solicitation from the taker, dealers provide restricted signed maker orders that represent soft-commitments to trades. If the requesting party agrees with the provided quote, they can (quickly) generate and respond with the signed 0x transaction message. The dealer, prior to submitting the transaction for execution, verifies the validity of the 0x transaction message which includes enforcing previously provided “soft-parameterization” constraints (akin to last-look). This scheme shifts optionality from the taker to the maker and results in competitive prices across trading sizes.

Learn more

Try it out

Kosu

Kosu is an experimental decentralized liquidity aggregation protocol that provides traders a consistent, curated source of off-chain orders. To do this, Kosu introduces a novel curation scheme for validators who are responsible for hosting a decentralized, proof-of-stake, order aggregation network. Kosu creates a public venue for makers to broadcast orders to potential takers, with the assertions that they will always be able to get an order included in a block for a certain token-based lockup, and that all takers on the network will see their order at a similar time.