Arrow Decentralized Financial Markets
The Arrow platform is inspired in part by Nobel laureate economists Kenneth Arrow and Gerard Debreu who showed prices of financial assets can be reduced to prices of a core set of contracts called Arrow-Debreu securities. The economies of scale suggested by this irreducible set of securities are embodied in the Arrow decentralized financial market (DFM).
Each Arrow DFM is built around an underlier. The underlier can be any quantity that can be accessed by a public API. The DFM contains a standardized set of claims whose payoffs are contingent on the value of this underlier. As an example, a quant user might choose to build a DFM around a portfolio of stocks, and create call options that payoff if the value of the portfolio of stocks is above some level in the future. Similarly, a defi user may build a DFM around ETH and create put option spreads on ETH.
DFMs live on the blockchain network where they were deployed. The Arrow application layer allows users to access, deploy and combine DFMs. The DFM can be used to replicate any claim to the reference assets or their derivative claims, including puts, calls, option spreads and synths. Assets can also be combined across DFMs to create fixed-floating swaps, total return swaps and a complete array of hybrid claims.
DFMs are long-lived, so after initialization their main purpose is to facilitate the creation and settlement of claims written on their underliers. The adjacent figure illustrates that once a user has initialized a DFM around an underlier, say 'U', other users can create and exchange options on 'U' by accessing the initial user's DFM.
In a traditional markets setting, obligations promised by a contract that are contingent on the value of some underlier are enforced by a centralized clearing counterparty or a brokerage service. Brokers typically require max-loss collateralization. In the case of U.S. futures markets, brokers can use a dynamic rule like span-margining. In a traditional over the counter (OTC) setting, settlement is enforced by explicit covenants between counterparties or between a dealer and a client.
In contrast, payments promised by claims written with an Arrow DFM are backed by a decentralized aggregate counterparty (DAC). The DAC stores balance sheet data aggregated at the level of the DFM and comes with standardized methods for handling settlement and liquidity provision. The contingent payments promised by contracts written on the DFM are financed by a special liquidity vehicle (SLV) that incentivizes users to overcollateralize the contingent obligations. Users are incentivized to provide liquidity by earning a pro-rata share of fees generated by asset creation on the DFMs.
The primary pool governed by the DFM's DAC functionality pools buy-ins posted by contract creators in order to cover the call option liabilities at the level of the underlier. In the case of insolvency, the pool pays all obligations up to some level S*. If the pool is not insolvent, writers receive the residual equity at expiration. The threshold S* is pinned down according to a deterministic algorithm.
We couple the DFM's primary pool with a tail pool designed to fund its shortfalls. The tail registry uses an independent SLV to fund the obligations to pay the aggregate counterparty's shortfalls in the event of insolvency. Investors capitalizing this pool are entitled to a pro-rata share of the fees generated by creation of assets in the DFM and their exchange.
The primary pool sets collateral according to the marginal contribution of each new contract to the risk profile of the DFM. This risk-reconciliation circuit is illustrated in the adjacent figure. Given an initial distribution of contract liabilities in the pool, the entropic value-at-risk - a particularly tractable risk measure - can be evaluated. Then, each new contract liability is charged to offset its incremental contribution to the liability distribution's entropic value-at-risk. When necessary, this calculation also triggers a modification to the buy-in rate for the tail pool. The circuit is repeated, keeping the risk profile of the DAC stable over time.
The interaction between the risk measure and the pool is designed to keep the primary pool insolvency probabilities very low. This is a novel interaction between optimization and automated market making to our knowledge.
A key advantage of building asset creation and settlement capabilities on blockchains is that users of the transaction application comprise the transaction infrastructure, greatly reducing barriers to innovation and access. On blockchains like Ethereum and Avalanche, balances of financial assets are associated with an address and transfer approvals are under control of the user who has a cryptographic key associated with their address. Often, secure wallet applications interact with financial applications to automate this approval process.
Advanced-generation consensus technologies such as Avalanche, Ethereum 2.0 and others make the underlying transaction protocol feasible for Arrow applications at scale. Arrow (v.1) is designed for Ethereum and the Avalanche C-Chain, which both interpret Ethereum virtual machine (EVM) code directly. Future versions of the Arrow protocol will be extended to incorporate streamlined virtual machines (VMs) like the Avalanche X- chain and custom VMs that target oraclization of underlier data. The protocol can also be implemented in private networks.
While the DFMs and their progeny contracts live on the blockchain network where they were deployed and instantiated, respectively, a subset of the calculations the protocol needs to make are handled by specialized nodes. For price oracles, we use Chainlink pricefeeds, which can be setup for any price reference with a public API. For our collateralization calculations, we use our own collateral oracles that run reproducible computations using the pool data and the risk measure which both have pointers on chain. Our collateral oracles are called to generate the marginal collateral requirements for new contracts.
Customizability and Extensibility
The scope for customizing assets by combining existing DFMs and making new ones for specific applications is broad. Once a user creates a DFM around an underlier, any other user can build claims on that underlier without having to recreate the DFM. Because users can access any DFM, they will be able to form assets that span a variety of markets without having to create DFMs themselves. Many users may just use and combine existing DFMs.
The ability of users to experiment contributes to the expected success of the platform. There are many examples of useful DFMs to create that we can predict, but we expect users to create and combine novel markets that we cannot predict. That this type of extensibility and experimentation leads to adaptability is an exciting feature of the system.
The protocol is not limited to creating DFMs around financial outcomes in principle. Any well defined outcome space can be built into a DFM. For example, election results, regional temperature outcomes and competitive sports outcomes can all be converted into a DFM. For Arrow (v.1), the subset of these scenarios that can be referenced with public APIs can be used as DFM underliers.
Disclaimer: Nothing in this document should be construed as a market investment recommendation or as accounting, legal or tax advice. This document is a description of technological capabilities and a vision of how these capabilities can flourish. Specific capabilities or reference outcomes may be subject to regulatory and/or licensing constraints. The designs reflected herein are subject to change.