Defining Institutional DeFi:
Firstly, and for the purpose of this series, I would define ‘Institutional DeFi’ as:
A. Crypto Native Businesses
Institutions / businesses already in crypto whose organisations are heavily exposed to the asset and its markets and therefore likely ‘underbanked’ in a conventional sense. Because of the degree of their crypto nativeness they are excluded from access to typical banking and financial services like lending / borrowing and navigating a fiat / crypto world challenging;
e.g. DApp startups, Middleware service providers, crypto day traders, VCs or foundations
B. Financial firms
Sophisticated and professional investors from traders to medium to large financial institutions that would not otherwise be interested in ‘crypto’ both generally as an alternative asset class, but especially its subset ‘DeFi’, either directly by taking custody or indirectly through derivatives and ETFs;
e.g. professional traders, family offices, hedge funds, banks, pension funds
C. Fintech firms
Making proven DeFi financial products easy to plugin to Fintech brings a B2B2C approach to onboarding new Retail demand through regulated on and off ramps. Given most Fintech companies are loss making with heavily subsidised user acquisition the ability to offer the kinds of products and yield (even as it normalises down from today’s 7-10% on stablecoins) seen in DeFi would not only make them profitable but be a further incentive for new user acquisition.
e.g. Fintech startups like Square
D. General Commercial Underbanked
There is an argument to say there could be a market for defi today for sections of both 1st, 2nd and 3rd world economics that are, for their own particular reasons, underbanked, typically sole traders / SMEs, in particular those that can not otherwise access affordable debt
e.g. sole traders, SMEs, entrepreneurs, gambling / adult industry
Sequencing Institutional DeFi Adoption:
All the stakeholders outlined above operate on a spectrum where it’s more likely DeFi will be adopted by crypto natives and smaller day traders followed by the underbanked SME world and eventually large financial institutions over the next decade. It could be said the easiest low hanging fruit are the 95% of institutions already exposed or directly using crypto that are not using DeFi.
More widely adoption will be driven by how native a given stakeholder group are to crypto and/or how underserved they are by the existing financial system leading to either high levels of risk appetite or desperation. But it also comes down to technical proficiency and the DeFi industries ability to lower its thresholds as well as be able to handle the typical size and volume of transactions an institution and their counterparties require.
It’s perfectly natural that DeFi and crypto today are generally just too small for most institutions to even bother upgrading their knowledge and understanding let alone IT systems, and even if they could the markets are likely too thin for then to deploy meaningful amounts of capital into. That said, DeFi adoption beyond today’s 5% of crypto has to start somewhere and there is clearly a strong financial incentive to solve its problems in one of the most innovative and fast paced fields of financial technologies. Importantly, you simply can not underestimate the superpower of DeFi’s composability which means no one entity alone needs to solve all these problems but each can be tackled by a specialist protocol which quickly combine into a stack that together to create a powerful form of open finance.
Below are several innovation triggers which when applied to DeFi promise to kick start what I call DeFi 2.0.
As discussed in the previous article (linked above) every hype cycle is usually a consequence of a combination of innovations coming together in a timely way. I propose the innovations that will drive demand for Institutional DeFi, in a DeFi 2.0 context, are as follows:
(Supply) NFTs, Confidentiality Solutions, DIDs and Verifiable Claims, AEAs + (Demand) dPrime Brokerages & ETFs
A non-fungible token (NFT)
Also known as a ‘nifty’, is a special type of cryptographic token which represents something unique; non-fungible tokens are thus not mutually interchangeable by their individual specification’
The two Ethereum NFT standards allow for all kinds of unique programmable digital goods of various characteristics beyond the use case of a currency. In the previous Retail DeFi article we discussed them in the context of the ‘consumerization of crypto’ including:
social currencies, rewards, collectibles, access tokens, digital art, loyalty points, and digital to physical goods redemption.
But in an institutional context; it is mostly digital representations of / or access / ownership rights to something unique, entirely digital or as a twin to physical assets (eg. property like a car or real estate).
The most important question is when is an NFT a security, and therefore a ‘security token’, which creates considerations around custody and activities that may make an intermediary a regulated counterparty. All this places a burden on the operator of a protocol, no matter how decentralised they may claim to be, to comply. A process which invariably seeks to make them more centralised.
Confidentiality solutions: private data, transactions & compute
DeFi protocols up to now have been radically open, like the public and unpermissioned platforms they are built on. At best pseudonymous, all transactions and balances are visible and traceable to the public. This is partly due to the ethos of the space, and partially due to sheer necessity: there weren’t any technologies available for on-chain confidentiality.
However, there are now several tools available for confidential transactions, some already on mainnets and others getting close. On-chain confidentiality will remove another barrier for institutional players to get on board where there is a business requirement for degrees of privacy.
Privacy primitives become common
Various forms of privacy primitives such as zero-knowledge proofs (ZKPs) have gone from theory to becoming practically usable both on-chain and off-chain, and are already being deployed in DeFi. Out of the enterprise camp there’s EY Nightfall, enabling standard fungible and non-fungible tokens to be transacted on the Ethereum blockchain with complete privacy. The drive for privacy and scaling has accelerated innovation in cryptography, which has led to different forms of ZKPs and similar techniques like ‘bulletproofs’ with ever improving performance. ZKPs are now finding their way into DeFi projects such as Loopring, and indirectly through scaling solutions using zkRollups.
Transaction mixing techniques can be used to conceal the history of a particular token.
L2 scaling solutions
Layer 2 scaling solutions can offer increased privacy over layer 1 networks by design: transactions don’t end up in the layer 1 blockchain. Some approaches make significant confidentiality possible. Notably Aztec recently released their v2 testnet, enabling private transactions with significantly lower gas costs than regular transactions, compatible with the current Ethereum mainnet.
General purpose confidential computing
Beyond keeping some data confidential sometimes, new crypto networks enable keeping everything on-chain confidential with built-in privacy functionality. The Secret Network (previously Enigma) has general purpose confidential computing on its mainnet today called ‘secret contracts’ and Oasis Network is close to launching its mainnet with many privacy features at its core.
Decentralised Identifiers (DIDs) & Verifiable Credentials
DIDs are a new type of identifier that enable verifiable, decentralized digital identity. A DID identifies any subject (e.g. a person, organization, thing, data model, abstract entity, etc.) that the controller of the DID decides. DIDs have been designed so that they may be decoupled from centralized registries, identity providers, and certificate authorities. The design enables the controller of a DID to prove control over it without requiring permission from any other party and is a standard being rolled out at W3C, The World Wide Web Foundation .
When combined with DIDs, verifiable credentials allow information to be exchanged where the provenance and integrity of that information can be verified (ie the information can be traced back to the issuer (via their DID) and verified as unchanged (ie prove that the DID’s private keys signed it).
E.g. The Sovrin Foundation and Evernym have been working on DIDs for several years, alongside IBM / Hyperledger and Microsoft in a public / permissioned context and there is now a token optimised network being rolled out called P-DIDI with an open RFP for founders
Autonomous Economic Agents (AEAs)
An intelligent agent acting on its owners behalf with limited to no interference of that ownership entity, and whose goal is to generate economic value for that owner. In a blockchain context, an AEA can be considered allowing for an order of magnitude more complexity on top of the smart contract logic and execution layer of DeFi.
In the same way ERC20s and smart contracts have allowed a composable stack of specialist DeFi protocols to exist and interact on top of Ethereum to form a kind of ‘Money Lego’ AEAs could be considered an ‘AI Lego’ overlay. Whereby they complement Ethereum and other layer one protocols as a layer two technology which:
- a) simplifies user experience through automation
- b) supports modularisation and reuse of complex decision making and machine learning capabilities off-chain (optionally with on-chain components)
- c) allows for proactive behaviour facilitating autonomy (vs smart contracts which are purely reactive to tx input/contract calls)
This is highly relevant to DeFi where there is an increasing requirement for complex off-chain logic in addition to all the on-chain smart contract logic and where you need to proactively “do stuff” in response to events external to the blockchain typically via oracles. Without the same open and accessible off-chain lego, we now have in the form of smart contracts on-chain, we will likely see DeFi look more and more like CeFi as it matures.
E.g. Fetch.ai a leading project working on cross blockchain AEAs and who have produced multiple papers on the promise of their technology.
Power and limitations of DeFi
I have previously described DeFi as an aggressive game of ‘hyper competition’ where its distributed system of participants seek equilibrium between yield and efficiency in direct contrast to the existing financial system. Ruthlessly removing inefficiencies and aggressively seeking yield, in a very fluid market environment.
Clearly a new financial system like this offers the potential of significant efficiency gains, but also new possibilities for an increasingly digital economy. However, one person’s inefficiency is another’s business or job and so DeFi is and should be treated as both transformational and disruptive.
Equally, the nature of DeFi’s relentless and fast paced innovation happens precisely because of its permissionless and open source nature. Anyone technically able can launch or copy and fork a new unaudited, and unregulated smart contract or protocol as a financial product or service at a global scale, which (at least in principle) can subsequently act with degrees of independence of its creator/s.
The question is to what extent will regulators allow this to continue to happen is to be seen. But the meta question is to what extent can they stop it, or at least find ways to contain it, versus engage positively with it or more succinctly: is there a threshold where its net benefit becomes irresistibly better than the current system in encouraging innovation, competition and economic expansion?
Many will dismiss DeFi as an edge case much in the same way they did Bitcoin or Ethereum but regulators have retrospectively come to see both as net beneficial and even declared them as operating ‘’outside of their jurisdiction’’. In fact, they have taken them as inspiration in the development of a more centralised and therefore controlled equivalent in CBDCs (Central Bank Digital Currencies).
So in fact when we talk about DeFi we are talking about how today’s dominant top down fiat system interacts and merges with bottom up crypto. The question will be whether or not there will be bridges into and out of their jurisdiction that will allow for free flows of capital.
And clearly degrees of decentralization are seen as not only beneficial but preferable when compared to parallel approaches such as Facebook’s Libra.
The question is: how can DeFi onboard new institutional users to benefit from its innovations, without becoming too centralised in the process?
Below I have laid out a series of opportunities in DeFi which leverage combinations of the key technology innovations listed above that promise to both increase liquidity and the institutional demand and usage of DeFi but in a sufficiently decentralised enough way to avoid censorship for it to still qualify as ‘DeFi’ at least on a sensible spectrum.
Lowering barrier to participation
Generally speaking institutions report they aren’t coming into Defi because of a lack of compliance, institutional experience and security / risk controls. But also because of (a lack of) accessibility and the unfamiliarity of the new asset class on offer. So I will use these barriers as a starting point to categorise each opportunity.
These are all opportunities we are or would like to actively invest in. If you are building a startup in any of these domains get in touch we would like to invest.
KYC & AML Passive participation
The primary concern of institutions is they want surety their counterparties in DeFi are legitimate and compliant actors. This manifests at two levels:
- KYC / AML at protocol side
- KYC / AML on sale execution side
Unless resolved this creates two extreme silos; one where regulated centralised entities perform compliance and the other where there is none. The former loses many of the benefits of DeFi’s permissionless nature and the latter is not something institutions can or would engage with. The goal should be a single system and pool of liquidity (capital and assets) which can automatically recognise the status of a wallet’s owner and advise their counterparty as to recommended permissions.
Currently, if you are building a decentralised protocol and wish to carry out KYC and AML on users and their funds you become exposed to becoming a counterparty yourself which would force you to become a regulated central party.
One approach would be to let DeFi users prove things about themselves such as KYC and AML using DIDs and verifiable credentials. Using zero-knowledge proofs, the users’ privacy is preserved, yet DeFi protocols and applications by adopting the standard can provably ensure all users are compliant without them directly being the ones carrying out the compliance function themselves. Through a credentials marketplace, DeFi users can choose the issuers they trust to perform KYC/AML checks for them. Institutions could choose to trade only with those counterparties that match their criteria, without the details of those counterparties being exposed.
However, a big factor in the financialization of assets and rise of DeFi will be providing new mechanisms around the transfer of liability for third party reliance from a compliance context. The challenge being the ultimate responsibility for Customer Due Diligence remains with the Financial Institution (FI) relying on the third party (identity verifier and credential issuer which provides insights about the identity holder rather than data). The FI remains criminally and civilly liable if the third party messes up. This may require the third party to be a regulated FI institution itself (reducing or eliminating the criminal liability) and for the civil risk to be sold to an insurer (who relies upon regulatory and likely auditor oversight).
Thus it enables essentially a risk transfer that makes it easy and seamless to rely upon third party provided identities. This will allow people to verify their identity once and use it in many places greatly reducing the friction created due to global compliance requirements.
(2) Risk Management
Insurance / Reinsurance as NFTs
There is huge risk in DeFi both from a custody and security perspective with many layers of unaudited code and opportunities for malicious actors to exploit a financial system in entirely new ways such as ‘flash attacks’. DeFi needs a native form of insurance contract to help its participants minimise their financial exposure to risk where it can’t be removed technically.
Like most nascent industries the existing insurance industry would struggle to price risk and it would not initially be big enough to justify the investment of time and money to develop the expertise. Historically in these instances communities insure themselves via mutual organisations which we are beginning to see with projects like Nexus Mutual with policies like; smart contract risk, hacks etc
One of the more useful areas to explore here is parametric insurance which maps quite well to smart contracts and simplifies things considerably. Basically a parameter based with a specific payout on a triggering event.
This kind of fixed payout contracts allow people to readily buy the risk against an event occuring. The key questions are how to govern the inputs such as oracles to validate the event if the data source is not contract driven itself (as covered in Trustful Oracles below).
However, these could also be issued as NFTs, which can be underwritten, bought and sold in an open reinsurance market based protocol, and potentially collateralized with loans and underwritten (ideally by multi-asset baskets leveraging staking like methods) to back risks.
Not only does this solve a problem for crypto and defi but it could be the kind of asset that draws in institutional investors and finance and lead to the transformation of insurance and re-insurance more generally.
Eg Nexus Mutual, Paragon Brokers, Marsh’s DART
Administration / Custody
Currently custody is either something you do as an organisation yourself or you outsource to a centralised party. It is one of the main barriers to institutional adoption. The more complex the organisation, and the more fiat based they are, the more complex the permissioning of wallets. Self-custody allows for full control and usage of funds in DeFi and other on-chain protocols, but requires significant ongoing in-house resources. A multitude of centralised custody providers has been established over the past years, some of them institutional grade, which can offer insurance and fine-grained permissioning but only allow access to a limited number of yield generating opportunities that the custody provider supports, like simple staking.
There is the possibility of leveraging DIDs for permissioning of ‘organisation wallets’, containing digital portable credentials that confirm an organisation’s identity and verify the authority of employees and other representatives to act on behalf of the organisation. These credentials can be used to securely identify authorised representatives when they execute an increasing number of digital business activities such as the movement of crypto assets but also approving business transactions and contracts.
Work is already underway in this area with initiatives like GLEIF.
For institutions to trust more economic load to DeFi, and therefore automation via smart contracts, they need increased surety in oracles and their truthfulness to achieve compliant levels of reporting should they be / become a regulated entity, but more generally security and resilience to manipulation.
For example the most likely way to onboard trillions of dollars into DeFi by asset managers would be via ETFs (Exchange Traded Funds) that allow institutions to not have to take direct custody of the underlying asset but trade in managed instruments like indices which could be built on top of protocols like Set Protocol.
And whilst there are a handful of regulated crypto indices, regulators need to be sure of the quality and integrity of the data that informs them, and in a DeFi context goes into the oracles that would inform them. Innovations in crowdsourcing and data markets can leverage the power of the crowd to curate and validate data sets provenance and integrity secured through game theory.
e.g. DIA are building truthfulness for institutional grade oracles by leveraging crowdsourcing to source data feeds as well as ensuring integrity through crowd-validation.
Native Credit Scoring
If users could attest things about themselves and their assets, or that of their organisation, via verifiable credentials linked to DIDs as NFTs a native onchain credit profile could be built and credit scoring could even be carried out by staking from their network to allow for risk adjusted credit profiles. These could be further augmented by oracles to prove things off chain.
Analytics & Fundamental Analysis
As the complexity in the tokenomics of DeFi increases, performing fundamental analysis gets harder and requires new analytic approaches leveraging the decentralized nature of crypto-networks and the power of automation and artificial intelligence.
To be adopted by CeFi entities financial security services within major private and public banking institutions require innovative monitoring solutions to manage structural financial and compliance risks around DeFi. To allow for the complexity of DeFi tokenomics, they need modeling tools to understand, analyse and monitor the activity around major dApps and associated assets.
Sophisticated crypto-assets analytics like Nyctale are needed to monitor investment and usage behaviors and on-chain activity patterns around DeFi services where institutional adoption requires a qualitative and quantitative understanding of economic and business mechanisms at stake.
Commodities / Derivatives
Most institutions will only come into DeFi if there are assets they know, understand and can price such as real world commodities, derivatives or things like real estate. Projects that can create synthetics of these assets that can be freely traded in DeFi and benefit from its efficiencies will bring with them new demand.
Income bearing NFTs
The problem with DeFi yield, as outlined in post (1/1) was the yield was not linked to actual income, rather a form of subsidised user acquisition and therefore unsustainable. However, there are new forms of income bearing instruments wrapped as NFTs coming to market such as invoice factoring or SME wrapped debt.
For long term compliance income bearing NFTs might themselves require a given wallet to have been whitelisted, linked to or signed by a verifiable identity to create more of a fluid system of compliance management depending on the associated risks and thresholds via decentralised compliance.
How do we swap STOs, CBDCs and crypto and vice versa? If NFTs can each carry a permissioning as to who can use it e.g. a wallet unless the owner has provably passed KYC they can exist in the same decentralised environment and be freely exchanged with one another in pairs or mixed into baskets and traded.
Equally, the ability to create investment pools and fund structures will allow for greater holdings of underlying assets and potential for algorithmic market making between them to provide liquidity for swaps including managed (Mutual Funds, Investment Funds) structures whether managed by individuals, entities or through collective governance and unmanaged (ETFs, ETNs) which track indexes and the value of underlying assets.
Access tokens, give access to digital vaults with collateral inside. Rather than a digital representation of collateral or a right (potentially treated as a security), NFTs could be access tokens to underlying collateral and be burnt / consumed when used. Legal corollary would either be rental arrangement (temporary ownership) or a trust arrangement (which separates legal & beneficial ownership). So access token allow the holder to rent collateral for a certain period (followed by forced return); and/or grants temporary legal ownership (where beneficial ownership stays with original owner).
(3) Accessibility Usability
Administration: Accounting & Taxes
Because institutional DeFi’s users today are likely crypto native companies or small boutique / individual day traders using increasingly complex multi-protocol products the administrative burden of filing end of year tax returns alone will put off many or limit their trading activity and frequency.
This is solved by projects like Cryptio an accounting API that integrates into existing accounting solutions.
CeFi to DeFi Bridges
Traditional finance has amassed decades of complex legacy IT systems that are often antiquated and difficult for newer external technologies to directly integrate with. Requiring new bridges to be built between CeFi and DeFi. Open source BaaS solutions for banking like Corda have been custom built with financial institutions over several years, through consortia initiatives like R3, and now allow for new institutional-grade DLT connectors to simplify integrations between legacy systems and newer DeFi protocols.
Projects like Bond180 and Alkemi allow for CeFi to bring liquidity to DeFi and offer financial products
UX / Prime brokerage
Much of what has already been discussed somehow needs to be aggregated into UXs that abstract complexity. This is especially true in what has been called ‘deep defi’ where a financial product is dependent on multiple defi protocols bringing increased complexity and risk and if you consider the longtail of smaller institutions described at the beginning including smaller teams of traders.
Projects like Key Tango and Alkemi are solving the UX problem of deep defi products turnkey ‘defi market’ liquidity and leverage for our institutional customers
(5) Market Efficiencies
Liquidity and AMMs
Crypto, and therefore DeFi, as a new capital market is still generally speaking inefficient with levels of liquidity too low for most institutions however innovations that address this problem that can aggregate liquidity in a non custodial way across wallets, centralised exchanges and DEXs enable ‘liquidity super highways’ that promise to catalyse the growth of DeFi tapping into dormant pools of assets in crypto.
E.g. Alkemi have built liquidity super highways to bridge cefi and defi with a growing consortia parties
Currently AMM (Automated Market Makers) one decentralised DEX (Decentralised Exchanges) alternative to order book matching seen on centralised exchanges suffer from several inefficiencies. Most notable is the issue of impermanent loss. Put simply, a liquidity provider would tend to lose out if the price of the pair changes significantly. This can be compensated for through fees and in some cases is subsidised through the issuance of rewards, but this is a factor that requires active monitoring and management by liquidity providers with risk controls.
Another point that is often overlooked is that price discovery in AMMs occurs through arbitrage often between more liquid centralised exchanges and the less liquid DEX. Traders are profiting from this arbitrage and thus the price discovery is costing something that could be seen as an inefficiency. There is innovation with new types of AMM algorithms being tried but this is an experiment still in its infancy.
Equally, maturation of derivatives markets and fund structures will bring more liquidity to the market. And there is also potential for markets themselves to drive market making through a ‘liquidity mining like’ activity that rewards participants for how much, how often, and how broadly they buy/sell as a mechanism to drive higher liquidity amongst illiquid assets.
And there is the growth of what has been called “transaction scoped credit” – the ability to leverage flash loans to back a transaction that can help arbitrage price differences across assets and add liquidity.
Dark (Secret) Pools
Today much of the institutional interest is focused on acquiring assets to diversify portfolios so there hasn’t been great demand for dark pools as of yet, particularly given the limited market size and shortage of large holdings across a broad array of assets.
As the market matures and has a wider array of widely traded assets the demand for the ability to do an off market swap of assets will grow, particularly as there are larger fund structures that need to quietly move into and out of positions in large volume without telegraphing their intentions.
Again one example is Secret Network, a privacy-preserving smart contracting platform that enables use of synthetic assets, such as ETH, as privacy tokens. Privacy tokens (SNIP-20), combined with privacy preserving AMMs, order-book trading and auction tools build the basis for decentralized dark pools.
This is not a new thesis but one that has evolved from a powerful feedback loop with a maturing portfolio of over 30 startups. It’s also by no means a comprehensive list just ideas and observations from across the Outlier ecosystem as we collectively seek out to build this bottom up permissionless capital market. We have almost definitely missed many themes and opportunities. So if you are working on any of these challenges directly or indirectly please get in touch, especially if you are at the pre-seed / seed stage and fit with our accelerator dedicated to DeFi and Web 3.
Portfolio in this segment:
Below is a summary of those projects mentioned in this article.
- Fetch.ai – AI Lego for Money Lego
- Metalex – A tokenized commodities platform
- DIA Data – Verified and Transparent Oracles for the DeFi Economy
- Key Tango – Prime brokerage for DeFi
- Alkemi – Decentralised Finance for Centralised Institutions.
- Bond180 – Corda based CeFi bridge for DeFi fixed income
- Nyctale – AI-powered crypto-assets analytics to manage structural financial risks in DeFi
- Cryptio – Accounting solution for crypto assets
- Evernym – The Self-Sovereign Identity Company
- P-DIDI – The token optimised verifiable claims market protocol
- Secret Network – Decentralized applications to perform encrypted computations.
|If you are working on something that accelerates this vision please APPLY to our 3 month accelerator program BASE CAMP to join this wonderful and highly synergistic portfolio.
If you are a later stage project but want to explore the power of NFTs to gamify your platform we have a ‘12 week token fast track’ program for basecamp applicants who meet a readiness criteria
I’m grateful to all the contributors from both with Outlier Ventures, our portfolio and wider network.
Andrew Tobin (Evernym)
Aron van Ammers (Outlier Ventures)
Blaise Cavalli (Nyctale)
Carl Bruns (DIA Data)
Chris Donovan (Outlier Ventures)
Conor Diviney (Bond 180)
Dan Danay (Key Tango)
David Minarsch (Fetch.ai)
Denis Pitcher (Chief Fintech Advisor to Bermuda)
Mudit Marda (Outlier Ventures)
Tor Bair (Secret Network)
Walid Al Saqqaf (Insureblocks)