June 7th 2020
Cryptocurrency advocate and analyst, growth hacker
Crypto’s growing attractiveness stems from digital assets’ unique capacity to eliminate pain-points peculiar to certain industries. Without any doubt, one of the defining successes of crypto in 2019 is the emergence of DeFi technology, especially in the lending and investment market. Together, cryptocurrency and its underlying technology enable lending ecosystems devoid of challenges prevalent in legacy systems. With this setup, lenders can easily invest in the budding crypto lending market, while borrowers can evade the bureaucratic structure of traditional lending facilities and enjoy lesser interest rates.
Fast-forward to 2020, and all market indicators project that the emerging crypto lending sector is not slowing down any time soon. Stakeholders are aware of the nascent status of the crypto lending niche and are constantly introducing new ways to ensure that lenders and borrowers continue to savor the benefits of crypto loan facilities, without foregoing the security of all parties involved. However, as it is with other sectors, there are certain areas that crypto lending platforms ought to improve upon.
Can you please tell us more about your background and why you joined the blockchain and crypto industry?
I hold a masters degree in computer engineering. Throughout my career I have held various engineering roles. Prior to joining Raise I had a leadership role in a fairly larger product development company. I am also the co-founder of Clovr Labs, which is a cyber security and data analysis boutique firm I have running with an old friend of mine.
I first became interested in the blockchain around 2012. I had had to deal with the interbank network and card processing schemes before and I was attracted by the inherent simplicity of Bitcoin’s value proposition. I started working on Blockchain related products from circa 2015. Mostly focused on payment processing, asset treasury and blockchain analytics.
Up to 2019 though, I hadn’t really looked at the possibilities provided by the Ethereum lego ecosystem. I used to think – and still do in a sense – the market wasn’t ready for real adoption yet. When I started speaking with the founders at Raise I knew they were onto something.
As the CPTO of a blockchain-focused platform, what is the most challenging aspect of creating and offering a viable product in the crypto landscape?
I’ll just speak about that subset of the crypto landscape that has do with financial instruments.
I believe it is clear by now we – as an industry – are capable of solving virtually any problem, so I wouldn’t say creating a product is a challenge per se.
If anything, we are at this stage where we’re taking on too many challenges at once, and that sometimes has an effect of making us lose focus. One thing is to say we can solve problem X, another thing is saying it makes economic sense for us to solve it.
Incidentally, this is where organizations like the Ethereum Foundation go a long way as they effectively subsidize the equivalent of basic research in the blockchain space.
Then, if by viability we are talking about economic viability, right now the challenge is in finding that intersection between problems worth solving and products that non crypto people can relate to.
Onboarding “non crypto-native” users is the number one leverage to making much of what we do viable in the long term. And we’ll be able to attract those not only thanks to better UX but – most importantly – by clear, relatable value propositions. The ultimate goal here is driving more usage, more volume.
Taking lending as an example close to us, the value locked in the whole DeFi market is what a major player in the traditional p2p lending space would do in a month. Some of the loan originators we’ve spoken with are looking at borrowing 100M a month and we are several orders of magnitude away from being able to compete if we’re just looking at crypto users.
DeFi is arguably the most compelling selling point of crypto and blockchain technology. How has Raise leveraged on this concept?
The truth is we’re not leveraging the #DeFi tag a lot these days. The thing is the DeFi community has matured a very specific concept of what the “De” in DeFi means: something that is trustless, borderless and open to anyone. All these properties are worth pursuing and as an operator we’re committed to their pursuit. At the same time, we don’t feel it all needs to happen at once.
For example, at Raise we’ve made the decision to adopt a compliance-first approach with regulators. This translates in things like requiring Id verification for our users and shying away from US investors (besides the rest of the OFAC list). Do these measures make our product less DeFi? Absolutely. Could have we grown faster without them? For sure. At the same time I believe our approach puts us in a much stronger position if we’re to build a stable, viable business.
The last topic in this area is to what extent our protocol can be called decentralized – in the sense it can’t be arbitrarily changed by a single actor.
As of today, do have an admin key. That is used to vet protocol access for verified accounts and to make certain parts of our contracts upgradeable. At the same time, investors and loan originators don’t need to trust us for loan execution as that executes algoritmically without us being able to intervene. I believe that strikes a good middle ground: it minimizes complexity while providing features traditional crowdlending products don’t have.
We will look at more decentralized goverance structures in the future but our focus right now is getting the basics right without adding more moving pieces just to claim extra points on the DeFi scale.
It is a known fact that there is competition brewing in the burgeoning DeFi market. How have Raise managed to thrive in this ever-changing terrain?
After much mulling on this, I’ve come to think most, if not all our DeFi counterparts aren’t actually direct competitors for our offering. Fully, or over-collateralized loans for instance are a very different instrument than what we do and is likely to attract a different segment of investors than the one we’re targeting.
I wouldn’t say we’ve found the magic formula here yet but it’s becoming clearer our target user doesn’t necessarily come from the DeFi space. As of today, ~50% of our users say they hadn’t used a DeFi product before and 23% of the wallets connected to our DApp hadn’t been used before.
And on the other hand, I see more conversations coming out on how over-collateralization is good for risk management – and thus has its own utility – but is also very capital inefficient. Requiring loan originators to lock 100% of currency as collateral wouldn’t make sense if you are a business that needs capital to actually do something with it.
Do you mind explaining to our readers the intricacies of your platform’s interest rate? How does Raise determine the ROI of each investment?
When a borrower requests a loan through Raise, they set the minimum and maximum monthly interest rate they’re willing to pay. These two parameters, alongside with requested amounts (minimum and maximum) and loan term (for how long they’re borrowing money) are passed to the loan smart contract when it’s deployed and can’t be changed afterwards. There is an additional parameter though, which determines for how long the pricing phase will last.
The loan contract will run a reverse dutch auction for the length of the pricing phase, using the minimum and maximum interest rates as the starting and finishing points. This means the loan interest rate will be linearly increasing during this pricing phase and investors will commit their funds at an APR that is determined by the point in time their transactions are mined.
The catch 22 of this type of auctions is at the end of it all investors get the rate of the last one having come in. The idea behind this is it allows the market to self-regulate: investors can enter their positions at a price they feel comfortable with without suffering losses for their early commitment. Loan originators on the other hand can still control their worsts case.
I am sure that you are aware of the security issues associated with your sector. A case study is BlockFi, which recently experienced a breach in its security protocol that led to the loss of users’ private data. How does Raise ensure that its clients are unsusceptible to a data breach?
When it comes to data breaches, it’s actually pretty simple. All we store about a user is their email and wallet address. We don’t store PII data gathered via account verification.
Personal data is secured by our two KYC providers: Sum & Substance and Bloom. The latter in particular allows users to retain control of their personal data so you could say there is conceptually no data breach to be had there – it represents a neat security model.
Our own surface area is limited to emails. When it comes to securing it, we have implement a set of good practices like network segregation and avoiding shared keys.
At the same time I believe users should stop using personal, catch-all emails to access all their services and instead get into the habit of using disposable ones. It’s time to do with emails what we have learned to do with passwords.
What are the risks and benefits inherent in unsecured loans?
The main risk is unsecured loans are well, unsecured 🙂 That means they expose investors to full principal losses in case of a default. Nevertheless, there are a couple of things we can do on this front. More on this later.
The main benefit is higher yields for investors. Our APRs ranges from 7 to 20%, which offers a significant premium over other investment instruments out there.
I think it’s important to highlight there is no high reward / low risk investment; the primary tool for investor risk management is a differentiated portfolio. This is well understood for example amongst stock traders. I have a hunch this is maybe less so in the current DeFi space.
Do you believe that enabling an uncollateralized lending marketplace is an effective solution against flash loan attacks?
I think they are very different solutions, I’m not sure there is actually a link between the two. Traditional lending executes over a period of time measured in months. As such, a flash loan attack would need to cause effects that last that much – something that I find unlikely. I’d actually say that if that was to happen then we’d have a much bigger problem to solve.
In addition to smart contract functionalities, what other way does Raise ensure that loans are repaid?
First of all, let me clarify we are currently doing unsecured business loans. This is a key difference from lending to individuals because one of the pains of personal lending is actually malicious actors with no intention of repaying. Businesses offer a wider array of data points in general.
To that extent, Raise current business model is to only lend money to businesses with the following traits – which we establish during our financial and legal due diligence.
Well-understood business models. We only onboard lending institutions that will use funds to finance growth as opposed to debt consolidation, for example on their way to profitability but whose financial statements still don’t make them a good fit for traditional banks.
We are also in the process of partnering up with a risk scoring start-up that has developed models targeting this space. This will allow an easier comparison between different loans in our application but also between our offering and other financial instruments by providing an equivalent of a bond credit rating for each borrower.
Other than evaluating and disclosing risk, we have skin in the game: we are currently funding 10% of each loan ourselves, something that aligns our interests with the ones of investors. We are also about to deploy a new type of loan contract, which will support monthly repayments. This way, investors will start collecting their principal and profits in instalments as opposed to at the end.
Finally, we are right now in the process of designing an insurance fund, whose financing will initially come by contributing a fraction of our transaction fees (which are paid by borrowers).
The extensive use of smart contracts to optimize the operations of financial platforms is fast becoming a technology trend. How can developers ensure that their smart contracts are unsusceptible to hacks?
At the risk of voicing an unpopular opinion, I believe the best measure to mitigate the risk associated to hacks or defects is adopting good development practices.
We have learned how to develop robust – financial and non financial – software for decades and I honestly feel all of this learning ins’t necessarily being applied in the crypto space right now.
Too much weight is given to third party audits, something that has created several negative effects, the most obvious one being there is now a perception that’s the only thing a team needs to do to avoid hacks.
Audits have their space, and our contracts were audited. At the same time, we have invested in testing, we monitor coverage and our changes are continuously deployed to the Kovan testnet.
Testing and rolling out small changes with robust tooling is the way to minimize risk. Tinkering for 6 months and unleash all of that at once isn’t – regardless of external audits.
We could go more in depth on how this is affected by immutability – immutable contracts are by definition deployed once – but this is would probably not fit here. I’ll just say that deciding which contracts to keep immutable and which to make upgradeable is part of how the protocol is designed.
Another talking point in the DeFi market is the perceived intricacies of decentralized financial products compared to their traditional counterparts. How should developers counter this prevailing notion?
I’d say product developers – and I include myself into the lot – need to think more about market fit and not necessarily about solving a problem just because we can.
Our current developer-led space is cool because it really sparks creativity and innovation. Nevertheless, I think it appeals a niche and I resist the idea of a DeFi landscape that only caters some. It’s as if the traditional financial system had been designed with the arbritrage stock trader in mind. What about everyone else? It’s not about the diffusion of adoption curve – which is natural. It’s about having relatable products.
In the aftermath of the crypto’s crash in mid-March, we witnessed the risk inherent in DeFi-based systems and how the MakerDao community responded to such threats. Do you believe that it is a tad difficult to create a lending platform free from all elements of centralization?
All form of underwriting lending – personal loans, business loans, invoice factoring – are subject to centralization to the extent money is lent to an entity that is then supposed to pay it back.
When I look at MakerDAO’s distributed governance experience I see two things
There is no reason why an operator such as Raise should retain the sole ability to tweak protocol parameters. This is something that can, should and most likely will be decentralized in the future. It’s really hard to constitute a voting system where no one actor has 80%+ of the weight.
One element of decentralization Raise has, which sets it apart from traditional crowdlending platforms is investors don’t have to trust anyone for the flow of funds or with the rules governing loan execution. I believe this approach of focusing on decentralization where it brings a market differentiator first is the way to go. I don’t believe in maximally decentralized products as the only worth shooting for.
What are your thoughts on the surge in mainstream interest in crypto? Do you think that this trend will rub off on the crypto lending marketplace, and is the space technically capable of scaling existing infrastructures to meet an explosion in
The biggest benefit for us is more people being exposed to the general concept, which lowers the barrier to adoption. At the same time I don’t think we’re a stage where people become interested in crypto because of a use case other than currency trading yet.
I’m not really concerned about scalability of the existing infrastructure for Raise as operating model doesn’t need high transactionality. Sure, a bit more snappiness would be nice 🙂
Since Raise is based in Barcelona and offers its service to several other territories, can you tell us more about the complexities involved in delivering multi-region financial service infrastructures?
For the sake of clarity, we run our product development efforts from Barcelona, Spain. At the same time, our marketplace is operated from Estonia.
Before going live, we invested considerable effort in understanding how we could best operate our product. Financial services regulations vary greatly from jurisdiction to jurisdiction and we’ve always wanted to adopt a compliance-first approach.
Two factors of simplification in any regulatory analysis are custody of funds and whether trading of assets is allowed. In our case we’ve circumvented both by implementing our protocol so it’s strictly non custodial and by not offering a secondary debt market. The third lesson learned is operating safely in the US is tricky and so far we’ve made the decision of leaving that market out. We believe Europe, Asia and LatAm provide enough opportunities for us to tap on.
Currently, developers are working round the clock to improve various aspects of blockchain and crypto technology. Which development are you most excited about?
I’m excited about many things! I’ll mention two applications, leaving out all the cool things that are taking place in the base layer development space.
The first is decentralized identities – products that give people control over the use and sharing of their personal data. This is important not just from a data privacy standpoint but because it provides a solid foundation to build personal credit information that is independently verifiable and hence generally available – regardless of demographics.
Somehow a generalization of the above is Oracles. It is becoming clear that access to real-world data is one of the stepping stones for blockchain applications to be truly useful. And, as we have seen, this is a high-stake topic. We are still in the early days in terms of the variety of information available – take company shares for example – but I am very excited to see this field picking up momentum.