Decentralized Finance

While we wait for the blockchain/crypto technology to scale to the point where it can be the foundation of mainstream consumer applications (games, social media, e-commerce, etc), there is a sector where scalability is a little less important and where blockchain/crypto is starting to show some real signs of life.

In the crypto space, it is called Decentralized Finance, or DeFi for short. It includes, of course, all of the ICO activity largely built on top of Ethereum and the ERC-20 token. But it also includes thinks like Maker which is both a stable coin and a collateralized lending sytem. The collateral for the loans is what stablizes the Maker stablecoin. We also are seeing other lending offerings develop in the DeFi world and we are seeing things like hedging, shorting, derivatives, and more, all built on a decentralized platform where there are no intermediaries, no clearinghouses, and the need for trusted third parties is much less, sometimes not at all.

This makes sense for a number of reasons. While the transaction requirements of financial services applications are not trivial, they are also not as demanding as mainstream consumer applications where millions of users are transacting with each other and the system in real-time.

It is also the case that, unlike many of the new architectures that emerged over the years (mainframe, mini, client-server, web), the blockchain/crypto space has always had money at its core and making money, transacting in money, and everyting that goes along with that has been an early use case and for most people, the driving use case for this technology.

All technologies need early use cases. I do not think DeFi will be the only thing that blockchain/crypto is good for. I think we will see blockchains scale in the next few years to allow mainstream consumer applcations to be built.

But until then, DeFi is a good place to hang out. It uses all of the same technologies, architectures, and value systems that we have come to know and love in crypto. You can learn to build applications, use applications, and generally come up to speed on the sector while serving real customers, building a business, and, hopefully, making money.

#blockchain#crypto#hacking finance

Comments (Archived):

  1. William Mougayar

    The level of automation that’s inherently built-into these DeFi capabilities and instruments is frightening to the un-initiated.

    1. JamesHRH

      A year ago, we made a major purchase in South America. I made a point out of executing the transaction with my Visa card, as we were concerned about the integrity of the vendor. Our concerns were valid.The vendor refused to honour the unsolicited guarantee of the quality of the good matching the price and would only replace the item. We asked for a full refund, as the guarantee stated that we would be ‘completely satisfied.’The Large Transactions Dispute mechanism of VISA INTL was invaluable to us.If I had Blockchained that Tx?The thing about Blockchain is that its main features are automation & non-reversibility. It also suffers from lack of certainty / delay.How many transactions value those features?Even in finance, transaction permanency seems a bug.

      1. Lawrence Brass

        Major purchase? Here is my advice, never buy a hill or a bridge in South America. And also, llamas and alpacas will spit you in the face, everytime. It is a product feature not a product defect, señor James. 🙂

      2. F G

        Your point about intermediaries in transactions is spot on. For many transactions, the intermediary isn’t dead weight but in fact, a guarantor of the transaction. The guarantor removes the friction from the transaction as the arbiter.Even in straight financial transactions, data is not always “right”. Imagine a derivative transaction based on an underlying stock price. If the stock hits a certain price, the derivative outcome is paid. During the day, the stock hits the price and the contract is paid. Two hours later, the exchange reverses the transaction as erroneous. You now need to back that transaction out, but can you?

      3. Salt Shaker

        If you did an attribute sort of the credit card industry 20+ years ago, convenience/ease of use would likely rank quite high. Today, with the rise of identity theft, database breaches, implementation of 2FA and other security related issues, convenience is no longer a big selling point. It’s all about security and safeguarding one’s liability. The CC industry’s main selling point is the largest impediment for acceptance of crypto/blockchain. A non-starter until/if that issue can be adequately addressed. The CC industry today is as much about insurance as it is finance.

      4. William Mougayar

        Your case is like comparing apples to oranges. DeFi isn’t about sending a transaction into hell and having no control like you’ve hypothetically alluded to. DeFi is more like codifying a process and allowing it to execute well. If we were to compare to your case, the blockchain would have identified your seller as fraudulent and not allowed you to make that transaction. So, it would be helping you earlier.

        1. Salt Shaker

          Fraudulent is one thing, unethical is another. Will there be buyer protection on blockchain transactions? When a seller offers a money back guarantee and doesn’t deliver on that promise, then what? It’s all caveat emptor. Obv on smaller, everyday transactions buyer protection is less of a concern.

          1. William Mougayar

            You could have escrows too and multiple approval levels coded inside a blockchain enabled transaction.

          2. F G

            The difficulty that I’ve seen in the Brokerage space (a mature industry/set of processes) is that new, never before seen issues/conditions continue to arise. How do you code for events previously never seen? It took the industry weeks to sort out the issues that arose from the Facebook IPO.

          3. William Mougayar

            You still need to apply common sense reality within these contexts.

          4. JamesHRH

            That’s really interesting. I have never heard that answer. Crypto from the cradle, sort of.But……You are talking about reinventing the entire world economy so that a specific type of transaction structure with no history of usage ( to be fair, due to impossibility ) can be used.I am talking about solving the same, occasional occurrence problem with a 3% fee that I never see.Hmmmmm.

        2. JamesHRH

          How so would it have helped me sooner?BC is Transaction Workflow Automation now?

          1. William Mougayar

            It would have identified bona fide participants, so you aren’t doing business with crooks. The assets they would own are certified a priori and the blockchain can automatically initiate the transfer to you. They can’t reneg on their commitments. That’s why the blockchain is also referred to as a trust intermediary.

        3. cavepainting

          How do you think the blockchain could have identified the seller as fraudulent? It is possible onlly if the seller used one of the identity services and his reputation is tracked on the chain?

      5. gbrandonthomas

        This is a function of immaturity, not the tech or emergence itself. No doubt even in the new system intermediaries are needed to support trust issues such as the one you describe. The point is that it becomes your choice whether or not to opt in (and pay) for them, vs being obligated.What if the process you described was handled by smart contract? You click a button to say you are dissatisfied. The merchant can choose to protest, elevating to an arbiter of some form. Your case is pleaded, and the arbiter judges (e.g. Visa in this case). But, instead of Visa charging 3% on all transactions to merchants for this service and taking days or weeks to settle the issue, the cost is far less and it takes a matter of minutes / hours.Rather than think of it as permanency, I’d alter it to transparency – “changes” can always be made to blockchain-based transactions, just in an open format visible at least to the parties of the transaction…

        1. JamesHRH

          For 3%, this job is done for me.Why would I want to waste my time and energy fighting with losers?

      6. Stefano Vettorazzi

        Using Bitcoin you can create a multisig address which is useful for your case. Let’s say you (buyer) don’t trust me (seller) and I don’t trust you, but we both trust Fred (Visa in your case). If everything is okay, you and I sign a transaction that “gives me the bitcoins”. If you complain and I don’t want to give you a refund, you go to Fred. Fred looks to what’s going on, and if you are right, you and he sign a transaction to “refund the bitcoins to you”.I think that the same is possible using Lightning, but I haven’t read about it yet.About refunds in South America (where I’m from): it’s not common practice to give a full refund. If it happens, it’s very-very rare. What you described is exactly how it works here, based in my experience and what I know.

        1. JamesHRH

          You get that VISA already does fhis job for me, right? At no cost to me? And with 50 years of track record?That’s a BTC attack, not a personal one.I was aware that full refunds were rare – although I did not know about the very, very, very part! – but that is what was offerred.

      7. Oink!

        If I had Blockchained that Tx?With a blockchain, a smart contract would be your middle-man. The seller is not honoring the contract, so the contract is not executed and your money stays with you. Nice, simple and probably cheaper (since you wouldn’t have to pay for Visa’s mediation).

  2. Rob Underwood

    Fred, thanks for highlighting the work going on in this space.As some on this board may know about a year ago I got involved with FINOS, the Fintech Open Source Foundation. We recently launched a new Decentralized Ecosystem program, led by IHS Markit together with Amber Baldet of Clovyr (who previously led Quorum at JPMC), to give a forum for the DeFi community focused on open source projects (which is much of the work in DeFi).Check out https://decentralized.finos… for more info.Our panel discussion from last week’s Decentralized meetup at Moody’s may also be of interest.

  3. Lawrence Brass

    It is an interesting field indeed. What would be a good starting point?Just read about fiat-collateral implementations through a single entity and frankly it smells like cheating the model. Perhaps hybrid implementations is the only way to go.

  4. jason wright

    So this is the best use case for the tech in its present ‘state of the art’ form, that volume and speed remain words and not deeds, and that DeFi (money) drives the tech to drive itself.

    1. Girish Mehta

      Narrative Violation.

      1. jason wright

        The best science (and technology) fiction is that which in time becomes science (and technology) fact.As an aside, the warning lights went off in my head with Ethereum when i saw the design its ‘issuers’ had chosen as its logo, a diamond.

  5. Joshua Cyr

    It wouldn’t have been that long ago that a post like this would have featured a significant portion of it’s time and follow on comments discussing the [lack of / uncertainty] in a regulatory environment. It certainly appears that there is satisfactory confidence in the regulatory environment for movement and discussion to be centered around value / opportunity / problem vs risk and barrier.

    1. Tom Hughes

      Well… in institutional finance (a.k.a. capital markets) there is a great deal of self-regulation: parties agree on standards (like the standard swaps contract promulgated by ISDA, or the CUSIP Bureau) and government regulators leave it pretty much alone. So the regulatory barriers are less high in general, within that constrained space, because there’s a reasonable assumption that these are mature institutions capable of looking out for themselves. The regulators look more at the properties of the institutions overall (like leverage and asset quality) rather than the minutiae of how they transact.Given their greater freedom to maneuver, lots of institutions are looking at blockchain solutions not because they will upend the status quo, but just because it may be an easier way to get some things done that are long-standing annoyances: know-your-customer compliance, continuous settlement, things like that.

  6. Jacek Czarnecki

    DeFi is a real thing. You may like my presentation on the associated legal challenges from the Ethereum Community Conference in Paris last week.

  7. kidmercury

    dai is the best stablecoin i’ve seen but like the others it still has a target price that is a nation-state currency (us dollar). dai has a governance mechanism that allows this target price to be changed, and that is the part i find most interesting, as when the US debt crisis reaches manifests itself fully i believe a new target price will be desirable. i doubt, though, that governance via a token is the best way to determine the new target price. i think a committee that possesses the necessary specialized knowledge should be in charge of price stablization. essentially more like how central banks currently operate.facebook’s stablecoin is purportedly about being pegged to a basket of currencies, like the IMF’s Special Drawing Rights. if they do that, how that currency will be managed will be fascinating to watch.IMHO this is a key basis of competition amongst stablecoins. i’m looking forward to the emergence of “actively managed stablecoins” in which a peg is not maintained to an existing currency, but rather to a basket of assets that can include currencies, cryptos, stocks, bonds, etc, with open market operations being used to achieve slow growth over the long term with minimal volatility.

    1. JamesHRH

      That’s a derivative.Those are not mainstream

    2. Matt A. Myers

      The only blockchain currency that should exist IMHO is one that good actors of democratic nations are part of, linking their trusted real-world network to a blockchain; this way a blockchain structure/method can be used also doesn’t depend on its energy/CPU consumption to fend off 50% + 1 attacks, and so on.

  8. Adam Parish

    I’m really interested in Binance in the decentralized finance space. Looking forward to seeing if their DEX takes hold and how the decentralized validators work.

  9. iggyfanlo

    I agree that DeFi is the core of blockchain. It follows the transitive properties of mathematics. Since the protocols are the first ones where trust is native, it follows that the best use cases are the ones where trust is essential, but it also follows that it would be fo digital items, not necessarily for e-commerce.Hence money, financial transactions, virtual goods (e.g. CryptoKitties) are the logical domain of blockchain.

  10. cavepainting

    The lending on defi (currently) is primarily about locking up 2-3x liquid collateral to get a loan. There is surely a niche market for people who do not want to sell certain crypto assets, but still want to borrow against them.But the “world changing” part can happen only if Defi grows beyond this into reputation based lending and physical asset based lending. If all that Defi does is be a toy for borrowing against liquid crypto collateral, it does not solve meaningful problems. Of course, how exactly that comes to pass in a non custodial model without intermediaries is not clear, but that can fundamentally transform finance.

  11. gbrandonthomas

    It’s great to see these ideas starting to (finally) take shape. I am starting to believe that the real ride begins when DeFi and tokenization merge, such that assets available in “any” token form can ultimately be “financialized” simply and easily. Long road ahead…

  12. cyrilhouri

    Check out (disclaimer: I am a member of the team). It’s a decentralized platform to trade derivatives running on the Ethereum blockchain. Yet another illustration of DeFi applications.

  13. sigmaalgebra

    I’m failing to see much of the harm: The US is just awash in education, especially freshman college and beyond, readily available to anyone who wants it.If a student learned a lot of academic material in high school, then they can show that on their SAT scores. High school grades? Getting straight As in high school, no matter HOW good a student is, is at best a crap shoot because the grading is just not THAT reliable. Of course, if the school is small and the student’s father is head of the local school board …! Or each report card, the family has a lawyer ready to sue the school and the teacher for any grade less than A.Maybe home school, don’t go to high school, or even if did then lie and say were home schooled, or just be home schooled for one year and claim to have been home schooled.Much the same for college: Do well on the GREs. I got 800 on the math knowledge GRE, and that pretty well sealed the deal.Graduate school? They don’t much care about grades in grad school. They are really short of students who can do the work at all and generally have more tuition scholarships than qualified students. I paid no tuition in grad school. I got accepted to Cornell, Brown, Princeton, and more. If go to grad school at a state university, then teach some freshman courses and get paid room and board plus enough for some movies, taking a pretty coed to a fancy dinner, and savings. For a Ph.D. the main wall of knowledge, learning to jump is just the qualifying exams. Course grades and even courses count from little to nothing. At least at one time the Web site of the Princeton math department stated that students were expected to prepare for the qualifying exams on their own, that no courses were given for preparation for the exams, and that graduate courses were introductions to research by experts in their fields. So, Princeton was going to ignore course grades: Just pass the qualifying exams from knowledge obtained however.To do well in grad school? To heck with courses, credits, and grades. Instead do some research and then publish it. The main requirement for a Ph.D. is the research. Eventually in academics people understand that no one can hope to carry the library around between their ears and give up trying and expecting the students to try.The high school system of courses, AP courses, etc. is junk, make work, busy work, junk think, nonsense. The high school teachers rarely know material well enough for their courses to be taken very seriously. The AP courses are total junk — just start with the best college materials. The AP course materials I looked at were written by high school teachers who didn’t know the material well. In very strong contrast, the best college freshman and sophomore materials in math, physics, etc. are beautifully done, polished, expert, etc.If a student is really ambitious, then they should be ready for college work by age 12-14. By age 17 they should be ready for grad school. Sooooo, THEN they can apply for a freshman slot at a competitive college?Sure, the big deal is getting an Ivy League Bachelor’s degree. It remains: About the only way to be really sure of getting in as a freshman is to have Daddy give whatever the amount is, $100 million or so, for a new building. Everything else is a crap shoot.That Ivy League Bachelor’s is not better academically than students can do a long way from Harvard, etc.So, the goal of the Ivy League is not academic but joining an exclusive social club? Okay, then, SAY so. And don’t be more concerned about the filtering rules of that social club than those of a local country club.Getting an Ivy League Master’s or Ph.D. is MUCH easier than getting accepted as a freshman.Some of the scandal is about USC — come ON guys, since when was USC the Ivy League?For anyone shooting for the high end of academics, the target is SIMPLE and NOTHING like what is from the newies on their current scandal: The target it to PUBLISH really good RESEARCH. That’s the target. Then for a good career as a prof, the target is to get RESEARCH GRANTS, that is, MONEY, that is, in US dollars, for doing more research. Uh, there are much easier ways to make money, and this fact is fully clear from profs and people who did make money.But the newsies are concentrating on AP courses, athletics, captain of the football team, honors, e.g., voted most intellectual, charity, e.g.. in a soup kitchen, tutoring disadvantaged students, …. So, some high school students miss sleep, strain, etc., but really for what the heck? The Ivy League freshman courses are not very different from freshman courses at any good four year college.E.g., there used to be a good text for freshman physics, Sears, et al. Everyone knew. It was widely used. I used it. It was okay for freshman physics. There is a big shelf of beautifully polished texts for freshman calculus. I used one (actually at least two), the one the same as Harvard used — it was good, and Harvard didn’t have anything better. Yes, Harvard had their challenging freshman math course Math 55. It wasn’t challenging: I was just the usual, good junior, senior math taught all across the country but at Harvard taught to some math students who were about 2 years ahead. The books Harvard used for Math 55 were by Halmos, Rudin, and Spivak, yup, right out of the mainstream available in universities all across the country. I had Rudin as a senior and did Halmos and Spivak on my own, together with MUCH more, e.g., Fleming, Royden, Coddington, … No way had to go to Harvard for Halmos, Rudin, or Spivak. A student with some talent, good guidance, and some work, not very hard, can be ready for Math 55 at age 14.So, we come to an absurdity: A student does well, say, via home schooling, and by age 17-18 is ready for grad school, maybe ready for Ph.D. qualifying exams, maybe has published, so applies to Harvard as a freshman? Good grief.We’re talking about people who want Harvard to be a social club, not academic education.Want to show really good academic progress? Then in one word, hit the target — PUBLISH. Want to show really good academic knowledge? Fine, the two keys are SAT and GRE. The rest is worrying about some social club stuff-o.Made As in high school? So darned what the heck: The high school teachers rarely know enough to get serious about, rarely could have gotten into Harvard themselves, and their grading is not very reliable anyway. A student with good talent, good guidance, and some work can from home schooling by age 14 know more math and science than nearly any high school teacher.Let’s be a little more clear: A student wants to do well in academics. Okay, one way in K-12 is to follow the advice, program, teachings, etc. of the K-12 teachers and get their praises and As. Bluntly, in this, especially in K-8, the main way the student was pleasing the teacher was not via good, objective, academic learning but by being good at seeing social cues that would please the teacher. So, it was seeing just what the teacher wanted and giving it to her (nearly always female). That’s not much of a good, early path to academic learning. Moreover, are pleasing someone with fewer accomplishments than are the real objective anyway.So, need (i) a good way to know what the heck material to learn and how well, i.e., some good guidance, (ii) an effective way to learn that material, and (iii) an objective way to demonstrate that know the material. Well, in (i)-(iii) K-12 is at best an extra load of useless bricks for the student to carry on their back and otherwise a slog through a fetid, dangerous swamp.As is common, to plan, start at the end and work backwards: So, start with (iii), the means of objective demonstration. Well, not commonly in life but in this case for the academics we have reasonably good versions of those, SAT and GRE. So, for planning, for (iii) investigate and get the suggested, etc. topics, hopefully texts, that are recommended. Then for (ii), get the materials and learn how to learn from them. Pro tip: Likely can get all or nearly all the texts in plenty good enough form USED at MUCH lower prices. Can get some quite good texts just for free as PDF files via the Internet, but need real EXPERTS to say what the good texts are or know from what is famous, used at Harvard, etc. But calculus hasn’t changed much in decades! In simple terms in math, study the next lesson, work the exercises, check with the back of the book or the teacher’s guide, etc., rinse and repeat. Then for (i), the guidance, have access to someone who cares and has been there, done that, and done well at it. Have a parent or other family member be encouraging, solicitous about any problems, clearing the way, e.g., have a quiet place to study and keep the other kids from interrupting with too much loud TV or video games, etc. And find some one, two, or three people to meet with about each three months, hopefully a well qualified prof at a good research university, to provide some overview, big picture, check for and then correct for getting way off track, etc., keep aiming at the target of learning the material as needed at least for the SAT and GRE exams.(i)-(iii) should often be quite sufficient — I did a lot more with a lot less.Here is an example of one of the dangers of K-12: My eighth grade arithmetic teacher gave me a D, at the end of the course took me aside and one on one and fervently advised me never to take anymore math and mentioned that all I needed was high school arithmetic.I was fine at math, and at the math in her course, fine, right away, no hesitation, in roots, exponents, ratios, proportions, areas, volumes, compound interest, etc. What I was not good at was handwriting — as is common for boys at that age, my handwriting was AWFUL. With no effort from me, it got fine by college. She was wrong, badly wrong, wildly wrong. Fortunately Dad was MUCH better at education than that teacher was, laughed at her advice, and fully encouraged me to continue on. The high school teachers were better, also, and sent me to a math tournament and a summer NSF program. At the end of high school when my SAT scores came back, some of the K-12 teachers it appeared had to have lots of long gossip sessions in the teachers’ lounge about how a bad student like me could do so well on the SAT. Well, with good SAT scores, the teachers could gossip, be confused, etc. all they wanted, but they could do NOTHING about my good SAT scores. The SATs were objective, and the teachers were not and, instead were heavily just social.So, sure, I ended up with a Ph.D. in pure/applied math from one of the world’s best research universities. Lots of good students didn’t make it; the dropout rate made the Navy Seal training look like fuzzy, bunny play time. The key to my success was independent study.Net, there was no reasonable way for me to do well in math by pleasing those high school teachers. They weren’t very good at math themselves, e.g., were all afraid of calculus. In contrast I taught myself freshman calculus on my own and in college started on sophomore calculus, from the book Harvard used, and did well. I’ve learned calculus, and several steps beyond, taught calculus, applied calculus, and published research in calculus. I know some calculus; those high school teachers didn’t.Big Lesson: A good student, especially one eager to strain to get into Harvard, needs a MUCH better learning environment than available in formal K-12 education. Basically they need home schooling with (i)-(iii) as above.

  14. sigmaalgebra

    “Decentralized, distributed, crypto, blockchain, finance” — sounds like sizzle to me. I’m less sure about the steak. Maybe for my startup I should take lessons in sizzle, not let the significant steak get in the way of selling both the sizzle and the steak.Some people make a lot of money out of sizzle ….

  15. CB

    “I think we will see blockchains scale in the next few years to allow mainstream consumer applcations to be built.” – Thoughts on what that could be? I struggle to come up with use cases that require decentralization and offer a 10x value proposition for consumers.

  16. Pointsandfigures

    Yup, looking to invest in it. Seen some stuff but haven’t pulled the trigger yet. A lot of the stuff is way too early. Some of it seems to be working on problems that don’t really exist. At the same time, sometimes some centralization in finance is worth the opportunity costs. Craig Pirrong at has written about this. https://streetwiseprofessor… We are investors in the first securitized token ATS ( and if you are interested in trading go to their website and sign up.

  17. Ajian

    Dear AVC,I hope this message finds you well.This is Ajian, from, which is a China based Ethereum developer community, dedicating to promoting Ethereum and blockchain technology in China. One of our main projects is translating selected English posts into Chinese and circulating them on our website and newsletters, where we have more than 8000 daily visits.I am wondering if we can ask your authorization to translate this post for non-profit purpose. If permitted, the Chinese version will be posted on and our wechat daily newsletter. We will specify your authorship of the post, and put down a link to your original post.Please let me know if you have other requirements regarding the authorization. Looking forward to your reply.Thank you.Best regards,Ajian

  18. Lawrence Brass

    Don’t worry, at least up to now they are contained experiments safely stored. Like VX gas capsules in Arizona.Pat a cat, be happy. 😉