The Future of Money


I took a two-week hiatus to focus on some career moves and mental health during the pandemic. Things are trending back to normal, and I should hopefully have some great news to share in the week to come. In other news, this newsletter grew ~50% thanks to Peter from Metacartel Ventures giving a strong endorsement on Twitter.

Now would be a good time to share some of my thinking on the evolution of money, how DeFi fits within it and what is to be expected from the space over the next decade. Consider this an alpha version of an evolving thesis for DeFi.

Note : If you are new to the space, I suggest beginning with reading this piece I wrote as an introduction to DeFi last year.

A Brief History of Money And Networks

Money is the instrument with which we drive influence upon social networks. It curates labour, goods, services and reduces friction in trade by acting as a measurable form of value. In studying the history of money, it becomes clear that networks have accelerated its use and innovation. Before American Independence, settlers in Colonial America conducted trade in Spanish dollars due to deficits in acquiring British Pounds in the region. The desire to set up East India Corporation led to the issuance of stock for venturing in pursuit of profit. Venice and Florence benefitted from the active trade going on in the region – leading to the birth of double accounting systems and foreign exchange. The large banking networks set by the Rothschild family also relied on an intricate network of reputation and networks to function between the 15th to 18th century.  With the benefit of hindsight, it is safe to say that the complexity of the functions money serves evolves as the strength of the network within which it is used. Why? Because as a network grows, the complexity of interactions within it increases. As a network evolves the amount of trust in the monetary base of its economy needs to increase. Nial Ferguson’s Square and the tower explores this phenomenon in good depth.

The birth and scale of the internet came at a time global economies were coming to grips with seismic shifts. The United States, fresh out of the cold war, began seeing its technical prowess as the means through which it influenced much of the world. China was evolving towards being a capitalistic, industrial nation as a generation began forgetting starvations after the benefits of the great leap became evident, and India’s markets opened after a financial crisis in 1991. Capitalism and money became the means through which markets expanded and democracies secured themselves. This context is essential when we think about how money changed with the arrival of the internet. “Globalisation” was the hot new thing, and we needed platform layers that enabled this new “village” as Thomas Friedman would put it to be able to interact with it. While traditional banks and regulators saw threats from the “dot com” frenzy, they did not see it eating them up in the early 2000s.  Dot Con is a good book on the exuberance of the time

Where the early 2000s was about money becoming digital through the adoption of ATMs and digital banking, the 2010s have been about programmable money. Interfaces that can speak to one another through APIs have been the foundation of what is fintech today. Robinhood, Stripe, Transferwise, Google Pay, Apple Pay – the behemoths of the fintech world rely on the ability of traditional banking interfaces to communicate with one another. So we have had three broad things happening in the past 30 years, which may have done more to influence the way we perceive, use and think about money than in the past 5000 years. Yes, thats a reference to the title of the book named Debt.

To re-cap, “digital money” came along-side
1. The opening of global markets towards the end of the cold war (the 1990s)
2. The belief that we do have a global village thanks to the power of the internet (the 2000s)
3. The arrival of money platforms that can communicate with one another (the 2010s)

The challenge here is that while the internet has given us a network layer to transfuse culture, thought and services across the world, the means through which we pay and receive money is built on legacy systems. They continue to be programmed by a handful of entities that started with a vision for what money should be but became duopolies or monopolies in different markets. When we consider Visa, Mastercard, Swift and Western Union, we are looking at entities that are comfortable with their share of the market as regulations ensure it is virtually impossible to compete against them. Entry barriers for creating new financial platforms and instruments have been kept high through licensing regimes and capital requirements. Until Bitcoin came along.

Value Transfers in Web 2.0

When thinking about the value on the internet, attention has been the most common system of payment. Tim Wu’s Attention Merchants gives context on this. Platforms like Youtube, Facebook and Reddit have scaled to size by relying on the cheapest commodity on the internet today – a person’s mind-space. Part of the reason for this is effective systems of monetary transfers don’t exist for internet-scale yet. Which effectively means business models cannot iterate at the pace at which the internet evolves. While Paypal did an incredible work in moving money to its digital medium and the likes of Monzo, N26, Revolut and Transferwise have made age-old banking interfaces move into the 21st century. The fee models they use leave much to be desired.

Internet-scale money transfer systems need to be handling the following to be truly effective :

1. Cost-effective – for micro-transactions
2. Global and inclusive – since our economies are now intertwined
3. Programmable – to account for the unique use-cases changes in business models bring.
4. Secure – Because we have banks getting hacked frequently enough

Stripe powers much of what powers digital commerce today. Their launch of API layers to enable digital commerce empowered developers to be able to begin collecting payments with a few lines of code. Faster onboarding, dashboards to verify sources of payments and ability to collect payments from anywhere in the world made it possible to build global-scale businesses. As the number of people offering the primary form of value (attention) on the internet increased, the infrastructure needed to derive money from them evolved.

Bitcoin’s emergence on the internet over the past decade relied on this macro-trend of our economic interactions going increasingly digital. Music, television, newspapers, movies – even dating, has entirely gone digital. The last front for social transformation today is money. Given that governments maintain power, influence and social integrity through their control of money, changes with our stores of value have taken longer. Over the past decade, Bitcoin showed us that a monetary system could be designed such that it is governed by the people, with levels of transparency higher than that of its traditional peers while being drastically more inclusive. While there have been monetary experiments that tackle each of the challenges mentioned in the previous statement, what may have empowered Bitcoin the most as v1 of an internet native monetary system is the fact that nobody has been able to take it down so far. While Bitcoin serves the function of a store of value quite well, the ability to represent customised stores of value on its chain has been restrictive. This is the gap the likes of Ethereum has found as an opportunity over the past five years. There are, however, several other trends that are contributing to the rise of internet native, digital-first assets and what we are now calling “decentralised finance”. I see them as a precursor to what Web3.0 will be.

The “new internet” will be one of ownership

As digital native economies came of age, the age-old question of “who owns what” came of prominence. Regardless of how good a creative is, without the distribution of Facebook, Youtube or Twitter – it becomes hard for the individual to reach their audience. This has turned on against the user in multiple instances. Amazon has launched its variations of goods that sell the most. Google competitively ranks businesses based on who’s paying them. Even our democracies have gone on sale as firms prioritised profits over sane societies. The quest to own one’s data and identity is the transitionary phase of Web2.0. Facebook allowing individuals to export their data or Google drive being able to import external files is the slow transition towards modularity of data on the web. Over time, it may not be surprising to see individuals being able to export their data and use it in a different client.

However, money and data are not the sole underpinnings of a digital native economy. Where Web 2.0 platforms like Facebook relied on user eye-balls, Web 3.0 ventures rely on giving individuals the right to owning the produce of their activity. In other words – we see a structural reform in how incentives on the web work. Reddit’s foray into using community-based tokens is an early sign of this. Through rewarding individuals in tokens for their activity on the website, Reddit is evolving beyond its reliance advertisement revenue. They are creating an entire economy that empowers creatives around the product. Users with a high amount of influence and reputation could trade reward tokens from their subreddits or be incentivised assets that can be converted to actual dollars. By using a blockchain, Reddit can enable micro-transactions from and towards individuals in any part of the world in unique assets that may or may not be traded against the dollar. In contrast, Medium – inspite of its $5 pay-wall and intention of rewarding content creators in proportion to the time individuals spend on the site has seen user activity decline. Why? Because unlike a blockchain native product like that of Reddit’s – the underpinning of Medium’s is on Stripe. One that by default, makes it considerably harder for creatives from much of the world to sign up given the amount of AML/KYC that is needed to get one’s foot in. More importantly, it is an ineffective tool for micro-transactions. Not every read converts to a $5 tip. Sometimes, people may want to pay just $0.10, and there needs to be infrastructure to enable that 20 years since money went digital through the likes of Paypal. 

I focus specifically on Reddit in the above instance as textual content is the most common form on the web today. However – across mediums, the quest to own the rewards that come from user-generated content will be the key driver for businesses adopting tokenised forms of value on their platforms. Platforms that enable individuals to receive rewards in shorter, more frequent spurts in forms of value that can be redeemed for money will over-rule traditional systems of incentives such as likes and re-tweets. We see an early version of this already. 

Here’s how the past decade has changed for prominent segments

  • Music – Artists skip the record labels and find an audience via Spotify and Soundcloud. Rewards are denominated on the basis of number of plays.

  • Gaming – One time payments for ownership has transitioned to free access and multiple micro-transactions. 

  • Journalism – Prominent writers and analysts break away from traditional publications and launch their own paid substacks.

  • Video – Subscription-based video access is slowly becoming the norm. Masterclass, Onlyfans, Skillshare being some of the instances.

In each of these instances, the creative is still at the mercy of the platform’s algorithms, but the web is moving away from low-value rewards that provide dopamine hits to ones that can sustain a person’s livelihood. A16z captures this trend perfectly in what they have been calling the “passion economy”. What has enabled this is the fact that the internet has reached a global scale. Half of humanity came online in the past decade, and that number will only increase. What is lagging, however, is the monetary rails to enable this new transition to a web built on ownership and new incentive mechanisms. 

DeFi As Infrastructure For Internet Scale Money

In thinking about DeFi today, individuals see it from the lens of an innovation in cryptocurrency. We are still struggling to see where people are going to find value. Much of the challenge with that lies with the industry’s focus on short-term rewards like staking and yield. Things that generate noise but make no meaningful impact outside niche communities. The truth of the matter is we are inching closer to a point where the end-user no longer realises they are engaging with a blockchain and exponential growth is already kicking in. One instance where this is taking place is with stablecoins. While much of the use for tokens like USDT and USDC is related to trading, there are users at the fringes using wallets like Argent to send money around. If you don’t believe the exponential growth claim for stablecoins – here’s what it looked like as of Mid June.

Much of what is holding back stable-coins from massive adoption is the application layer around it. Where Stripe makes it easy to take subscriptions through credit cards, a variant of the same for stable-coin based invoicing system has not scaled in the market yet. This, in turn, relies on on-ramps for the ecosystem. While it is easy to argue that on-ramps won’t scale without regulatory clarity from the government, there is ample evidence to believe that rhetoric may change too.

1. Traditional banking platforms have begun enabling users to have access to digital assets

2. What is “fintech” today is also onboarding users to digital assets. (e.g., Paypal shortly).
Interestingly enough, the industry already has precedent for native applications over-taking its traditional peers. Here is Coinbase’s growth chart for reference.
Source : CB Insights

What we are missing out between the noise around yield and staking is the fact that we are creating an entirely new financial stack. One that is more inclusive, efficient & programmable. When we speak about “democratising” finance, we ignore the fact that for decades the ability to build on it has been restricted to the few that could acquire the licenses needed to build on it. More importantly – functions such as interest rates and monetary policies have historically been handled behind closed doors with the public only able to see the final output of what is decided on. Decentralised finance is changing all of that – one application at a time. Part of the reason much of the functions a bank serves has been costlier than its peers is that profitable avenues of the bank have subsidised the non-profitable ones. The gradual unbundling of banks have been occurring over the past decade. DeFi is a natural extension of that. 

Image Source : Andrew Wong’s writing on Open Finance. Strongly recommend reading it.

DeFi allows developers and creatives to take any part of what is the traditional financial stack and build for it. Since stand-alone applications will need to find their unit economics right early on (for funding and making paycheques), they may be able to function better than traditional banks. One way we see this occurring is through liquidity mining today. Compound unlocking over $500 million in total value locked through their liquidity mining provision is an incentive mechanism at play. (More on Compound here). By rewarding early liquidity providers with an asset that can determine governance on the project, the venture is doing two things at once

1. Ensuring people have the right incentives to provide liquidity

2. Aligning those with the right incentives to the long term growth of the network so long as the underlying asset is held onto. 

The volume and speed of money moved through DeFi ventures will be in multitudes of what it is in the traditional financial stack over time. One instance where this has already played out is with stablecoins. Global C2C remittance volume annually is at around $500 billion. Stablecoins are already set to cross that threshold in 2020. The moment the middlemen are replaced by code, there is ample profit to be moved around to other stakeholders or cost efficiency that can reduce the burden on the end-user. 

There are other trends on the internet that will further accelerate the pace at which DeFi becomes internet-scale money. Since each of these are long-forms in their respect, I will give a brief introduction here and dig deeper in future issues.

The Passion Economy

Web 2.0 relied heavily on the influencer economy to come of scale. Facebook, Instagram, Twitter are platforms that enable the dispersion of altered realities that can then be commercialised through sales. However, income from eyeballs has been declining on these outlets for quite some time. The inability to directly monetise is what has lead to the birth of Patreon and Substack. Even then, the ability to customise and receive payments at scale is hindered by the amount of paperwork and documentation needed if you are not in specific parts of the world. Much of the passion economy today is fuelled by Stripe’s payment APIs. If you are not in the USA – opening a banking account to receive payments through Stripe is a painful three to the four-week process, starting with registering a company in the USA to having your account approved. DeFi can cut down that process to a matter of hours.

Projects like Unlock Protocol are already laying the foundational layers for it to happen through empowering creators to charge in digital currency. Similarly, the likes of Roll are making it possible for creatives to release their own NFTs. Historically – releasing goods associated to your name was only for athletes or great musicians. Jay Z has Rocawear, Jordan had Air Jordans, and Rihanna has Fenty. Physical products have added to their cash-flow. Soon enough, the average creative will be able to create non-tangible, digital-first assets and monetise them. This transition towards the digital has already begun with Travis Scott recently having a concert on Fortnite. These digital assets will not have platform lock-ins, allowing individuals to monetise on multiple platforms with a single asset. Sales of things like tickets to a live event on Instagram or rewards for engaging with their content is how Web3.0 will sneak into the existing web. 

One Person Corporations

In the next decade, the theory of the firm will likely evolve to meet the scale of the internet. Where historically, individuals rallied around a large corporation for the trust and efficiency that came with scale, in the next decade – the best in each task will rally around their personal brands due to higher profitability. The best operators will work with others that are just as good at their work in contracts that may last weeks or months to solve for a specific problem instead of setting up a firm and hiring full-time employees. Part of the fuel for this will be the on-going lay-offs in the pandemic. Many young individuals are receiving a harsh lesson that you could be the best at a place in terms of productive output and still be at the mercy of random events when lay-offs do occur. Paul Jarvis’ book – Company of One, explores this concept in great detail. 

The challenge here is that existing laws in much of the world have not evolved to meet this transition. Things like “freelancer” and “gig economy” are more associated with job insecurity even while half of the US economy is hired under those terms. In developing economies like India, up to 80% of the work-force is in “alternative work agreements”. The move to digital-first money and more importantly, digital-first organisations will empower economies to make this transition occur faster. The business model Uber relies on was enabled with rising usage of credit cards. With a DeFi first approach, we may likely see this occurring at an international, more global scale. Specific use-cases I see happening are around tutoring, consulting and likely therapy. Individuals from around the world will be able to pay by the minute for access to the world’s best. 

These two will be the most powerful means through which DeFi will enter the mass-markets. Influencers will make it possible for the average individual to know about and acquire digital assets – thereby being the route through which individuals are onboarded to the space. One-person corporations will imitate existing B2B interactions in traditional financial rails. They are thus bringing in liquidity and scale. As the number of one-person corporations using DeFi to handle payments scale, SMEs will follow suit and by extension, we will have audit firms offering customised offerings for this new form of money. Jurisdictions such as Singapore have much to gain through creating regulatory frameworks that allow individuals from different parts of the world to register in the region, receive payments in stablecoins and handle their business with access to Singapore’s fintech ecosystem without ever moving to Singapore. 

Crossing The Chasm Requires Building Foundations

If we look at what is capturing attention within the industry today, much of it has to do with foundational layers that are making it possible to create experiences that traditional financial stacks cannot. When we see the likes of UMA Protocol and Synthetix being in public discussions on Twitter, it is because they do what traditional financial platforms offer very niche, highly rich clients. Projects like Uniswap are breaking down the barriers that traditionally existed for a token to list and find its initial liquidity. In other words, they are ‘disrupting” models already. It is now a matter of time for them to see the scale. 

As with all technology trends, there will be a chasm within DeFi too. Much of what captures public attention in the initial days will be highly customisable, technical tools the average individual may not care about. That is why news about DeFi feels so esoteric today. As the number of hacks on the foundational layer reduces, and trust in them increases, we will see applications that interface directly with the end-user. The slow transition from traditional centralised exchanges (e.g.: Bitmex) to a decentralised alternative (eg: DyDx) in terms of user-count will be the first signal of this occurring. Once liquidity on the application layer increases, we will see massive community-based DeFi applications emerge. This is already occurring through traditional finance giants collaborating through the likes of Chicago Defi Alliance. These will be digital native cooperatives that function without regional jurisdictions interfering with their operations. Once each of these elements is in place, a vision for a full-stack DeFi based banking product could occur. APIs and composability within DeFi will empower users to hold, remit, trade, lend and purchase exotic instruments from their wallets. In the words of Ryan Adams – this is when we will go truly Bankless. 

Beyond “De”-Fi

Every once in a while, I see individuals on Twitter (and myself) confused about how certain aspects of a decentralised product will be optimised to reduce regulatory interferences. A common example of this is when the front-end of some decentralised exchanges are blocked for access from the United States. The ideological underpinning is that if we are creating truly decentralised interfaces then blocking certain regions is against the ethos of what this stack is built on. There will likely be much to debate about platforms practising censorship once layer two solutions are used in a hack in the near future.

My understanding is that when it comes to technology, people do not make their choices on the basis of ideology but on convenience. The television has been proven to be a terrible way to rest one’s mind and yet most individuals spend hours on it every day because being on the couch is convenient. We know Facebook and Instagram makes us depressed, anxious and frustrated at the state of the world but we would rather use it than go out and hang out with people in real life. Finance has its own share of similar instances. Buyers choosing to finance a purchase with a credit card at higher interest rates for faster approval is one instance. My point is – as more users enter the arena, there will be a spectrum of decentralisation focusing on different needs.

For the user in Iran trying to escape sanctions that are crippling their economy, Bitcoin may still be the epitome of money that is independent of state actors. For someone in India that is looking for quick remittances, a balance between handing over personal identity-related data (for AML/KYC) and speed may better suit their needs. Ultimately, unless there is a sweeping political movement to make money independent of the government, DeFi will be one of the many stacks that empower finance in different forms. Which means it will have to compete against peers on convenience and efficiency to capture a meaningful share of the existing financial markets.

Ironically, blockchains like Ethereum are not the only ones looking at empowering the new financial stack. This transition to an open, immutable and verifiable network to handle digital currency is going on in other forms too. On the enterprise side, we have “money networks” bridging economies through digital currency. Facebook’s Libra and initiatives like JP Morgan’s Interbank information network are some instances of this. Governments seem to be keen on buying into this shift to the digital through digital currency initiatives. The rise of central bank digital currencies is only the beginning of the journey to a cashless society. In this battle between enterprise, decentralised and government variations of currency, distribution will play a key role. Zynga benefitted hugely from the distribution Facebook offered. Before that, Paypal became prominent thanks to E-bays network of digital auctioneers. As nothing can compete against governments when it comes to adopting a new form of currency, it will be essential for builders in DeFi to take a step back from the rhetoric and look at what the financial world truly needs. There is much to be built doing things that don’t scale.

I will be resuming the usual publishing routine from the coming Monday. Wear your masks and stay safe!


Written with help from Alex from Nansen, Ashwath Balakrishnan, Ivan from LTO Network, Dary Lau from CoinGecko, Luciano From DeFi_LatAM, Siddharth Jain and copious amounts of coffee mixed with just the right amount of lockdown frustration.

Consider sharing this if you enjoyed reading it


And subscribe so I can notify you when the next piece is out

Subscribe now

Stablecoins And The Fat Protocol Thesis

In my previous post, we explored how stablecoins are exploding in multiple adoption metrics. In this piece, we will explore whether stablecoin adoption is fuelling a rise in the price of Ethereum as an asset. A commonly held belief in the industry is investing in layer ones will likely create more value than investing in applications. This is because as adoption in the application layer increases, transactions on the main-chain, which in turn increases the demand for the asset powering it. Joel Monegro captured how speculation, boom-bust cycles and early adoption leads to loops that create a rise in the price. I strongly suggest reading his piece from 2016 for more on this. For now – we will look at a critical statement from the piece. 

What’s significant about this dynamic is the effect it has on how value is distributed along the stack: the market cap of the protocol always grows faster than the combined value of the applications built on top, since the success of the application layer drives further speculation at the protocol layer – Joel Monegro

With that in mind, let’s take a look at what has been going on with stable-coin market caps.

A likely Flippening

Tether’s massive issuances over the past few months have meant that the gap between the total market capitalisation of major stablecoins (combined) and that of Ethereum’s has been slowly closing. To see how this has panned out – I considered exploring what percentage of Ethereum’s market cap have been held by stable coins. The idea was quite simple. If the fat protocol thesis does stand, Ethereum’s market-capitalisation should be increasing alongside those of the stablecoins on it. Why? Because coin issuance is plugged to consumer demand, which in turn means more transactions and thus more demand for gas. The interesting bit is – this much has already transpired. Tether is routinely one of the highest gas-consuming applications on Ethereum today. However, that has not caused a rapid increase in price even as DeFi establishes itself as a new market segment. I believe there are two reasons for this

1. Unlike “utility” tokens that provide compute or storage-related services, stable coin market-capitalisation represents a dollar figure held in a bank account. A rich enough entity (like an exchange) can choose to collateralise with their idle assets (e.g.: Bitcoin) and release stablecoins without (i) network effects or (ii) token economies a conventional asset has. Both of which are critical components for spurring activity in layer one.

2. Stablecoin issuance can increase rapidly (eg: the past three months) without the base layer reflecting a change in price due to the activity. My belief is that there is likely a lag in the layer one token price catching up the newfound demand.

Looking at market-capitalisation on its own does not give a holistic picture. Therefore, I considered exploring the volume & token-velocity as a measure of transaction frequency.

The Moneyness of ETH

The next bit to consider exploring was how on-chain volume has changed over time. I understand this is a bit of an unfair chart as I am not considering the dollar value of the ERC-20 assets (e.g: Golem, Augur, SONM, Synthetix etc.) that is moved on-chain but simply Ethereum against stablecoins. What I wanted to know was how the gap between the volume of on-chain transfers of Ethereum and stable coins has evolved. Part of the reason to explore this is to consider the moneyness of Ethereum. If public perception of Ethereum is that it is indeed money, then it is fair to assume that there may be a higher volume of Ethereum moving on-chain than that of stable-coins. However, this has not been the case. Individuals seem to be more comfortable holding and remitting large chunks of stablecoins instead of Ethereum. If we notice the chart, this transition happened around July last year as (i) transition of USDT to Ethereum’s standard occurred (ii), and the market rallied. The gap between dollar value moved on Ethereum and those on stable-coins every day has steadily increased. The wider the difference, the lower chance that individuals choose to switch to Ethereum.

Does this mean Ethereum is not money? The argument stands that ETH can be used to receive DAI and therefore is an abstraction of money. More importantly, it is used as the enabler for the transaction and settlement of stablecoins and is consequently still useful. As retail users understand the broader ecosystem in DeFi – they likely switch to the base layer asset. For now, traction is sticky on the application side. We have some reason to believe the trend may remain that way for now.

Playing Catchup

To understand how stablecoins are contributing to Ethereum’s ecosystem, I studied the number of active addresses over the year. To make a fair comparison, we need to observe 

  1. The growth of Ethereum’s active addresses against that of stablecoins over the year

  2. The percentage of active addresses on Ethereum that comes from stablecoins. 

If you notice the chart, you will observe that stablecoins have gone from sub ~10% of total active Ethereum addresses to close to 40% at this point. As the dominance stable-coin addresses have in the network increases, it becomes clear that much of the growth in Ethereum today is fuelled by stablecoins. I assume that if other “utility” networks like those focusing on storage, network and the like is a decade away from existing don’t come to market, DeFi applications will be the source of much of Ethereum’s usage. As the dominant layer one chain today, Ethereum will also likely influence where many of the newer players like Polkadot will also focus. Finally, even in terms of actual usage, stablecoins like Multi-collateral DAI is being “used” more than Ethereum today. To check this, I compared the velocities of Ethereum and DAI over the past six months. From Santiment. They define velocity as a measure of the number of times a token is used in a day.

One way to argue against this is that much of ETH’s supply likely lies idle as individuals hodl it in secure wallets and avoid moving it around often. DAI in comparison is used by DeFi power users who move it between vaults, dex’s and other applications on a routine basis. That said – it does show that stablecoins like DAI are actively transacted with quite more than the base layer itself.

Food for Thought

As applications come to market, it is becoming clear that adoption alone is not driving the price of Ethereum any longer. There is likely a lag in price due to

(i) the sector being in the late stages of a bust and
(ii) value transactions occurring on stablecoins without Ethereum being locked up.

Even in DeFi platforms like compound today, lending demand for USDT is often higher than Ethereum. None of this is to suggest the Fat Protocol Thesis is wrong. However, it may need certain accommodations for the fact that application layers that are capital intensive like DeFi may very likely end up requiring it to be updated. My bet is that stable-coin market-cap will likely be higher than that of Ethereum’s over the next year as long as Tether doesn’t blow up. More importantly, if DeFi continues to deliver on traction, a basket of DeFi related projects (eg: Kyber, 0x, Snx) will likely have a higher combined market-cap than that of Ethereum over time. Maybe, we really do obsess over the base layer a bit too much. Maybe, the next generation of token enthusiasts will argue over applications instead of the settlement infrastructure.

In my next few pieces, we will explore a thesis for DeFi, investment trends in South East Asia and how the token economics of a few prominent projects are changing. If there is something you’d like to see me cover, do drop me a message

Have a pleasant weekend.
Joel John

1. Data sources – Nansen and Santiment
2. If you would like to discuss / debate application layer valuations, just hit reply to this e-mail.

State of Stablecoins – Q2 2020

Note: This is the first issue of the newsletter that is being sent out via e-mail. I have been using Cipher & as landing pages to build an audience. Those that signed up on have also been ported to this list as I will be retiring that domain. I will be launching a paid variant a few months down the line. Those that had signed up so far will have free access to all of it.

Now that I have reached a meaningful figure, I will be dropping in with numbers & start-up updates every alternative day. Drop me a line in case you need me to unsubscribe you from the list.

I dug into stablecoins for the first time in December 2019. If you had told me then that the market-supply of tokens in the space would cross $10 billion and user-count could be close to 100,000, I might have written you off as an optimist. And yet, here we are in June 2020 looking precisely at that at the late stages of a pandemic. As of writing this, combined stablecoin market-cap of the seven most significant projects is at $10.5 billion (across chains).

In writing this piece, there are a few considerations I made 

  1. You will often see me offer two separate charts for Tether and other coins. This is because Tether is in a league of its own and using logarithmic charts may not do justice on shedding light on leading contenders in the non-tether group of coins. In making this piece, except the market-cap, all data is derived from the on-chain activity on Ethereum. I have discounted Omni, Tron and Algorand for the time being due to their nascent ecosystem and aim to dive into them at a later point in time. 

  2. There is a sad irony at play in the industry today. Of the leading seven stablecoins in terms of market cap, only Multi-collateral DAI has some relation to being “decentralised”. I will be following up with a piece on the state of MakerDAO to shed more light on the project. The rest are tokens issued with reserves in the bank. I hope that reading this piece will make people more aware of how big the market size is and mindful of the need to consider non-banking alternatives to build this space.  

I have sliced data into market-caps, transaction count, on-chain volume, active wallets, tokens held in exchanges, token velocity, nature of transactional frequency and volume to get the numbers behind each of these chains. With that out of the way, lets dig in.

85% of all stablecoins are in USDT

The supply of Tether in the market has gone from $4 billion at the beginning of the year to over $9.2 billion as of writing this. The changes in supply came from issuance of $1.6 billion in addition to the supply and another $1.4 billion in May. USDC imitates this in March and April ($138 million and $148 million) but has a decline in supply for May. There is a distributed power law at play once you exclude Tether. The likes of USDC, Paxos and BUSD are beginning to have an equal split of market share between them which is healthy as customers have a mix to choose from. However, much of them are representations of dollars held at a bank and may come directly under the FSB’s ruling from April 2020. This is part of the reason why collateral backed and algorithmic stablecoins will matter quite a lot going into the future.

Key metrics set new highs

Count of monthly transactions soared from 2.65 million to 6.8 million

The total number of transaction on stable-coins have roughly tripled over the past five months. The bulk of that growth was captured by USDT in frequency. It could be because USDT already had an impressive ~2 million transactions being handled on its own in January. In terms of percentage growth, Paxos takes the lead with an almost ~600% jump in the number of transactions. If current trends hold, 2020 will see a doubling of on-chain volume enabled by stablecoins

Average daily on-chain volume has doubled
The amount of stablecoins moved on-chain in January was at $26 billion. For May, that figure stood at $49 billion. The bulk of this growth in dollar terms again comes from USDT. In other words – the amount of Tether moved on-chain has roughly doubled while its closest peer (USDC) has only seen a 50% uptick over the past six months. Strong contenders like DAI have seen a decline since the oracle issue of March and have been slowly recovering. For now, Tether dominates.

And retail activity may be driving numbers higher.
One of the ways to benchmark stablecoins is by looking at the size of the average transaction on each network. To do this – I have added the total volume moved on-chain, then divided by the number of transactions. In doing so, it became quite clear that while volume has doubled, the frequency of transactions has also increased incrementally. This meant the size of the average transaction has declined over time. In Tether’s case, this meant going from $24,000+ to south of $11,000 over the year. USDC has seen a similar decline making me think this trend is not specific to USDT alone. Higher frequency of retail transactions is a key contributor to this trend.

Thinking of average sizes without looking at how active accounts have changed would make no sense. To get a gauge on this, we measure the number of active recipients over a day on each chain. For a sense of scale, this figure peaked on 7th of June at north of 211,000 users across the prominent stable-coin projects. On the 13th of March – that figure was around 89,000. If we take the weekly figures, every week in the post-pandemic world has set a new all-time high for active recipients of stablecoins. It was inching close to 700,000 as of writing this. A good measure of how much interest in trading has emerged since the pandemic lock-down has kicked in.

DAI emerging as a leading contender

Naturally, the next question is whether the growth of stable-coins is concentrated with Tether today. The power-laws at play here is quite intriguing. On one end, you have Tether being the clear dominant leader with over 70% of both transaction count and volume. However, DAI has shown the more transaction count than some of its centralised peers and has been a leading contender in terms of volume too. This may be because of its usage in DeFi. Exchange-based stablecoins, being primarily used for OTC transactions have considerably higher supply (in market-cap) and larger average transaction sizes. However, since those using DAI typically use it to trade with leveraged positions or use in a DeFi protocol – the frequency of transactions is higher. In other words, as a currency, DAI is likely way more used than its peers on a routine basis.

It becomes evident once we account for the token velocity of the asset (in this case – defined as the total on-chain volume (6 months) / existing market-cap of the asset). DAI inspite of its low supply has far more usage than some of its peers. A significant contributor to this is the fact that DAI’s supply is relatively small as its issuance is reliant on the assets that are used for CDPs in MakerDAO. On the flipside, USDT has a considerably lower velocity. It makes me believe much of USDT is used for one-time transactions and storage of dollars on a digital ledger. Value is simply moved to an exchange and held there instead of being used in a decentralised product like a dex.

Retail users on the rise

One way to gauge the nature of users of stable-coins is to study how much money is being transacted through the network. We could look at both the volume and the count of transactions to understand what is driving value. My assumption here is that there is no reason individuals would keep doing hundreds of low-value transactions due to the nature of the Ethereum fee model. Therefore, higher the number of low-value transactions, higher the number of retail users doing these transactions. The largest sub-category here is individuals doing transactions of size between $100 and $1000. Over 750,000 transactions have occurred in this category most recently in a single weel. The figure is at around 450,000 for those between $10 and $100.

But whales dominate volume

While larger transaction sub-groups have not increased in the pace at which retail traders have been entering the market, they have not paled either. In fact for those transacting over $100k-$1mil a recent high for monthly transaction count was set in May at over 71,000. While these graphs don’t make me think “institutions are rushing in”, it does give some anecdotal evidence of stable-coins being used as a money remittance system by those that have the means to move weight.

A better display of the impact of this on the volume is evident when you look at the total on-chain volume contributed by each transaction sub-group. Transactions over $1million in size were responsible for ~$21 billion moved on-chain. In may that figure was at $18 billion. Those doing transactions of upto $1000 in size did a mere $1 billion in on-chain volume. Seems like whales dominate how value moves in the industry.

Exchanges are still a driver for adoption

Tether is a controversial project, but as I have said earlier – it has been critical for adoption. Almost “too big to fail” at this point. The ~20% or so of Tether supply being on exchanges also makes me believe much of the distribution to the end-user likely happens through OTC desks. For some strange reason, that reminds of Liberty Reserve. Bad memories aside – it is likely that by the time I write an update to this piece, the stablecoin ecosystem diverges into what could be trading and financial applications. We are seeing some instances of this with crypto native individuals using USDC for payroll management and DAI for purchasing NFT. That transition from a trading first utility to digital currency used for alternative use-cases will likely be the most significant leap stable-coins have to make going forward.

I will leave with this chart showing the exposure of major exchanges to prominent stablecoins on basis of their wallet balances. Yes, much of it is concentrated on Tether with the exception of Coinbase. Make of it what you will..

See you on Friday with some charts explaining how stablecoins may be upending the fat protocol thesis.

Data sources : Nansen and Santiment


Stable Coins In 2019

Stable Coins In 2019

Note : If I were to be idealistic, I’d likely have only DAI in this data-set. However, markets don’t really care about my idealism. As much as there is controversy around USDT today – fact remains that it is a crucial component of the stablecoin ecosystem. I would appreciate it if conversations that stem from this piece does not go into “is tether a stable coin”. My honest stance on the matter is that in the absence of verifiable audits, it is not.

Stable coins are what happens when crypto finds product-market fit. A pro-longed winter in 2018 combined with increasing scrutiny from banks around the world set the stage for adoption of stable tokens, which in turn has fed into the growth of the broader DeFi ecosystem. About a quarter trillion dollars had been moved on-chain through stablecoins when I began work on this piece (early November). However, little literature was available on who leads, to what extent and the nature of user-behaviour on these chains. What followed has been a month-long exploration of data from Token Analyst, Santiment and loads of procrastination as I struggled with the idea that currency kept in a central bank somewhere, represented as a token on-chain has seen the most traction in 2019. (Discounting DAI ofcourse). This is my attempt at summarising what has been going on in terms of volume and user behavior within stable coins.

Stable Coins In 2019

If you need any indication of the fact that stable-coins and DeFi as a theme are here to say, a look at the most active networks should lay the case for you. Six of the top 20 networks, ranked based on average active wallets over 30 days are engaged with directly or indirectly with stable coins. Tether takes the lead here with 40,742 wallets (vs Bitcoin’s ~750k). DAI is a distant second at 2752, followed by USDC and Paxos. This “trendsetting” done by USDT, and DAI’s attempt to play catch up while stable coins issued by established incumbents (i.e. – Circle, Gemini) lag behind DAI is a common theme throughout this data-set.

The Year In Volume

Stable Coins In 2019

Over $237 billion has moved through on-chain volume in stable coins over the course of the year. Much of this could be attributed to demand being driven by exchanges. While it is easy to jump to a quick conclusion that USDT’s two chains (Omni and ERC-20) dominate volume, we have for the fact that the ecosystem to use them today is much larger than those for other chains. What could very likely cause a change in this scenario is if the market for DeFi instruments increases exponentially to a point where it surpasses that of exchanges. This would mean easier on-ramps, a suite of products that use stable tokens as a payments instrument and wallets that make it easier for users to handle stable tokens. Projects like Argent and Mosendo are paving the way towards making that happen.

USDC and DAI for instance very likely see more volume being driven by lending markets (eg: Juno, Dharma, Compound) and exchanging requirements (eg: Uniswap). It will be interesting to see how this evolves over the course of the year.

Much of the market today is dominated by USDT. In volume terms – that figure comes to ~80%. The oddity of an ecosystem that claims to work towards decentralisation relying on a centralised currency with no verifiable audits should not be missed upon us. In order to account for that – I decided to take a look at what a world without USDT would look like in terms of market share. Centralisation (and brand) still plays a role here. USDC managed to have close to half of the entirety of the market’s share with 45%. DAI and Paxos were quite close to one another at ~20%. What did seem odd here was that GUSD, inspite of the brand and incentives that were released earlier in the year had 2% of volume without accounting for USDT, and 0.4% when accounting for it. The power laws at play here are brutal

Stable Coins In 2019Stable Coins In 2019

Figure to the right is market-share split without accounting for USDT. Brand and central custody, still a requisite for traction.

DAI is the only stable-token that has actually grown in volume over the course of the year if I am to discount USDT. It’s volume has seen a ~300% hike since January 2019. The introduction of multi-collateral DAI will very likely make this number jump further up. Its volume has also begun rivalling certain other centralised offerings such as GUSD. This is likely the earliest indication that a DAO run business, with an ecosystem around it can take on centralised alternatives and beat them in terms of metrics if additional products are built around it.

Stable Coins In 2019

The adoption that exchanges offer combined with the speed that ethereum offers has over-thrown both Omni and all other prominent stable-coins combined. If anything, 2019 is the year USDT-ERC20 established itself as a lead, MakerDAO explored its tao and other projects languished (on basis of volume).

Stable Coins In 2019

Product transition on point. 10/10

Stable Coins In 2019

Would sign up for lessons on blitzscaling from USDT-Erc20

Getting Transactional

Volume however, tells only one part of the story. In order to understand what may be happening in these projects, one needs to explore the number of transactions each chain handles and the frequency at which it does. In order to do so, I looked at

– Number of active wallets on each chain

– Number of transactions on each chain

– Amount of volume contributed by each chain.

One way to understand this data is that as the number of transactions increase in a network, the average value of each transaction may reduce. This is with the understanding that as adoption increases, individuals may not store large amounts of wealth on a stable token but instead “use” it as a utility. A chain may have very high volume on it during its early phase (eg: Paxos) as those setting it up issue assets and move it to partners. However, if adoption does not kick in, the average transactional volume per address will remain high indicating that whales dominate the network. The graph below shows DAI compares with its peers. For a reference of scale, USDT handled a total of 20 million transactions between the ERC-20 and Omni variants over the course of the year (till November).

Stable Coins In 2019

In conversations with a number of analysts about the average volume per user- there were two primary conclusions that routinely emerged. One was that the average volume on a chain being high for each user is an indication of the fact that confidence in it is high. By this logic, Paxos is likely a very preferred chain for institutional transfers as large volumes are moved on it. The other, is the utility argument that as retail adoption does increase – we will see a substantial dip in the average volume per wallet. For DAI and USDT this does ring true with an almost ~80% dip from their ATH figures to what they saw in August and September.

However regardless of how one looks at the data – the conclusion I have been coming to is the fact that whales still dominate both DeFi and dApps. And if an app or utility is unlikely to capture them early on in their growth cycle – they may likely not see much traction due to the small market that crypto today holds. Ideally, as this ecosystem evolves – the number of on-ramps increase, and the active wallet count does increase. If volumes remain stagnant (or grow slowly in comparison to user growth) the averages shown below should dip substantially. That is the idealistic scenario for 2020.

There are a number of things that could contribute to it. Some of them are

  1. Card linked crypto wallets being more common (eg:

  2. Mobile wallets (eg: Samsung)

  3. Browser wallets on the rise (Opera, Brave)

  4. On-ramps growing (eg: Local Crypto,

Stable Coins In 2019Stable Coins In 2019

What I did however find fascinating with the whole space is that the average transaction from each address on any given day was roughly around ~2. This figure was the highest on DAI, with it ranging all the way up to 5. To me this indicates that individuals still use stablecoins as a “volatility hedge” and don’t see it as a transactional layer yet. The average transaction per wallet on dApps are relatively high even by Ethereum standards at ~4. My hunch is that if EOS and Tron based stablecoins were included in this number crunching exercise, this figure would have been much higher.. but that is for another day. DAI does have a higher average transaction per month count due to the fact that it is used for use-cases other than exchanging alone.

Stable Coins In 2019

Beyond Exchanges

Personally, the question I have been asking myself has been what would it take to build a unicorn with DeFi and stablecoins at its foundation. If regulators catch up, and provide a stable framework for the growth of the space, many of tomorrow’s Stripe, Paypal and Monzo remain to be built. That can only happen if retail adoption does pick up massively. Some of the markets I have been tracking (and linking stablecoin uses) to are remittance, the gig economy, digital asset insurance, income share agreements and DAOs. As much as exchanges are fascinating, when I study B2C apps and last mile solutions that blockchain based products have built – there is ample space for growth left. The likes of Bitpesa and have not only pioneered in frontier markets but also set the stage for a new generation of fintech businesses built on blockchain first to take the spotlight. A very early instance of this has been LocalEthereum – it caters to the world, with a very small team, has consistently seen volume growth and runs most of its function on a blockchain. I believe stablecoins could hypothetically do to money, what the cloud did to data. It paves avenue for extremely lean teams to cater to millions with smart contract enabled interactions. Whatsapp and Instagram were hyper-lean teams built to cater to customers in the mobile era. I am still looking for that “aha” moment where a consumer based app does the same with stablecoins for banking.

Whenever you think all the good ideas are taken, remember this: We put a man on the moon before we put wheels on suitcases.

— ᴅᴀᴠɪᴅ ᴘᴇʀᴇʟʟ ✌ (@david_perell) December 2, 2019

Until that magical start-up lands up in my network or I end up setting out to build it -here’s some inspiration for you if you believe exchanges are the only ones that can drive stablecoin adoption.

If you are indeed one of them – make sure you slide into the DMs on Twitter. Would love to hear from you!