Tokens (Part II) - Ponzinomics or the Future of France?
Crypto Network Valuations vs. their Web2 Counterparts
Hope you all had a relaxing holiday period and were able to spend some quality time with family. Always goes by too fast.
Back in Singapore now & in quarantine this week, so no excuse not to post :)
This week, we explore:
Tokenomics: the raging debate between fatally flawed Ponzi’s vs. the Future of France
Why fundamentals still matter: valuing Web3 with familiar consumer internet metrics
Should multiples for protocols be lower than internet and software businesses because of more limited user lock-in at maturity?
Growing revenues by charging less - lessons from the video game industry
MCGA: Making Cashflows Great Again - the hunt for “fundamental value” in crypto amidst macro uncertainty
Going Micro
Last post, I wrote about the new paradigm for building communities. I wrote about tokens and communities superseding aggregators as the dominant mechanism for bootstrapping networks online. I wrote about digital land reform and property rights. About novel governance and transparent legal systems opening up the design space for billions globally this decade.
Macro bullsh*t as my pseudonymous pal Distilling Frenzy enjoys reminding me.
This week, we will switch gears a bit. We will don our skeptical boomer monocles to address many of the circulating concerns regarding the rising pseudoscience of tokenomics (a.k.a “ponzinomics”) in the crypto sphere.
“Sure - number sometimes go up”, you admit, “but is any of it sustainable? What are token values actually backed by? What are the barriers to entry if everything is opensource? Isn’t it all just elevated inflationary rewards riding a bull market backed only by the greater fool theory destined for a reflexive nose-dive whenever market sentiment takes an inevitable turn? You know… like the one we have witnessed in early January?”
In short, is any of it sustainable?
To assuage these concerns, we first need to admit many tokens today are vaporware. The token is unnecessary. The economics make no sense. The projects are riding speculative euphoria and will crash in a blazing fury of spectacular nothingness - like 98%+ of ICO projects during 2018.
However, like the internet champions, a handful of projects will prove transformative. Therefore, it is worth trying to parse through the noise to find the gems.
At risk of being dismissed as a dinosaur, I’ll contend we can delineate many gems using our same tried and true web2 investment frameworks.
Unit economics
Competitive Dynamics / Barriers to Entry
Total Addressable Market (TAM)
Valuation
Unit Economics
Many in crypto believe traditional financial analysis is dying: a relic of a bygone era of positive real interest rates, cash flows (lol), and “fundamental value”. The truth is, while a corrupted yield curve has distorted discount rates, traditional financial analysis is still quite relevant in valuing these novel networks. While almost all networks are cashflow negative during the growth phase, unit economics help to discern which have a path to sustainable value accrual.
If you are an entrepreneur, investor, marketer, or growth hacker, you are intimately familiar with the concept of unit economics.
I, too, spent the last six years of my life scouring the globe for companies exhibiting the magical combination of high growth, large addressable market (TAM), market leadership, and (the increasingly elusive) compelling unit economics. Yes, I need to get out more.
The practice generally entails modeling out the life-time value (LTV) of a single customer divided by the cost to acquire said customer (CAC). Best-in-class consumer internet companies tend to have LTV / CAC ratios of ~3:1 (i.e. the expected cumulative margins attained from 1 customer over the ensuing ~five years is 3x the amount paid (probably to Facebook or Google) in order to acquire said customer.
While Web3 represents a new paradigm for bootstrapping networks by baking in financial incentives to get the network off the ground…
Source: @ChrisDixon
… the exercise of valuing the network is similar to its web2 counterparts.
Control F: “Shareholder” → Replace with: “Token-holder”
Control F: “CAC” → Replace with: “avg. inflation rewards per user”
Control F: “LTV” → “Replace with: avg. protocol revenue per user”
Web3 networks are more like internet companies in the emerging world (like Shopee, OVO, Swiggy, GCash, Shopback, etc) who often find it more effective to offer subsidies to users directly as opposed to spending on Facebook ads.
The reality today is the unit economics for most projects don’t work1. Inflation rewards are dramatically outstripping protocol revenues and burns (token buybacks) on a per user basis. However, that doesn’t mean all are destined to fail.
Over the last decade, many leading aggregators had poor unit economics in the scaling phase (including almost all >US$5b companies in Southeast Asia today. In a low-rate environment, raw market share matters more.
In a Web 2 world, investors will pay more today for the market leader because they will reap monopoly profits on the back end.
The issue becomes, in a world of opensource software, are there any monopoly profits to reap at maturity considering user lock-in is much more difficult?
If not, shouldn’t protocol growth subsidization be lower today to account for lower switching costs down the line?
So, either…
1) the TAMs of these protocols are massive OR
2) there are barriers to entry at maturity OR
3) Relative valuations should not be as high in web3 as traditional web2 marketplaces or software businesses because user lock-in is materially weaker
Let’s explore…
Barriers to Entry
Given the blockchain is opensource, competition is particularly cutthroat. The space is prone to “Vampire Attacks” like the one between Sushiswap and Uniswap, where a project can “fork” the code base of another, tweak a few parameters, and ramp incentives to take away users and market share.
The possibility of a competitive fork serves to cap the level of extraction a protocol can generate leading to relatively slim take rates in web3. Still, there are barriers to entry for protocols which provide an element of defensiveness and should allow them to accrue network effects, while simultaneously capping the extraction value.
For Layer 1 blockchains, the primary barrier to entry is security. Security is driven by a combination of decentralization and value. The more decentralized and higher value a (Proof-of-Stake) chain is, the more difficult it will be to commit a 51% attack on the network2.
At the app-layer, Multicoin has a compelling series of writeups detailing key barriers to entry in Dapps (defi applications). The TL:DR:
“Unforkable state” - typically liquidity & collateral locked in a protocol, but can also include 3rd party integrations
Network effects from liquidity - preeminent example is Tether
Track record and trust - Lindy
Back-stopping of risk: insurance funds or governance tokens which backstop protocol risk which is difficult to fork
All present barriers to entry, up to a point, but still keeps protocol extraction in check as the community can always fork / exit.
The question then becomes, should investors pay comparable multiples to software or internet business today for protocols which are likely to be less defensible at maturity?
In my opinion, the answer comes down largely to the expected total addressable market (TAM).
TAM
The success a16z and Tiger Global have had as growth investors over the last decade has emerged from a relatively simple insight.
Internet and software TAMs are MASSIVE.
“Honestly, Ben, it doesn’t really matter if I pay 25% more at Series A, B, or C. Everyone is under-estimating how big this space will be. These are the guys to back
- Marc Andreessen, at a 2011 investment committee, while scribbling notes for “Why Software is Eating the World” … probably
The continual 20% growth YoY of FAAMG or BATs or enterprise software champions ~15 - 20+ years after founding is a screaming tribute to the accuracy of this thesis.
Through this lens, the only logical thing to do is to deploy. AGGRESSIVELY.
Most investors passed on Uber early because they projected the TAM as the Taxi market. Most investors passed on BABA, Grab, and SEA because eCommerce penetration curves were modeled on the West. Most investors under-estimated Amazon’s ability to expand into new business lines.
The TAM for new business models has a way of expanding in ways most investors cannot predict. This is especially true as networks open.
The video game industry is an amazing example of business model innovation expanding the TAM considerably by charging users LESS. High level evolution:
Buy a game for ~US$50 to play on your (~US$250+) console indefinitely
PC streaming: pay-to-play
Free-to-Play (largely now on mobile)
At each step, by charging less, the industry reduced friction to onboard new users, dramatically expanding the player base and growing the monetization considerably via in-app purchases.
Matthew Ball has a fantastic graph outlining the success of the gaming industry vs. other entertainment mediums like music.
The largest markets for gaming today (in terms of users) are in the emerging world: ~ 1 in 4 humans ON EARTH today plays video games regularly. And it turns out, once onboarded, these individuals are likely to invite their friends and spend considerably on digital goods.
SEA Free Fire’s mind-boggling 729 million quarterly active users with ~13% spending in-app (93.2 million paying users) is the perfect case study. With ~10% of the globe playing your game, you just don’t need that much revenue per user to prance into the ranks of top 10 gaming publishers globally…
The music industry on the other hand has shrunk - remaining closed to business model innovation. Until now :)
My hypothesis is web3 may provide a similar catalyst to many verticals, but financial services in particular. Financial services today are notoriously inaccessible, have siloed tech stacks, thrive on regulatory capture, and provide relatively poor services. By unleashing thousands of open-source developers with composable financial primitives and compelling financial incentives to early adopters, I suspect the TAM will expand tremendously.
Despite the lower expected take rates at maturity, a16z and Paradigm are now making the same bets in web3 that a16z and Tiger Global made in web2 from 2010 - 2020.
Once again, I suspect the thesis is relatively simple: The TAM of the leaders will be enormous.
Valuation
Alas, we arrive at the trickiest of topics. How much to pay today for the combination of existing traction, future market growth, and ultimate value capture. A combination of art and science, micro and macro, discipline and risk.
My own sense is crypto will play out a lot like the Nasdaq - but accelerated.
The trouble is I’m not sure if we are currently in 1999 or if 2017 was 1999 and we are now entering ~2014 where the money, talent, composability, and institutional adoption flywheel is in the early innings of a decade long crescendo.
A few data points:
~300m3 crypto owners today vs. an internet population of ~5 billion (~3.9% penetration)
~10m4 MAU for DEFI apps today vs. 3.8 billion5 accessing Tradfi (~0.3% pent.)
~18k6 monthly active devs in web3 vs. 26.9m software devs globally (~0.1% pent.)
By 2030, it’s obvious those penetration figures will increase dramatically; especially given the global proliferation of smart phones, data costs plummeting, and consumer time spent online (now ~6.6 hrs per day).
On the flip side, a market cap of ~US$2.1 trillion is a lot to pay for 10m monthly active defi users and ~18k active developers. A lot of future growth is already priced in for an asset class which is promising, but largely self-referential and speculative.
Despite obvious structural tailwinds, the adoption path is unlikely to be straight up and to the right with the revenge of Jerome Powell’s (potentially 4!) rate hikes in 2022 putting a serious damper on the speculative growth fervor of 2021.
In 2022, crypto exposure is a must for any serious investor given the massive disruptive potential and talent pouring into the space. However, the shaky macro backdrop, speculative price run up in 2021, and historic bouts of volatility all cause my “fundamentals” and “boomer” mentality to resurface. To hide in the only real solace for any investor: cash flows and “reasonable” relative valuations.
To the surprise of many pundits, there are real bargains to be had in crypto.
Price / Earnings (Fully Diluted Market Cap / Annualized Protocol Revenues)
Source: Token Terminal
While the metrics are not apples to apples (protocols vs companies), many will find it surprising Ethereum with a market cap of ~US$400b7 trades at ~29x annualized protocol earnings, roughly in line with the Nasdaq (~27x P/E8) despite Ethereum’s 2021 Q4 YoY revenue growth clocking in at US$4.3b (up 1777% from Q4 2020).
Or many yield aggregators like Yearn Finance (12x P/E) and Decentralized Exchanges (Pancake, Trader Joe, Spooky etc) trading between ~13 - 15x P/E, despite rapid growth in AUM and revenues during 2021. These valuations look downright cheap relative to equities.
For the masochists out there, please scan through Adam Cochran’s 2021 bird app magnum opus outlining his shopping list for 2022.
The TL:DR (for sane people who refuse to read 206 tweet threads… ): in a world with more conservative monetary policy needed to tame inflation well above target levels, protocols with actual cashflows are likely to outperform. In essence, crypto, like to the stock market as of late, is likely to see a return to value.
And despite the naysayers who think all tokens are Ponzis, there are a minority of projects with real cash flows to provide downside protection when sentiment turns. So, to correct the ledger of record, only MOST projects today are ponzi’s :).
In Short
The crypto markets, like the equity markets, are nuanced. At any given time, there are pockets of euphoria and soaring valuations beyond fundamentals and neglected pockets with depressed P/E multiples awaiting the inevitable investor rotation.
Annoyingly, many of the same skills in valuing higher growth network businesses - unit economics, competitive dynamics, TAM and market capture - apply to the crypto markets as well.
New ownership structure, new customer acquisition model, less room for extraction, but similar valuation techniques.
Despite the opensource nature of crypto putting barriers on extractive take rates, my own bet is by opening up these networks to new participants, the TAM for winning protocols and Dapps will expand dramatically; more than making up for lower take rates. The Ponzis will fall away and the gems will surface.
For those who can stomach the interim volatility, many retail investors today are following the Tiger and a16z playbook of the 2010s…
Don’t bet against the TAM of software markets - especially now that these markets are global, composable, and can be owned by their users.
Disclaimer: This is a gross generalization; all data is on-chain. TokenTerminal is an amazing resource for analyzing protocol fundamentals and showcasing those few tokens which trade at reasonable valuations today. https://www.tokenterminal.com/terminal/metrics/ps
Note to crypto-nerd readers: while other Layer 1’s are seeing high growth, I ultimately buy the “modular” blockchain scaling thesis and that over time, the vast majority of execution will occur on ZK-rollups and other scaling solutions, so the primary differentiation at Layer 1 will be security not execution over the long-term).
Global crypto ownership: https://triple-a.io/crypto-ownership/
Metamask MAUs >10m as of August 2021 https://consensys.net/blog/press-release/metamask-surpasses-10-million-maus-making-it-the-worlds-leading-non-custodial-crypto-wallet/
World Bank - Financial Access. https://www.worldbank.org/en/news/press-release/2018/04/19/financial-inclusion-on-the-rise-but-gaps-remain-global-findex-database-shows#:~:text=Globally%2C%2069%20percent%20of%20adults,crucial%20step%20in%20escaping%20poverty.
Electric Capital Developer Report (2021) https://medium.com/electric-capital/electric-capital-developer-report-2021-f37874efea6d
Coinmarket cap (Jan 16 2022)
https://ycharts.com/companies/NDAQ/pe_ratio (Jan 14 2022)