Non-Fungible Tokens (NFTs) have taken over the hype-zeitgeist of the moment. On the one hand: cool! I love it when weird new technologies upend the applecart of how we expect money and markets to work.
On the other hand: now those of us who consider us (with some small touch of irony) “Market Professionals” now need to understand these things well enough to at least hold our own in a conversation. After all, when NFTs are selling for tens of millions of actual U.S. dollars, it’s only a matter of time before a client (or your college-bound senior) walks in and declares that NFTs are going to cure portfolio cancer. Let’s cut through the hype, shall we?
First, a few resources:
- A really good 101 on what NFT’s actually are from Decrypt.
- My piece from a few weeks ago, on how IP-laden NFTs could radically transform investment management, in a magical world where regulators could keep up.
- Corey Hoffstein and Jason Buck’s video on NFTs is really informative (and entertaining).
- SNL’s rap-skit on NFTs is really entertaining (and at least somewhat informative).
But assuming you skip all that, the TL;DR on this is a single line from a brilliant article by Crypto-VC Nic Carter describing Grimes NFTs:
What I’m buying is effectively a digitized version of a signed setlist after a gig, or a signed, limited edition album cover. As I jokingly put it, the NFT should be understood as the autograph, not the art.
Under the hood, NFTs are smart contracts, generally on the Ethereum blockchain, which the author (an artist like Beeple, a celebrity like Elon Musk, or a company like the NBA) creates (and then sells). The contract links to some form of intellectual property (an image, a video, and/or various metadata) — but it isn’t the IP itself. Your NFT doesn’t contain, in any way, the video of that killer dunk, or the music of Kings of Leon, or even the 5000 images collaged into Beeples Everydays. It simply refers to that IP, often with some vague promises (this is 1 of 300, or this is 1-of-1 and I promise not to mint another one, etc.). Once a given address is established as the owner of a given NFT, that’s kind of it. The address owns the NFT, until the holder of that address sells the NFT (or theoretically, the whole address).
But the NFT itself? It’s just the signature, and like any signature, it’s value is going to be dependent on what that signature represents to the buyer.
Why People are Bullish/Incredulous About NFTs
People who are super bullish on NFTs argue that they’re a huge improvement to the infrastructure around owning collectibles: they’re easier to buy, easier to store, easier to sell. Collectibles have a huge history as an investible asset, and big businesses like Rally and Masterworks.io are taking the traditional way of owning collectibles (buy the thing at auction, pay to store it somewhere) and financializing it, selling shares in collectibles dark pools to create portfolio exposures. NFTs are, to the bulls, just the logical extension of this business, eschewing the physically scarce object (the car, the painting) in favor of digitally scarce objects (the NFT). After all, the thinking goes, you can download a GIF of a Picasso too.
People who are super skeptical of NFTs point out that all you “own” is a string of 1s and 0s. This is of course true — you’re not buying a physical object. Skeptics also point out that your NBA TopShot doesn’t give you any rights to the images or videos referenced, and even the buyer of Beeples $69 million masterpiece didn’t get any rights to the IP itself — no copyright changed hands. (This is, of course, entirely true of physically collected art as well. You don’t own the copyright to that Banksy you bought, or the one he painted on your wall.)
When your client or colleague starts talking about NFTs, chances are they’ve already bought into one side of this narrative: NFTs are either worthless bubblefuel, or they’re going to completely transform ownership. There are fragments of truth to both, but let’s turn down the volume and think about NFTs in a portfolio context.
All prices contain some level of psychology – the entire field of Behavioral Finance is based on studying just how much, why, how, and under what conditions. Whether you’re haggling over a piece of jewelry in a bazaar in Marrakech or buying a share of SPY, I don’t know many folks who would argue that those prices are perfectly efficient, representing the completely mathematical result of calculating the probabilistic current and future utility of the asset. Even old-school hardcore chart-jockeys understand this: they don’t believe support lines or pennant charts are somehow voodoo magic, they believe they proxy the psychological state of price setters and price takers. Technical analysis is replete with evidence for psychology.
The really important question to ask yourself (and more importantly, your client) is this:
How much of your portfolio value is based on psychology?
A quick exercise: take a look at a sample client allocation: maybe some equity ETFs, some bond ETFs, maybe an active manager or a thematic product, some gold, maybe a REIT fund. Ask yourself what portion of the value of each security you believe is based on psychology, versus based on a mathematically justifiable assessment of value. If you want to get mathy, you could use discounted cash flow based analysis with a valuation of book value, and you could come up with numbers. Regardless of how you decide to math it up, the difference between your B-school inspired “book value” and the current actual trading place is, essentially, the psychological value embedded in the asset.
Every asset you own contains some combo of this intrinsic and psychological value. It’s hard to come up with an asset with zero psychological value. An apple on your counter is pure utility. But if you bought the slightly more expensive Apple because you like the taste, or because you feel good about supporting the local farmer, well, you put psychological value into your purchase.
Collectibles like art, baseball cards, signed prints, etc. are almost entirely psychological assets. There are of course nuances: if you invest in wine futures, at the very least, you’d drink the wine eventually, and you may assign a huge aesthetic value to that. Similarly, if you buy that rookie baseball card and put it on a shelf, you may get enormous value from knowing its there, showing it off to friends, or just keeping that space in your head where baseball lives. But as a portfolio asset, the value of those things — just like the value of my NFT or a bar of gold — is entirely and completely psychological. There is no value in it other than what someone else would pay for it.
In between these two extremes are where most people actually build wealth.
Owning SPY gives you legitimate claims on the future earnings and assets of the firms in the fund (intrinsic value), but it’s pretty obvious that XOM’s price carries increasingly negative psychological value. Tesla contains a healthy heaping of positive psychological value. (Ocean Tomo did that excercise from the bottom-up accounting perspective, and they posit that 90% of all public equity business value is currently intangible.) Some commodities, like industrial metals and some aggs, have strong foundations in basic utility. Nobody’s stockpiling corn for a decade, it mostly gets consumed. Gold, on the other hand, is the mother of all psychological assets; it’s value determined almost entirely by the global conversation around risk. It carries so much portfolio respectability primarily because it’s our oldest fungible asset. Millenia of conditioning is hard to shake.
In no way am I being pejorative about this. Psychological value is important. I have certain physical objects I own to which I assign enormous psychological value, mostly physical assets that I have a history with: pictures, a revolutionary war powder horn that belonged to an ancestor, artists maquettes from the game Bioshock, Vinyl records. I am not suggesting things laden with psychological value have no value.
I would suggest, however, that assigning portfolio value to these intensely psychological assets is tricky. We don’t accurate measure either the potential return of psychological assets nor do we have a deep understanding — intuitively or academically — of their risks.
So How Do You Talk To Your Clients About NFTs?
Try this on for size:
Non-Fungible Tokens are a kind of signature, like the autograph on a game winning ball. Like all collectibles, then, any NFT has value determined by the psychology of like-minded collectors. How much of your portfolio exposure is currently psychological, and if we are adding NFTs, what are we selling to make room on the psychology shelf?
I really think it’s that simple.
The Big Postscript
(You can stop reading here. The rabbit hole lies below and I won’t be insulted.)
In the process of learning as much as I can about, well, everything, I started looking at precisely what’s built into most of the NFTs changing hands these days. To me, the two big critical flaws I’ve seen in the current NFT ecosystem are the legal framework and the reference framework. Both of these are solvable problems, but I do think they need to be solved.
The legal issue is quite straightforward. As I pointed out when I was minting my NadigHour tokens (which are still unsold, if you’re interested), the only actual value you could associate with the NFT is either purely psychological (you like me! You really really like me!) or it has to be based on some form of trust.
I promised, in the description of the NFT, to redeem the NFT for one hour of my time. However, you have no real legal recourse should you attempt to redeem it except that of the implied verbal contract. Maybe you’d win, maybe you’d lose. More importantly, in the process of minting the NFT, at no point do I actually prove that I am who I say I am, that I am intending the NFT to be a signature on a contract, or what the terms of that contract might be, or even how to reach me. Even the simplest real-world agreement covers these kinds of basics (to say nothing of issues like governing law). I could have minted the exact same token promising 1 hour of Elon Musk’s time, waited to get paid, then disappeared. Sure, Elon could come after me, and perhaps the buyer could come after me, but good luck trying to chase me back through various wallets and transactions.
It’s not just me. When you buy a moment from TopShot, TopShot and the NBA aren’t agreeing to much (the terms of service are a trip. Essentially while you own your moment — the token — NBA/Dapper has no obligation to do anything to maintain the Flow blockchain, to maintain the app, or to facilitate anything you might ever want to do with that token. If your sole proof of ownership is through the app… well…have fun with that.)
NFTs have the potential to be incredibly powerful tools, but at the moment, there’s scant legal framework to do much of real value, especially when the use cases cross over into anything that could be viewed as financial. That means everything is about arms-length trust. Beeple, for his part, has been doing the digital art→NFT thing for a very long time in Crypto-years, so people trust that the things he’s claiming about his NFTs will be true, and someone like Christie’s being the auctioneer adds to that trust. But if you read the Christie’s docs, it’s pretty clear you’re just buying an NFT from a dude, and its value is based on how much you value Beeple’s work and word.
Referents and Intellectual Property
If an NFT is fundamentally just an autograph, what that autograph is connected to is important. If I have a verified autograph of Ian Curtis, lead singer of Joy Division, well, that’s worth something — he died young, his fans are legion. If I managed to find a verified Ian Curtis autograph on, say, the factory pressing of Transmission, it would be incredibly valuable, despite the music being available on Spotify. It would be even more valuable because bandmate Peter Hook has admitted he used to forge Ian’s signature. In other words: an NFT here might actually be helpful, if Ian had been around to somehow verify his intentions on minting one.
But at the end of the day, the connection between the signature and the item clearly matters. And this is the second big issue: most NFTs connection to the IP they reference is very very sketch. In many, if not most cases, the actual image or video that’s been “signed” is just a link to a server somewhere on the internet, and those links break with breathtaking speed. This isn’t an unknown issue. Crypto circles are buzzing with both the observation of broken NFTs, and proposed new standards that could deal with the issue. I think it’s telling that a high-profile Op-Ed in the New York Times about NFTs had to be immediately amended on publication to correct the common misunderstanding that NFTs contained actual assets:
Correction: March 24, 2021 An earlier version of this article misstated a detail about a GIF sold by Li Jin. A unique link to the moving image in the GIF and its metadata were stored on a blockchain; the image itself was not.
The solution to this might eventually be IPFS (which stands for Interplanetary File System – welcome to nerdworld!) — a kind of universal file storage lookup system that doesn’t require a physical address, simply a hash (a mathematical proof a given file is in fact what you think it is) and referents to find the file out on the big cloud of IPFS servers.
It’s a cute idea, but ultimately, it relies on either the kindness of strangers continuing to host your MP4 video file for all-time, or paying a company to “pin” the file in the network so a copy is always available. So, A for effort, C- for actually solving the particular problem.
There are solutions to both of these problems, but they don’t mesh well with the ethos of the crypto community. Symbiont.io solves both of these problems right at the blockchain level. Symbiont runs high-value, small-scale blockchains, such as the one that Vanguard and CRSP use to disseminate index data. Their solution embeds the IP — index constituents and related data — right into the blockchain. This means that your local node is going to have big storage requirements, so you’re not setting up a node on your laptop. But that’s fine because they solve the other problem by being deeply non-anonymous. Their blockchains have authenticated nodes and users, and access to the IP (the index data) is permissioned the way a lot of things are permissioned — using key management.
Purists get nauseous at the very thought of this: it’s centralized! Anathema! But I mention it only to point out that these are solvable problems. If you want NFTs to be REALLY valuable, then we need to create a legal framework in which they exist, and a technological backbone that allows for significant secure storage of value.
The Psychological Asset Conundrum
Finally, it’s worth pointing out that we’re in the midst of a huge explosion in psychological assets. The entire crypto ecosystem started as 100% a psychological asset — it had value because of unproven potential. It’s evolved, of course. Even the most die-hard Bitcoin hater wouldn’t shred they keys if gifted a Bitcoin — they’d sell it fast to fund a semester of college.
What’s new here is network effects. The psychological component of value relies on commonality of psychology: you need to believe my bar of gold is worth something if I wish to use it as a medium of exchange, or anything else. One of the really exciting things about Crypto and De-centralized Finance is that it’s already carrying huge network effects. Millions and millions of folks are now wrestling daily with the value of dozens of digital-only psychological assets, and that increases the changes of success for the group as a whole (while acknowledging there will be crazy unpredictable winners and surprising losers here).
I suspect we won’t solve the legal side of this particularly elegantly. Instead, we’ll sit and wait until someone scams someone or something breaks, and the lawsuits fly, and then regulators and perhaps legislators will get involved. But we’ll get there eventually.
The Referent/IP issue will also get solved one way or another. Off-chain storage in a universally accepted manner is probably the ideal, but I suspect we’ll see trusted entities wade in sooner than later, acting as, essentially, the backstop for NFT IP. Kind of like a custodian that holds gold in a vault in London to back up your shares of GLD.
[Got some thoughts? Ping me @DaveNadig on Twitter. DMs are always open]