What NFT mania can tell us about market bubbles

Remember NFTs? They were like Pet Rocks but instead of a rock there was fraud. Now they’re dead, pretty much, and what’s left is data.

It’s data that might prove quite useful, in a way that proprietary digital receipts for JPGs of monkeys never were, because it promises a 360-degree view of speculation in its purest form.

As well as being absent of any fundamental value, NFTs were practically impossible to short or hedge, so they appealed almost exclusively to retail punters. Add in public blockchains that give each trade provenance and traceability, and we have what’s probably the first diagrammatic description of how the actions of individuals foment a madness of crowds.

That’s the premise of a paper from Swiss Finance Institute professors Andrea Barbon and Angelo Ranaldo, who trawled more than 15mn NFT transactions in search of what creates bubbles. Their conclusions — that in markets rife with pump-and-dumps the same few traders can make out like bandits — may not be wholly surprising, but it’s good to look at the evidence.

From more than $18bn of NFT trades between January 2021 to September 2022, Barbon and Ranaldo identify around 1,000 events where the average sale price of an NFT collection at least doubled within 24 hours. Slightly more than half the time, a the pop was followed immediately by a drop.

High volatility on low turnover is a clear warning that a crash is coming, irrespective of what’s happening in the wider market. Barbon and Ranaldo also find that price pops involving fewer participants are less tenable, perhaps because they draw in a higher concentration of optimistic and overconfident investors.

Since these same dynamics are also factors in tradfi bubbles, the authors to argue that “the NFT market is not inherently more irrational than traditional financial markets”. Okay.

Where the paper builds most on existing research is its study of whales versus fish. Wallet analysis identifies the NFT traders who were consistently best at timing the market without getting drawn into momentary hype. They were the ones who traded in relatively greater volume across multiple exchanges, held their positions for longer, and were more likely to apply leverage.

The implication, that sophisticated retail investors are more talented at making money than speculators, might appear obvious to everyone other than a student of price discovery. Yet the conclusion, that it’s safest to follow whales, will probably sound like nonsense to anyone who’s been paying attention over the past few years.

Fraud is of course the giant pixelated elephant in this particular paper. Barbon and Ranaldo don’t venture into causation, making only passing references to the likelihood that the sophisticated cohort disguise their intentions by running multiple wallets and avoid crashes by causing them.

A section on wash trading notes its deleterious effect on market quality, but also finds that the whales fare no worse even after these bogus trades are excluded. The below chart shows an NFT portfolio picked for investor sophistication and ownership diversity (“agent-based variables”) does better than one picked on the traditional bubble signals of high volatility on low volume (“aggregate variables”).

Conclusions are twofold: market quality measures can “significantly predict bubble formation and price crashes”; and “sophisticated investors” use “superior information and skills” to avoid the damage.

How much these findings are shaped by the unique pointlessness and lawlessness of NFTs is impossible to guess, however. Weren’t all NFT markets a bubble, with duration the only variable? Is there really much to learn when the findings are applied to established, regulated markets in assets that have some real-world utility beyond money laundering?

The full paper is here; our comment box is below.

And in case you were wondering: yes, there are Pet Rock NFTs. Their peak was in August 2021, when one sold for $1.3mn and another was bought by crypto shillmaster Justin Sun for $500k. A fractional share of an original EtherRock can be bought today for $0.00002667, having lost 99 per cent of its value in a year. This, apparently, is what counts as “not inherently more irrational than traditional financial markets.”