Non-Fungible Tokens, Explained

“Cambrian explosion” of disparate currencies
Most of these early blockchain innovators just took the original cryptocurrency’s source code, made their preferred changes, and launched their alternative versions as distinct cryptocurrencies; it was as if they’d modified the DNA of an existing species to create a new, reproductively isolated branch of the family tree.


What does non-fungible mean?

To fully appreciate what makes these tokens special, it’s worth taking a look at the difference between “fungible” and “non-fungible.”

When something is fungible, in this case a token, it means it can be easily replaced by something identical – and it is interchangeable with ease.

Real world examples of something fungible could include grains of rice, or the $1 bank note in your pocket. If you were to lend that $1 to someone, it wouldn’t matter if they didn’t return the exact same one.

This all changes when something is non-fungible. Although two items may look to be identical at a glance, each will have unique information or attributes that make them irreplaceable or impossible to swap.

One physical example of a non-fungible asset could be a plane ticket. Sure, they look the same as other tickets, but each one has different passenger names, destinations, departure times and seat numbers. Exchanging yours with someone could have serious consequences – not only could you end up thousands of miles away from where you wanted to be, airport security might not be too impressed either.

Why are non-fungible tokens different from other tokens?

Because they can offer unique characteristics which make them different and digitally scarce.

Many tokens – and indeed cryptocurrencies – are fungible. If you send someone a Bitcoin, and get one back, you wouldn’t notice any difference.

A lot of the time, fungible tokens are built using a standard called ERC-20. For the sake of simplicity, let’s imagine each of these tokens is a $10 bill. If you sent a token to someone, and got another one back a week later, they would be identical. (That said, there might be some fluctuation in price.)

This all changes with non-fungible tokens, many of which are ERC-721 compliant. These can be compared to baseball cards, as each has unique information and varying levels of rarity. If you were to accidentally send one of these tokens to someone, and get a different ERC-721 token back, you might be very upset.

There’s one more crucial difference you need to bear in mind. Fungible tokens are divisible – meaning you can send a fraction of one ERC-20 token. (Like cash, where you can pay with a $10 bill and get change.) On the other hand, non-fungible ERC-721 tokens cannot be divided and must be bought or sold whole. (Like baseball cards, where no one in their right mind would want to buy half.)

What can non-fungible tokens be used for?

Collectibles are a common use for non-fungible tokens.

One early pioneer of non-fungible tokens was CryptoKitties, a blockchain-driven platform where players have the chance to collect and breed digital cats.

If you’ve ever owned a cat, you’ll know that they are not easy to replace, as their appearance and personality make them unique. In this case, CryptoKitties replicated this concept in the crypto world – with each cat’s digital genetic material being stored on the blockchain. They can be bought and sold using Ethereum, and some are rarer than others. Indeed, as reported by Cointelegraph, sales hit $12 million last year – with the most expensive CryptoKitty reportedly going for $120,000.

Other games have promptly followed, such as fantasy titles where fighters are collected for battles. And in another apt development (almost like we planned it,) Major League Baseball in the US is planning to launch a game where baseball cards can be exchanged on blockchain.

What are their advantages and disadvantages?

Although the non-fungible ERC-721 token has offered improvements on ERC-20, there are a few setbacks.

As we recently explained in an article looking at alternatives to ERC-20, advocates of this token protocol believe it could become “the ultimate vehicle for putting every significant asset on a public or hybrid blockchain with 100 percent immutability and security.”

Non-fungible tokens allow you to detail more of the attributes that make them special – far beyond the name, balance, token supply and symbol. This is because you can include rich metadata about an asset and include information about ownership – and these authenticated details can ultimately add value because investors can be confident about its provenance.

That said, there are downsides. Non-fungible tokens have not been embraced as speedily as some advocates hoped, in part because the ERC-721 protocol is so new. It can also be tricky and time consuming to develop decentralized applications for non-fungible tokens properly.

So why not carry on using real world solutions?

The world is moving on from paper – and supporters believe there are many use cases for non-fungible tokens beyond collectibles.

For example, it could become a secure and immutable way of storing birth certificates, academic credentials, warranties and identities – even artwork and property ownership.

These real world assets can then be properly digitized and stored in a wallet, keeping them safe and ensuring that they cannot be altered or counterfeited by a third party.

But how are non-fungible tokens created?

There are concerns that the use of non-fungible tokens could end up becoming fragmented, with different standards and varying degrees of certification.

At the moment, as we mentioned a little earlier, it can be an expensive and complicated thing to achieve. It can sometimes take months to develop a DApp, and in the fast-moving blockchain world, that could cost an entrepreneur their competitive advantage.

Instead of every outlet interested in non-fungible tokens creating their own framework (which would result in a breeding ground for inconsistency,) some platforms are trying to create a technological layer which unifies and standardizes these tokens.

One of them is 0xcert, which offers a “plug and play” framework which means that a non-fungible token can be developed and verified quicker – and the platform claims it is achievable in days. The company says it eliminates the need for in-depth blockchain knowledge, while stopping precious data from being siloed and preventing developers from having a lengthy, expensive and insecure process.


Consumers might be the biggest winners when crypto tokens are designed correctly.

At least that’s the case made by a new paper by MIT’s Christian Catalini and the University of Toronto’s Joshua Gans, which describes a simplified model that illustrates what might be a valuable price discovery role that utility tokens, or those that operate as true commodities in the spirit of bitcoin and ether, might enable.

Not only that but the paper, called “Initial Coin Offerings and the Value of Crypto Tokens,” goes so far as to predict a world where tokens empower consumers to choose an optimal price for a service collectively.

From the introduction:

“This paper provides the first economic analysis of the ICO funding mechanism and how it relates to traditional equity financing.”

It goes on to tease out intriguing benefits that might accrue to society through the sale of and trade in utility tokens, one of the most controversial topics in crypto today, suggesting that risks could be balanced by returns if entrepreneurs are permitted to try the token fundraising model so the public can see how it plays out.

Most people looking at the initial coin offering (ICO) trend see extraordinary sums of money coming in — $8.84 billion as of February, according to the CoinDesk ICO tracker, and no doubt that’s why U.S. regulators, such as the Securities and Exchange Commission (SEC) have raised so many questions about this new industry.

“The problem the regulators have is they don’t know what the goals are,” Gans told CoinDesk in a phone call. “Instead the regulators are coming in saying ‘I don’t really know how the market should be working, but it smells terrible.'”

What tokens do

In this way, the paper was meant to begin a conversation about the right way to think about tokens so that societies could rationally consider the correct approach to managing them.

“You have to create the economic theory to understand what’s going on here to even know what category of regulation to choose,” Gans said.

“The ICO mechanism allows entrepreneurs to generate buyer competition for the token, which, in turn, reveals consumer value without the entrepreneurs having to know, ex ante [based on forecasts], consumer willingness to pay,” the paper argues.

Gans said:

“We were trying, with our paper, to ask why would these tokens be of any value.”

The paper plays out a simple model of a company offering a token as the sole means of paying for some new technology platform (like a video site or malware detector), with a founder who has no intention of cheating or bending on commitments. The idea is that the issuer can have some idea of how much it would cost to build whatever it wants to build, but the firm can’t reliably know how the public will price it.

The hard thing about technology projects is that they often have very high upfront costs to write the code, test it and get it working (fixed costs). On the other hand, once it’s built, each new user usually doesn’t cost the company much (the marginal cost).

Gans believes that entrepreneurs are thinking about costs a lot more than demand when they run token sales.

“The real value of the tokens has really nothing to do with the amount of money you want to spend, but with how much money people want to spend with you,” Gans said.

That said, Gans argued that it didn’t matter if a token project gets specific early on about pricing new services in terms of its token. For example, if Netflix launched today with an ICO, it might say in its white paper that one NFLX token would be good for one month of streaming video.

That level of specificity isn’t necessary before a project gets started, Gans said, but it might help if the company spelled out some of its thinking about how many users it might get over time and what factors might influence it growing or declining.

Such projections would give the market more insight to price the products accurately before they launch, including future gains.

“Maybe people who are doing these white papers could provide a more clear path to whatever they are doing,” Gans said.

On the other hand, he added:

“I don’t think for a moment that anyone who’s buying these tokens has worked all that out.”

Founders mustn’t

One caveat: it only works if founders don’t cheat, and there are two key cheats that consumers and speculators both need to worry about.

Acknowledging the fact that no one has had a chance to see an ICO roll from a token sale to launch and product-market fit, theoretically, this basic model works fine, assuming a good faith founder. Of course, the biggest danger is a founder who presents a compelling vision, runs a token scale and absconds with the earnings.

And here Gans welcomes the present scrutiny by regulators. He said, “Their instinct to stop people making false promises is of course correct.”

But what if the issuer breaks the commitment to only accept the token as payment on the platform. If they did that, the price of tokens has the potential to completely collapse.

For another, the issuer could also just issue more tokens. Then all the tokens in the world would drop proportionally in value.

Most issuers hold tokens, supposedly to align their interests with their users, but this is really no protection because the issuer will still come out ahead issuing new tokens, even if their reserve loses value. As Gans said, this is why some countries get into out of control inflation situations, because every time a treasury prints more money, the state comes out ahead (until it all completely unravels).

That said, this danger is only contingent on projects that announce a fixed supply of tokens that will never change again after the token generation event (as most ICO projects do today).

“What if during that growth cycle you realize you don’t have enough money and you need more?” Gans asked. “You’re going to be a bit stuck.”

He added:

“I wonder if they are tying their hands way too much?”

While Gans and Catalini identify these sorts of dangers, their paper doesn’t suggest solutions. The idea here is to begin laying out this basic model to facilitate further research and discussion.

“This is a very simplified model,” Gans said of their paper. “I don’t know how everything’s going to work out.”