Original title: "NFT Sales: Clearing the Market, Avoiding Gas Wars"
Original author: Scott Duke Kominers, Tim Roughgarden, a16z
Original translation: Lao Yapi
This article comes from WeChat official account: Lao Yapi
A key feature of Web3 is to enable early fans of creators, products or platforms to form a community through ownership of related tokens on the blockchain. When it comes to NFTs, they uniquely represent ownership of media such as code, digital art, characters, memes, music, text, videos, games, virtual real estate, etc., and creators often build important expectations and participation before their initial sales. This can foster community building among early collectors, which can sometimes in turn increase the value of holding a Token .
However, rather than letting the market determine the price of their primary sale, many NFT projects choose to initially sell their NFTs at prices below market clearing levels. Sometimes, this is because the team did not realize the huge demand for their product, or because their initial sale mechanism was poorly designed. Often, however, selling NFTs at prices below market clearing is motivated by a straightforward desire to make the sale accessible to more people, or to achieve a specific distribution goal.
But what happens when market designers trade efficiency for fairness (where “fairness” refers to the fairness of the market and the market clearing mechanism); or when demand far exceeds supply, as in the case of high-demand NFT launches and drops?
When people want to buy scarce resources in the market, this inevitably leads to competition among potential buyers in other ways. In the Ethereum ecosystem, this competition creates so-called “gas wars,” which escalate when would-be buyers bid up gas prices as they compete to be first in line for a sale. So how do developers avoid this? More broadly, if developers really want to offer NFTs at a price below market clearing, how do they meet market demand? Here, we offer some guidance to developers from the perspective of mechanism design in auction theory and practice—particularly in the unique context of blockchain.
But first, some context.
When there is a high-demand post or buy limit, buyer demand may far outstrip supply. Market pressures then drive prices toward a level where demand and supply match (or are equal), the “market clearing price.” This inevitable drive toward the price where supply and demand meet is one of the most core principles of economics. Even if “pricing” is somewhat hidden, e.g., competition over gas fees, ultimately market forces result in some form of payment that ultimately helps match supply and demand.
For example, imposing price controls often results in competition for opportunities — longer lines, more time spent searching for buying opportunities, or other costs. This has happened in non-blockchain contexts, with rent controls and (literal) gas price caps, both of which are well known.
Similarly, as mentioned above, Ethereum’s “gas wars” are a market equilibrium phenomenon. They are generated directly through market forces in response to constrained supply. (“Gas” here refers to the amount of computation required to perform a specific action on the network). More specifically, gas wars occur during NFT mints, when buyers try to jump ahead of each other in the queue, which causes a sudden surge in direct demand for block space; this in turn drives gas prices up and leads to increasingly higher transaction fees.
A common misconception that often comes up in discussions about avoiding gas wars is that the smart contracts that run sales need to be optimized. While it is certainly good to do so for other reasons, such as user accessibility, more gas-efficient smart contracts do not in themselves prevent gas wars, as gas wars are generated through market forces. And paradoxically, if the contracts are better optimized, gas war prices will need to be higher to achieve the same effective price. (Though the optimized contracts at least have more sales per block, which makes the entire process – including gas wars – end faster).
Ethereum co-founder Vitalik Buterin also pointed this out last year in a post about market clearing prices in the context of falling NFTs (and the ICO phenomenon before it), and the negative externalities that arise from this. In the post, Buterin also proposed alternatives to selling below the market clearing price, whether the purpose of doing so is to achieve fairness, community sentiment, or fun.
Relatedly: one of the authors of this post (Scott) also recently published work with Mohammad Akbarpour and Piotr Dworczak on "Redistribution Mechanisms," a theory of how market designers think about the tradeoffs between efficiency and "fairness" or redistribution problems, and how the same approach to designing revenue-maximizing auctions and efficient allocation mechanisms can be applied when there are market-level redistribution goals. In a non-blockchain context, this work has implications for policy issues such as in-kind transfer programs for food and healthcare, and vaccine distribution; in crypto, it applies to organizing NFT sales, DAO Token governance allocations, and more.
Now, with that general context, let’s talk more specifically about the tradeoffs between efficiency and other goals in market design for NFT sales. First, we’ll talk about auction designs that rely on market forces to efficiently clear the market. Then, we’ll talk about what to do when an NFT project or creator still wants to conduct a primary sale below the market clearing price, and how they can manage demand in that situation.
Perhaps the simplest strategy to avoid gas wars is to rely on market forces to conduct NFT sales in a way that clears the market. For example, here are two auction mechanisms that can achieve this goal and are currently used in NFT sales.
One option is to conduct a descending price auction, where the sale starts at a very high price and then slowly decreases, with bidders claiming units at the current price until they exhaust the available supply. This "pay as you bid" auction format is easy to understand, as bidders pay exactly what they bid for each unit. This also mitigates gas wars somewhat, as it spreads out the bids over time, although we sometimes see gas wars at the end of an auction as bidders try to buy in at whatever lowest possible realized price.
But there is an interesting incentive problem in this type of auction, in that participants have an incentive to wait if they believe that others are unlikely to bid, because bidding later, all else being equal, lowers the price you ultimately pay. Waiting tends to turn descending price auctions into something like a game of "chicken," where everyone waits to see when others will start bidding. This can paradoxically lower the clearing price (and lead to a gas war when everyone ends up bidding at the same time).
People also often point out fairness issues with this type of auction: the biggest fans and supporters of a project are the ones who are most motivated to lock in their bids at high prices early on, meaning they may end up paying more than others. While it may make sense from an economic perspective for those who place the highest value on the Token to pay a premium to ensure they win the Token , it may also be counter-intuitive to the spirit of the community to have the most influential backers pay, for example, 10 times the final sale price of the auction.
The fairness issues and gas wars that occur at the end of descending pay-per-unit auctions can be avoided if the team adopts so-called “clearing price” auctions (or “uniform price” auctions). In this auction mechanism, the price starts again from a high and falls, and people submit bids for units at any price along the way. The auction ends when the last unit is claimed, but unlike a “pay-as-bid” auction, here everyone pays the same price per unit at the end: specifically, the “settlement price”, which is the last (and lowest) price displayed before the auction terminates.
Note that in a clearing price auction, while no one pays more than their respective bid, if people bid for units when the price was high but the final price is low, then in principle people could end up paying much less than their bid. So, for example, if a clearing price auction starts accepting bids at 2ETH but eventually exhausts the supply at 0.42ETH, all bidders end up paying 0.42 per unit, even those who bid, for example, an initial price of 2. This makes the process easier for bidders, because if someone's value is 2, they can immediately bid at 2 without worrying about getting a worse deal relative to if they waited and entered a bid at a lower price. (This isn't entirely true when you're the only one bidding very close to the clearing price, but in oversubscribed NFT auctions this isn't usually a problem).
From a market design perspective, clearing price auctions make sense in many ways. In particular, they encourage economically efficient outcomes, especially when there are a reasonable number of units being sold. It’s important to note, however, that running them on a blockchain introduces some complications, as bidders must effectively hold funds in escrow while bidding, and then receive a refund once the final price is determined. The refund function must be carefully specified, for example, to avoid accidentally locking additional funds into the sale contract.
If a team really wants to sell at a price below the market clearing level, then the way to avoid a gas war is to constrain demand in some way, preventing buyers from having to compete with each other in order to buy.
But this is difficult to do, as it usually requires limiting who is allowed to buy, and how many units these potential buyers are allowed to buy. The total demand from all eligible buyers needs to be less than or equal to the available supply. This can be particularly difficult to ensure in the cryptocurrency space, where it is common to maintain multiple pseudonymous accounts.
Still, there are options for limiting participation, and we outline three below. Some projects have used these techniques successfully. (Note, however, that while we include a section on lotteries, purely for educational purposes, there are some questions about the legality of lotteries and “sweepstakes” as a mechanism for narrowing the pool of participants in a sale; so builders should consult an attorney before using such a mechanism.)
One way to limit participation is to require some form of “personal identification” as a condition of participating in the sale—pre-registering buyer accounts to guarantee that each account represents a unique person. This limits the pool of buyers (or, more specifically, buyer wallets) to those who can prove unique identity, perhaps through a “know your customer” (KYC) policy requiring the upload of identity documents. If implemented well, this can screen out botnets and greatly reduce the number of potential buyers.
However, proof of unique personhood alone is not always sufficient to achieve market stability and fairness. In practice, people often manage to get around implicit purchase limits by hiring people to KYC additional wallets. But even if it were possible to truly restrict sales to unique buyers, this would still not solve the problem if demand simply far outstrips supply.
As we highlighted above, market forces are likely to drive prices up as long as demand outstrips supply. There is not a huge difference between an infinite number of bots ready to buy and just 20% of registered demand outstripping supply, and in both cases the incentive to be first will result in a gas war that drives the total price paid per unit toward the market settlement level. This is often clear to experienced NFT buyers before a sale is made, and blockchain detectives often carefully track the total number of approved wallets and keep this in mind.
Another way to limit participation is to create “whitelists” — explicitly curated lists of potential buyers. Many teams offer access to token holders associated with partner projects, or to those who are particularly active in the community before launch. This approach does have the potential to limit demand below supply, completely alleviating the problem of gas wars, at least as long as project creators don’t go too far and whitelist too many people. However, this approach does not escape the problem of market pricing. In fact, the price competition here just happens in a different unit. Potential buyers don’t just pay with currency, but pay part of the fee through work, that is, the effort they put in to hope to get a place on the list. This means that the process of getting whitelisted can become a grind. However, whitelists do create opportunities for more incentives, in the sense that the activity carried out by potential members of the NFT community increases the participation of the entire community, and ultimately, the opportunity to buy NFTs goes to those who are the most active participants. There is also the case that whitelists are designed in a way that allows access to those who have the time and energy to spend, but less liquidity. At the same time, however, the race to the whitelist is often “all pays” rather than “winner pays”. Unlike the auctions above, where only the winner ends up paying any price, in the context of whitelists, everyone who participates ends up paying for their efforts (or receiving an NFT from a partner project), regardless of whether they ultimately receive the right to purchase the NFT. This can lead to negative feelings from those who worked hard but were left out, which can be tricky for the NFT community.
Another way to limit participation (and potentially avoid gas wars) is to randomly assign buying opportunities via a lottery. Here, potential buyers have the opportunity to register for a lottery over a long period of time (perhaps a week), and then at the end, a randomly selected group of registrants are given the right to buy the NFT. (Some of the selected registrants may opt out, meaning that whitelisting may subsequently be required).
This type of mechanism effectively mitigates gas wars by first decentralizing the registration process (so there is no need for all registration transactions to occur at the same time) and then by limiting the number of buyers to exactly match the available supply.
However, like whitelisting, market-clearing pricing again works through a form of work: in this case, the effort put into participating in the lottery.
In a lottery for an anticipated sale in high demand, potential buyers may create a large network of bot registrations. In this case, the lottery may devolve into a competition between powerful bot networks, and the average individual buyer may not actually be able to get any position in the sale.
Participation in lotteries could be more effectively controlled by using unique individual identification strategies (again, like KYC). Alternatively, the ability to create absurdly high numbers of participants could be reduced by requiring people to escrow a deposit for each entry into the lottery. The escrow approach makes it expensive to submit a large number of entries, but at the same time, buyers with excess liquidity will be more able to enter than others (and still likely to enter multiple times), limiting the overall accessibility of the sale.
It is worth noting that many of the mechanisms described above can be chained together for a single sale.
For example, Tally Labs’ Jenkins’ Valet Bored & Dangerous mint was run in three stages. First, the team held a descending liquidation auction for 2,367 units; then, they held a whitelist sale for various members of the existing community at a price lower than the final auction clearing price. Finally, holders of the team’s creation of a series of “Writers Room” NFTs were able to receive an additional copy of the Bored & Dangerous NFT for free (one per Writers Room NFT).
Market clearing prices are like gravity: inevitable, inescapable. Anyone who builds a market mechanism (whether on a blockchain or not) must deal with the forces of balancing supply and demand.
But at the same time, thinking about market forces and incentive design can help us build the right mechanisms and shape the ultimate equilibrium point for sales. We have seen that a top-level decision is whether to rely on market forces (in which case the clearing price auction provides a good starting point for sales); or to stick with the more challenging approach of selling below the market clearing price (in which case controlling for eligible demand, such as through whitelisting, is necessary but may not be sufficient).
The design of NFT distribution provides another example of how difficult mechanism design on blockchains can be… As NFT markets develop, we expect to see new mechanisms that are both guided by and contribute to classical theories of market design. However, as more researchers and builders embed classical mechanism design and the idiosyncrasies of Web3, we are starting to see a lot of experimentation around various blockchain-based distribution mechanisms. Progress is inevitable… We are excited about what comes next
Acknowledgements The author would like to thank Chain Runners Architects, Chris Dixon, Far, Flashrekt, Foobar, Eddy Lazzarin, SAFA, Steve Kaczynski, Valet Jones, and Vivek Ravishanker for helpful comments. Special thanks to our editor Sonal Chokshi!
Original link
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Official Twitter Account: https://twitter.com/BlockBeatsAsia