Disclaimer: This article is part of the Industry 4.0 Open Educational Resources (OER) Publication Initiatives jointly supported by Duke Learning Innovation Center and DKU Center for Teaching and Learning under the Carrying the Innovation Forward program. This article belongs to the OER Series No. 2 Computational Economics Spring 2022 collection. The Spring 2022 collection is partly supported by the Social Science Divisional Chair’s Discretionary Fund to encourage faculty engagement in undergraduate research and enhance student-faculty scholarly interactions outside of the classroom. The division chair is Prof. Keping Wu, Associate Professor of Anthropology at Duke Kunshan University. The co-author Tianyu Wu was the Teaching and Research Assistant for Prof. Luyao Zhang in the course: COMPSCI/ECON 206 Computational Microeconomics at Duke Kunshan University Spring 2022, when he completed the joint article. The co-authors are forever indebted to Prof. Vincent Conitzer, who presented “Computer Science Meets Economics” as a distinguished guest lecture for this course on Apr. 19, 2022. Tianyu Wu thanks Prof. Fan Zhang’s insightful comments on this article during his summer experiential learning activities jointly supervised by Prof. Luyao Zhang and Prof. Fan Zhang in their project entitled “Understand Waiting Time in Transaction Fee Mechanisms,” supported by Ethereum Foundation.
Introduction
An auction is a transaction in which products or services are offered for bidding (Myerson, 1981). In the auction market, potential clients place competitive bids according to the value of their willingness to pay for the items, while sellers submit competitive offers to indicate the lowest price that he/she is willing to accept. As an important component of market design, auction theory has been well-studied by hundreds of Economists in the past decades. Among them, two distinguished scholars, Dr. Paul Milgrom and Dr. Robert Wilson were awarded to appreciate their contributions to improving auction theory in 2020 (Nobelprize.org 2020).
At the same time, blockchain, a decentralized technology, made by a Peer-to-Peer network of nodes, is becoming more and more popular right now. It prompts us to seek more opportunities between blockchain and auction theories or models from the perspectives of research and innovation. For one thing, auction design in the traditional market tends to be constrained by the inefficiency and inflexibility of centralized systems, thus many advanced auction theories cannot be genuinely implemented in the current market. But it is expected that the blockchain tends to enable more advanced auction designs to be implemented in online markets (Shi et al. 2021). Taking the Ethereum blockchain as an example, its variant block generation time makes the idea of candle auction design, where the end signal is not necessarily predictable, possible. For another, market designers can make full use of auction theories or models to manage the dynamic relationships in the blockchain world (Ai, Liu, and Wang 2020; Gao et al. 2019). For example, the blockchain's transaction fee [1] market can be modeled based on the auction theory (Chen, Chen, and Lin 2018).
In this article, we will further investigate and elaborate on how blockchain meets auction at the interface of computer science and economics. First of all, we will focus on blockchain technology and introduce how auction-based mechanism design [2], such as transaction fee mechanism (TFM) [3], block reward [4] distribution, etc. can empower the consensus layer [5] of the blockchain, and how they are and expected to be applied to layer 2, the scaling solutions for blockchain. Then we will make a comprehensive analysis of auction models in categories. Last but not least, we will look into the relationships between these two and investigate how those two concepts are well integrated and empower each other. Figure 1 summarizes the content of the article.
Figure 1 Content Summary
I. Blockchain Technology
1.1. Layer 1
Figure 2: Incorporating Auction Mechanisms across Different Blockchain Layers (Shi et al. 2021)
The common architecture of a generic blockchain includes five layers: Hardware, Data, Network, Consensus, and Application Layer (Shi et al. 2021). Figure 2 demonstrates examples of different blockchain layers. In this section, we mainly focus on the Consensus Layer, which typically contains an incentive mechanism to motivate participation in validating the block records, especially under permissionless blockchains. Typically, the main components of incentive mechanisms include transaction fees and block rewards. Currently, several incentive mechanisms based on auction theory have been used to help with the development of blockchain (Kolb et al. 2020), among which one of the most important application scenarios is its application in the design of Transaction Fee Mechanism (TFM).
Taking the most popular permissionless blockchain, Ethereum, for example, its legacy TFM adopts the first-price sealed-bid auction model: users submit a fee quotation on the premise of mutual confidentiality of their bids, and miners preferentially pack transactions with higher bids into blocks to obtain transaction fees (Zhang, Li, and Wang 2017). Under the assumptions of traditional economics, the sealed first-price sealed-bid auction model can achieve Pareto optimality (i.e., high transaction efficiency). However, Ethereum, in reality, is beyond the assumptions of traditional economics, and the adoption of sealed first-price sealed-bid auction leads to information asymmetry, and it further 1) results in the ineffective allocation of resources and low transaction efficiency; 2) makes the bidding strategy more complex and increases the bidding cost of users. To deal with this situation, a new protocol, EIP-1559 started to be applied after the London hardfork, to mainly deal with the problem of information asymmetry in the Ethereum blockchain, to further optimize resource allocation. The proposal introduces a fixed fee to Ethereum that must be paid per transaction, known as the Base Fee, whose operating mechanism is well specified in the protocol. Specifically, the Base Fee fluctuates up and down based on the previous block, with the goal of an average block utilization of 50%. The fixed fee increases proportionally (capped at +12.5% per block) when the usage of the previous block exceeds 50% and decreases when the usage falls below 50%. We can see that because the Base Fee is embedded in the protocol and is automatically adjusted according to the status quo of the block, which can be perceived as a price reference standard introduced to the original auction mechanism: if the user's reserve price is lower than the Base Fee, Then he will choose not to bid and wait until the future Base Fee is lower than his reserve price before bidding; if the user's reserve price is higher than Base, he will choose to bid higher than Base Fee. Therefore, EIP-1559 provides traders with a predictable transaction quotation by moving the Base Fee into the protocol, reducing the asymmetry of information, and to a certain extent alleviating the problems of inefficient resource allocation and complex quotation strategies. According to Roughgarden (2020), it can also be perceived as splitting the transaction fee into two parts: one is a base fee, a bottom line for the buyers to enter the auction and automatically burned once the transaction is included in the blockchain, and the other is a priority fee, similar to the tip which indicates the buyers’ willingness-to-pay for the opportunity costs [6] that are spent to enable a certain transaction to be included in the blockchain. Meanwhile, we also notice that some alternative auction models, especially the Generalized Second-Price (GSP) auction are used to investigate how it will empower and improve the efficiency of the blockchain transaction fee market (Basu et al. 2019; Yan et al. 2020). Last but not least, aside from the application in transaction fee mechanism, the auction design has also been heavily discussed in the block reward distribution, another important factor that impacts miners’ decisions in the ecosystem, which is subject to different types of the blockchain (Zhu et al. 2018).
1.2. Layer 2
Layer 2 scalable solutions functioned as processing new transactions faster while reducing the load on Layer 1 and typically taking much lower fees (ethereum.org). It enables market designers to implement more advanced auction models to improve their efficiency and avoids market congestion. There are several types of auctions that are popular and widely known in layer 2 as of now: Non-Fungible Tokens (NFT) [7] Auction, Decentralized Exchanges (DEX) [8] double auction, and Initial Coin Offering (ICO) [9] Auction.
A. NFT Auction
The bidding and buying process for an NFT is the same as most other lots. Some auctions may be online-only, whereas others may end in a live auction (Fazli, Owfi, and Taesiri 2021). However, the winning bidder must take possession of the NFT on the blockchain where it resides digitally without any excuses.
For the transaction to be held at the NFT Auction, the seller typically starts by setting a low price and a deadline. Bids for the NFTs can be submitted by interested buyers, and the NFT can then sell the property to the highest bidder after the designated time period has passed. Participants in NFT auctions may be from all over the world, and they may be in different time zones. As a result, this auction-style method ensures that all potential bidders have a fair chance (Arditi et al., n.d.).
Some common NFT auction marketplaces in the decentralized world include OpenSea, Axie, Rariable, etc. Taking OpenSea, the NFT market leader, for example, it has a wide range of digital materials available on its platform, and supports more than 150 different tokens for payment. People can sign up and browse those broad choices with a completely free account. It is also quite simple and convenient for artists and creators to create (known as “mint”) their own NFT for sale in OpenSea. Moreover, auctioning an item in OpenSea is also extremely convenient, and it also supports selling a bundle of NFTs with simple clicks on your account. Figure 3 illustrates how to create and sell the NFTs in OpenSea. More information can be found on the website.
Figure 3: How to create and sell the NFTs in OpenSea
B. DEX Double Auction
According to Malamud et al. (2017), agents in the exchanges can submit their long or short prompts by combining limit and market orders, where the whole process follows the uniform-price double auction trading. Traditionally speaking, uniform-price market clearing is almost comparable to the general one in centralized markets from a theoretical point of view. Typically, in blockchain layer 2, the decentralized exchanges (DEX) will commonly adopt this combinatorial double auction framework to let participants introduce an auction bid so that a genetic algorithm can be designed to find a solution to optimize the pairing in a reasonable amount of time (Lee et al. 2021).
C. ICO Auction
Initial Coin Offering (ICO) is another essential achievement of the ecosystem in the blockchain. ICOs enable projects based on blockchain to accumulate the flow of capital and avoid the more traditional financing methods. They also enable clients to become investors and for investors keen on cryptocurrencies, there are a number of factors that must combine to make ICOs attractive. Auction models are well-established and integrated into ICOs. There are three common types: capped sales, uncapped sales, and the dutch auction (CoinCodex 2017).
II. Auction Models
In this section, we classify the existing important auction models into the following four categories based on the bidding process, the number of items, bidding participants, and other more advanced or complex models (Klemperer 1999; Klemperer 2004; Krishna 2009).
2.1. Open-outcry Auction
There are two types of open-outcry auction mechanisms, where one is ascending price and the other is descending price (Wahaballa et al. 2015). The former is also known as English Auction, and the latter is named as Dutch Auction, both of which require bids and offers to be made in the open market without any excuses.
2.2. Sealed-bid Auction
Similar to the open–outcry auction, there are also two types of sealed-bid auctions, i.e., First-Price and Second-Price auctions, where the latter is also named Vickrey Auction. Both of them are well-considered in building the auction design in the marketplace (Galal and Youssef 2018). Figure 4 illustrates the process of sealed-bid auctions.
Figure 4: Sealed-bid auction
2.2. Pay-as-Bid Auction.
Besides the two introduced above, there is one more auction model which requires all the participants to pay-as-bid in the process by reporting their willingness to pay for each unit of items. Compared to the traditional first-price auction, extend the rules of the well-known first-price auction to the sale of multiple units of the same good. (Wittwer 2018).
2.3. Auction with the Time Dimension.
From Economists’ point of view, time is another format of wealth since rational people always take the opportunity cost, that is, the potential loss from a second maximal opportunity, into consideration. In the traditional auction market, it is difficult to take the effect of time into consideration when it comes to bidding, thus, the candle auction, adding the time dimension to the original auction models has not been discussed a lot due to this limitation (Patten 1970). However, the Ethereum Virtual Machine (EVM) can be perceived as a Markov process with states updated by blocks, thus this provides us with a benchmark of the eligibility that the candle auction can be applied in the Ethereum blockchain.
Last but not least, there are several more advanced or complex models that are available, for example, multi-attribute auctions and sponsor search auctions. They tend to rely on a higher-standard scenario to be well integrated. Figure 5 shows examples of several advanced auctions.
Figure 5: Advanced Auctions Models
III. Connect Blockchain to Auction
There is a ton of space for interdisciplinary research and innovation between blockchain and auction systems because of the ways in which they complement one another. On the one hand, the decentralized character of blockchains can offer a dependable, safe, and affordable method for administering auctions; On the other, blockchain development, especially from the perspective of economic incentives in the consensus protocols, indeed requires the development and application of advanced auction models. In the following part, we will look at some of the current approaches for fusing the auction models and blockchain, as well as some specific and realistic application scenarios and taxonomies based on current literature. These potentials have spurred a frenzy of research and development efforts in both academia and industry.
3.1. Blockchain-based Auction Model
Shi et al. 2021 introduce several widely used blockchain-based auction models in their survey papers. We notice that there are more auction models being developed based on the blockchain. Among them, some blockchain is now trying to make an integration of a two-stage and multi-attribute auction model to increase the trusted security between nodes (Devi, Rathee, and Saini 2020). In addition, some researchers set out to further make use of auctions to realize greener and more computational-friendly missions in decentralized energy trading (Hassan, Rehmani, Mubashir Husain, and Chen 2021).
3.2. Auction-enhanced Blockchain Technology
In addition, auctions can be applied to enhance blockchain technology to produce a more efficient, fair, and secure market environment.
A. Transaction Fee Mechanism
As we have mentioned in the previous paragraphs, the transaction fee mechanism, as one of the most important features of the incentive layer in blockchain, relies on a desirable auction model to incentivize the users to make transactions in the blockchain. In Bitcoin, we know that it is dominated by Priority Auction with VCG Mechanism, which means that each user is going to give a bid that meets one’s externalities. But in Ethereum, Layer 1 generalized and applied the GFP Auction before the London fork and now applies both EIP-1559 Auction and legacy auction methods for determining the gas prices. Figure 6 illustrates the Legacy (Left) and EIP-1559 (Right) auction.
Figure 6: Legacy and EIP-1559 Auction
B. Flashbots Auction
Confronted with the front-running behaviors in the Ethereum blockchain, many scholars try to find a solution to deal with those behaviors in order to add to the fairness and security of the chain. In the blockchain settings, it is also called Miner/Maximal-Extractable Value (MEV) (Daian et al. 2020). At this stage, flashbots auction fills the gap by using a dark venue, that is, a private channel between miner and user to enable the transaction only be seen between those two, free from potential front-running or back-running attacks, and in those dark venues, the auction also follows the first-price sealed-bid auction (Flashbots.net). It is expected to decrease the Ethereum network transaction fees thus further reducing Ethereum network congestion (Flashbots.net). Figure 7 illustrates the flashbots auction roadmap.
This article introduces how blockchain technology and auction models connect with each other. We expect the two to integrate further at the intersection of computer science and economics in theory and practice.
Annotations
[1] Transaction Fee
Transactions require a fee that users have to pay for the computation, and must be mined to become valid.
Mechanism design theory provides a coherent framework for analyzing this great variety of institutions, or “allocation mechanisms”, with a focus on the problems associated with incentives and private information. Nobel Prize - 2018 Economics
[3] Transaction Fee Mechanism (TFM)
A transaction fee mechanism (TFM) is a triple (x, p, q) in which x is a feasible allocation rule, p is a payment rule, and q is a burning rule.
This layer provides incentive mechanisms for a blockchain to motivate participants to validate the data and maintain the whole system. Incentive mechanisms are typically based on block rewards and transaction fees.
Non-fungible tokens (NFTs) are cryptographic assets on a blockchain with unique identification codes and metadata that distinguish them from each other. Unlike cryptocurrencies, they cannot be traded or exchanged at equivalency. This differs from fungible tokens like cryptocurrencies, which are identical to each other and, therefore, can serve as a medium for commercial transactions.
A decentralized exchange (or DEX) is a peer-to-peer marketplace where transactions occur directly between crypto traders. DEXs fulfill one of crypto’s core possibilities: fostering financial transactions that aren’t officiated by banks, brokers, or any other intermediary.
ICO refers to a formerly popular method of fundraising capital for early-stage cryptocurrency projects. In an ICO, a blockchain-based startup mints a certain quantity of its own native digital token and offers them to early investors, normally in exchange for other cryptocurrencies such as bitcoin or ether. As a type of digital crowdfunding, ICOs enable startups not only to raise funds without giving up equity but also to establish a community of incentivized users who want the project to succeed so their presale tokens rise in value. Coindesk
[10] Internet of Vehicles (IoV)
Internet of vehicles (IoV) is a network of vehicles equipped with sensors, software, and technologies that mediate between these with the aim of connecting and exchanging data over the Internet according to agreed standards.
Ai, Zhengpeng, Yuan Liu, and Xingwei Wang. 2020. “ABC: An Auction-Based Blockchain Consensus-Incentive Mechanism.” In , 609–16. IEEE. https://doi.org/10.1109/ICPADS51040.2020.00085.
Arditi, Andy, Pranav Garimidi, Dean Hirsch, and Iason Milionis. n.d. “COMS 6998-006 Foundations of Blockchains an Initial Framework for NFT Auction Mechanism Design: Impossibility Results and Solutions.” https://timroughgarden.github.io/fob21/reports/r2.pdf.
Basu, Soumya, David Easley, Maureen O’Hara, and Sirer, Emin Gün. 2019. “Towards a Functional Fee Market for Cryptocurrencies.” ArXiv:1901.06830 [Cs]. January 2019. http://arxiv.org/abs/1901.06830.
Chen, Yi-Hui, Shih-Hsin Chen, and Iuon-Chang Lin. 2018. “Blockchain Based Smart Contract for Bidding System.” In , 208–11. https://doi.org/10.1109/ICASI.2018.8394569.
Conitzer, Vincent, Tuomas Sandholm, and Paolo Santi. 2005. “Combinatorial Auctions with K-Wise Dependent Valuations.” https://users.cs.duke.edu/~conitzer/kwiseAAAI05.pdf.
Daian, Philip, Steven Goldfeder, Tyler Kell, Yunqi Li, Xueyuan Zhao, Iddo Bentov, Lorenz Breidenbach, and Ari Juels. 2020. “Flash Boys 2.0: Frontrunning in Decentralized Exchanges, Miner Extractable Value, and Consensus Instability.” In , 910–27. IEEE. https://doi.org/10.1109/SP40000.2020.00040.
Devi, Anju, Geetanjali Rathee, and Hemraj Saini. 2020. “Using Optimization and Auction Approach: Security Provided to Vehicle Network through Blockchain Technology.” In , 476–80. IEEE. https://doi.org/10.1109/PDGC50313.2020.9315804.
Fazli, MohammadAmin, Ali Owfi, and Mohammad Reza Taesiri. 2021. “Under the Skin of Foundation NFT Auctions.” ArXiv:2109.12321 [Cs], September. https://arxiv.org/abs/2109.12321.
Galal, Hisham, and Amr Youssef. 2018. “Verifiable Sealed-Bid Auction on the Ethereum Blockchain.” In . https://www.researchgate.net/publication/323976298_Verifiable_Sealed-Bid_Auction_on_the_Ethereum_Blockchain.
Gao, Zhipeng, Bang Lin, Kaile Xiao, Qian Wang, Zijia Mo, and Yang Yang. 2019. “A Dynamic Resource Allocation Algorithm Based on Auction Model in Mobile Blockchain Network.” In , 1605–10. IEEE. https://doi.org/10.1109/EITCE47263.2019.9095058.
Hassan, Muneeb Ul, Rehmani, Mubashir Husain, and Jinjun Chen. 2021. “Optimizing Blockchain Based Smart Grid Auctions: A Green Revolution.” https://doi.org/10.48550/ARXIV.2102.02583.
Klemperer, Paul. 1999. “Auction Theory: A Guide to the Literature.” Journal of Economic Surveys 13 (July): 227–86. https://doi.org/10.1111/1467-6419.00083.
———. 2004. “Auctions: Theory and Practice.” ID 491563. Social Science Research Network. https://papers.ssrn.com/abstract=491563.
Kolb, John, Moustafa AbdelBaky, Randy H. Katz, and David E. Culler. 2020. “Core Concepts, Challenges, and Future Directions in Blockchain.” ACM Computing Surveys (CSUR) 53 (1): 1–39. https://doi.org/10.1145/3366370.
Krishna, Vijay. 2009. Auction Theory. Google Books. Academic Press. https://books.google.com/books?hl=zh-CN&lr=&id=qW1128ktG1gC&oi=fnd&pg=PP1&dq=V.+Krishna.
Lee, Jung-San, Chit-Jie Chew, Ying-Chin Chen, and Kuo-Jui Wei. 2021. “Preserving Liberty and Fairness in Combinatorial Double Auction Games Based on Blockchain.” IEEE Systems Journal 15 (September): 3517–27. https://doi.org/10.1109/JSYST.2020.3027948.
Li, Honglei, and Weilian Xue. 2021. “A Blockchain-Based Sealed-Bid E-Auction Scheme with Smart Contract and Zero-Knowledge Proof.” Security and Communication Networks 2021 (May): e5523394. https://doi.org/10.1155/2021/5523394.
Liu, Keyang, and Yukio Ohsawa. 2019. “Auction Based Rewards Distribution Method in Pool Mining.” In , 103–10. Association for Computing Machinery. https://doi.org/10.1145/3343147.3343162.
Omar, Ilhaam A, Haya R Hasan, Raja Jayaraman, Khaled Salah, and Mohammed Omar. 2021. “Implementing Decentralized Auctions Using Blockchain Smart Contracts.” Technological Forecasting and Social Change 168 (July): 120786. https://doi.org/10.1016/j.techfore.2021.120786.
Patten, R. W. 1970. “Tatworth Candle Auction.” Folklore 81 (2): 132–35. https://doi.org/10.1080/0015587x.1970.9716671.
Roughgarden, Tim. 2020. “Transaction Fee Mechanism Design for the Ethereum Blockchain: An Economic Analysis of EIP-1559.” https://timroughgarden.org/papers/eip1559.pdf.
Shi, Zeshun, de Laat, Cees, Paola Grosso, and Zhiming Zhao. 2021. “When Blockchain Meets Auction Models: A Survey, Some Applications, and Challenges.” ArXiv:2110.12534 [Cs]. October 2021. http://arxiv.org/abs/2110.12534.
Shih, Dong-Her, Ting-Wei Wu, Ming-Hung Shih, Wei-Cheng Tsai, and David C Yen. 2021. “A Novel Auction Blockchain System with Price Recommendation and Trusted Execution Environment.” Mathematics 9 (December): 3214. https://doi.org/10.3390/math9243214.
“The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2020.” 2020. NobelPrize.org. 2020. https://www.nobelprize.org/prizes/economic-sciences/2020/press-release/.
“Types of ICO Auctions | CoinCodex.” n.d. Coincodex.com. Accessed April 17, 2022. https://coincodex.com/article/60/types-of-ico-auctions/.
Wahaballa, Abubaker, Zhen Qin, Hu Xiong, Zhiguang Qin, and Mohammed Ramadan. 2015. “A Taxonomy of Secure Electronic English Auction Protocols.” International Journal of Computers and Applications 37 (January): 28–36. https://doi.org/10.1080/1206212X.2015.1063248.
Wittwer, Milena. 2018. “Pay-As-Bid vs. First-Price Auctions: Similarities and Differences in Strategic Behavior.” ID 2937129. Social Science Research Network. https://papers.ssrn.com/abstract=2937129.
Yan, Guanghui, Shan Wang, Zhifei Yang, and Yi Zhou. 2020. “Dynamic Game Model for Ranking Bitcoin Transactions under GSP Mechanism.” IEEE Access 8: 109198–206. https://doi.org/10.1109/ACCESS.2020.3001157.
Zhang, Jie, Linjing Li, and Fei-Yue Wang. 2017. “A Probabilistic Mechanism Design for Online Auctions.” IEEE Access 5: 10782–94. https://doi.org/10.1109/ACCESS.2017.2705120.
About the student author
Student Author: Tianyu Wu
Tianyu Wu, a junior Math student at Duke Kunshan University (DKU), full admission scholarship recipient, and 2019-2020 National Scholarship recipient. He is an interdisciplinary researcher with his interest and research experiences in blockchain and cryptocurrency, game theory. He once served as the student project leader of “How FinTech Empowers Asset Valuation: Theory and Applications” for the Summer Research Scholar Program at Duke Kunshan University in 2021. He is also the Interim Chair of Human Resources in SciEcon CIC,Tianyu Wu, a junior Math student at Duke Kunshan University (DKU), full admission scholarship recipient, and 2019-2020 National Scholarship recipient. He is also the Interim Chair of Human Resources in SciEcon CIC, leading the development of the HR management system in the era of digital transformation. Contact him on LinkedIn or Twitter.
Creative Commons Attribution 4.0 International License (CC-BY 4.0)
