Auctions for Multiple Goods, and Understanding Consumer Preferences
Lead Research Organisation:
University of Oxford
Department Name: Economics
Abstract
In 2007, responding to the Northern Rock bank run--Britain's first bank run since the 1800's-- Klemperer invented the "Product-Mix Auction", to improve the Bank of England's allocation of funds to banks, building societies, etc. (Klemperer, 2008, 2010). An Executive Director of the Bank of England described it as "a world first in central banking" and "potentially a major step forward in practical policies to support financial stability" (Milnes, 2010), and the Governor (Mervyn King) told The Economist (Oct. 19, 2012) that "Klemperer auctions [are] a marvellous application of theoretical economics to a practical problem of vital importance to financial markets". The auctions improve efficiency by simultaneously allocating distinct goods - in this example, funds against different collaterals.
This project will further develop the Product-Mix Auction design to make it a practical tool in a broader range of contexts. For example, versions of the auction could be used for batch processing of stock trades to eliminate high-frequency trading problems. They could also be important in achieving the U.K.'s goal "to start auctioning electricity-supply contracts. ... by the 2020s,...across technologies. 'We would be the first country in the world to achieve that', boasts [the U.K. Cabinet Minister for Energy and Climate Change]" (The Economist, Dec. 1, 2012).
The theoretical work required to extend the range and flexibility of the auction will also shed light on a much wider range of issues. For example, our work will develop a better characterisation and understanding of consumer preferences when goods are indivisible, yield answers to some deep problems about the existence of economic equilibrium, and have applications to matching theory.
We will use recently-developed "tropical geometric", and other geometric, techniques as well as more standard economic tools to develop a better understanding of economic agents' preferences. We will develop our theory of "demand types" which provides a framework to check and compare the properties of agents' demands, and develop ways to depict preferences of different "types".
The project will thus have both theoretical and practical outputs: in addition to preparing academic papers for submission to leading economic theory journals, and presenting at academic conferences, we will also discuss our work with potential users of our auction designs, and will develop software implementations of our auction procedures that will be easy to use for demonstration and experimentation purposes and provide a platform on which problem-specific applications can be developed.
This project will further develop the Product-Mix Auction design to make it a practical tool in a broader range of contexts. For example, versions of the auction could be used for batch processing of stock trades to eliminate high-frequency trading problems. They could also be important in achieving the U.K.'s goal "to start auctioning electricity-supply contracts. ... by the 2020s,...across technologies. 'We would be the first country in the world to achieve that', boasts [the U.K. Cabinet Minister for Energy and Climate Change]" (The Economist, Dec. 1, 2012).
The theoretical work required to extend the range and flexibility of the auction will also shed light on a much wider range of issues. For example, our work will develop a better characterisation and understanding of consumer preferences when goods are indivisible, yield answers to some deep problems about the existence of economic equilibrium, and have applications to matching theory.
We will use recently-developed "tropical geometric", and other geometric, techniques as well as more standard economic tools to develop a better understanding of economic agents' preferences. We will develop our theory of "demand types" which provides a framework to check and compare the properties of agents' demands, and develop ways to depict preferences of different "types".
The project will thus have both theoretical and practical outputs: in addition to preparing academic papers for submission to leading economic theory journals, and presenting at academic conferences, we will also discuss our work with potential users of our auction designs, and will develop software implementations of our auction procedures that will be easy to use for demonstration and experimentation purposes and provide a platform on which problem-specific applications can be developed.
Planned Impact
Likely beneficiaries of our improved auction designs are in the following areas where the Applicant and Co-Investigator have already developed working relationships (see Pathways to Impact):
Central Banks
Within the Bank of England (which the Applicant continues to advise), one possible use of an extended Product-Mix Auction might be to implement recommendations in the October 2012 Winters Review; another possibility would be to run a future version of the Asset Purchase Facility to buy corporate bonds using a Product-Mix Auction approach. These applications require a thorough understanding of complications arising from using more than two goods (especially issues about representing bidders', and the auctioneer's, preferences in user-friendly ways).
There will likely be other applications in central banking and, of course, in other central banks.
Other Financial Applications
There are many other potential applications in finance. For example, a version of our proposed auction that could handle very many goods might be used to do batch processing of stock trades to eliminate high-frequency trading problems (as suggested by, for example, Professors Budish and Cramton).
Telecommunications Authorities
The Applicant and Co-Investigator have already had discussions with officials of the U.K.'s OFCOM about possible applications of potential developments of the Product Mix Auction in several contexts including spectrum allocation. The Applicant has had similar discussions with the US's Federal Communications Commission where he has been a pro bono consultant. OFCOM's Director of Competition Economics (Mr. Geoffrey Myers) has said he would like to arrange future discussions about specific applications of extensions of the Product-Mix Auction. A particular feature of these applications is potential complications arising from bidders viewing different goods as complements.
Energy auctions
Both the current and the previous Cabinet Ministers for Energy and Climate Change have consulted the Applicant about potential uses of auctions within the energy industry, and the Co-Investigator has spent two months consulting on this issue within the department. As noted above, The Economist (Dec 1, 2012, p33) recently wrote, the U.K. "is due to start auctioning electricity-supply contracts as early as 2017. ... by the 2020s, it is hoped, such auctions will be held across technologies. 'We would be the first country in the world to achieve that', boasts [the Cabinet Minister]." The Applicant has advised that more work on auction design needs to be done if the Minister's boast of a 'world-first' is to be achieved--an extension of the Product-Mix Auction is a natural candidate for this application. The Department's Chief Economist (Dr. Steven Fries) has said he is keen for the Department to continue to interact with the Applicant and Co-Investigator on these issues. In this context there are potential complications arising from the auctioneer viewing different goods as complements.
Other Public Sector Applications
Examples of other UK government departments which the Applicant has advised (pro bono), and which the project results might in due course benefit, include the Departments for Business, Innovation and Skills; for Communities and Local Government; and for Environment, Food and Rural Affairs.
Applications in Industry
The Applicant is routinely contacted by commercial firms, and occasionally provides pro bono advice. These contacts will give opportunities to introduce these ideas to industry via our easy-to-use demonstration versions.
Central Banks
Within the Bank of England (which the Applicant continues to advise), one possible use of an extended Product-Mix Auction might be to implement recommendations in the October 2012 Winters Review; another possibility would be to run a future version of the Asset Purchase Facility to buy corporate bonds using a Product-Mix Auction approach. These applications require a thorough understanding of complications arising from using more than two goods (especially issues about representing bidders', and the auctioneer's, preferences in user-friendly ways).
There will likely be other applications in central banking and, of course, in other central banks.
Other Financial Applications
There are many other potential applications in finance. For example, a version of our proposed auction that could handle very many goods might be used to do batch processing of stock trades to eliminate high-frequency trading problems (as suggested by, for example, Professors Budish and Cramton).
Telecommunications Authorities
The Applicant and Co-Investigator have already had discussions with officials of the U.K.'s OFCOM about possible applications of potential developments of the Product Mix Auction in several contexts including spectrum allocation. The Applicant has had similar discussions with the US's Federal Communications Commission where he has been a pro bono consultant. OFCOM's Director of Competition Economics (Mr. Geoffrey Myers) has said he would like to arrange future discussions about specific applications of extensions of the Product-Mix Auction. A particular feature of these applications is potential complications arising from bidders viewing different goods as complements.
Energy auctions
Both the current and the previous Cabinet Ministers for Energy and Climate Change have consulted the Applicant about potential uses of auctions within the energy industry, and the Co-Investigator has spent two months consulting on this issue within the department. As noted above, The Economist (Dec 1, 2012, p33) recently wrote, the U.K. "is due to start auctioning electricity-supply contracts as early as 2017. ... by the 2020s, it is hoped, such auctions will be held across technologies. 'We would be the first country in the world to achieve that', boasts [the Cabinet Minister]." The Applicant has advised that more work on auction design needs to be done if the Minister's boast of a 'world-first' is to be achieved--an extension of the Product-Mix Auction is a natural candidate for this application. The Department's Chief Economist (Dr. Steven Fries) has said he is keen for the Department to continue to interact with the Applicant and Co-Investigator on these issues. In this context there are potential complications arising from the auctioneer viewing different goods as complements.
Other Public Sector Applications
Examples of other UK government departments which the Applicant has advised (pro bono), and which the project results might in due course benefit, include the Departments for Business, Innovation and Skills; for Communities and Local Government; and for Environment, Food and Rural Affairs.
Applications in Industry
The Applicant is routinely contacted by commercial firms, and occasionally provides pro bono advice. These contacts will give opportunities to introduce these ideas to industry via our easy-to-use demonstration versions.
Publications
Baldwin E
(2019)
Understanding Preferences: "Demand Types", and the Existence of Equilibrium With Indivisibilities
in Econometrica
Baldwin E
(2015)
Understanding Preferences: 'Demand Types', and The Existence of Equilibrium with Indivisibilities
in SSRN Electronic Journal
Baldwin E
Strong Substitutes: Structural Properties, and a New Algorithm for Competitive Equilibrium Prices
in Mathematical Programming
| Description | This research produced two main outputs. First, we developed a new framework for understanding consumer preferences, which illuminates (among other things) when competitive equilibrium exists. Although the answer to this question is relatively well understood when all goods are divisible, much less has been known about the case with indivisible goods. Our methods obtained several new theorems that provided new answers and insights about the existence and nonexistence of equilibrium. A significant research paper is, we think, at the final stage of the reviewing and editing process. This work also opened up a range of important research questions for us and others to work on in the future. For example, our theoretical results can probably be developed to show that even when existence is not guaranteed, a limit on the "extent of failure of equilibrium" can be obtained, which may open up possibilities for market mechanisms in situations that are currently understood to be problematic, but in which limited failures of equilibrium can be tolerated. And, of course, our work identifying environments in which competitive equilibrium can exist also has implications for the design of practical auctions, as we discuss below. The application of our work to matching markets provides another avenue for research. Second, we developed and implemented a variety of algorithms to run efficient auctions when multiple differentiated goods are to be traded: --we made a series of extensions to the auction design that the principal investigator originally invented for the Bank of England in the financial crisis, and was further developed for the Bank by the principal and co-investigator; --these extensions also allow more complex bidder preferences for the case of multiple divisible goods --we used the deeper knowledge of the geometry of consumer preferences (as described above) to develop an auction for indivisible goods, incorporating a bidding language which may communicate all preferences from a certain important class ("strong substitutes"). Although others have considered this issue, the fact that we can efficiently find equilibrium from bids depicting any valuation of this class is a significant achievement; --we also developed a prototype auction for the case in which bidders' budget constraints are very important to them. This software will both allow practical applications, and also allow experimentation to better understand the specific circumstances in which these, and other, auction designs are most appropriate. |
| Exploitation Route | We envisage both the practical and theoretical strands of our research being taken forward. We will ourselves continue to develop theory based upon our new "tropical geometric" framework for understanding consumer preferences. Others are beginning to become interested in this area of research as well. The tropical geometry community are excited that their research has these applications. For example, recent work by Ngoc Tran and her coauthors use similar ideas in mechanism design, to explore characterisations of incentive compatible mechanisms, and moral hazard. We expect our computer programs to be used for experimenting on the merits of different auction types. Several academic researchers are showing interest in this. |
| Sectors | Digital/Communication/Information Technologies (including Software),Energy,Financial Services, and Management Consultancy |
| URL | http://pma.nuff.ox.ac.uk/ |
| Description | Bank of England's ILTR Auction |
| Geographic Reach | National |
| Policy Influence Type | Membership of a guideline committee |
| Impact | Prof Klemperer has advised the Bank of England on auction designs since 2007. In 2007-8 he invented a new auction design; the then-Governor Mervyn King said "The Bank of England's use of Klemperer auctions in our liquidity insurance operations is a marvellous application of theoretical economics to a practical problem of vital importance to financial markets," (as quoted in the Economist magazine, after the auction was fully implemented in 2010 ). The auction allocates funds more efficiently, and gives the Central Bank better information about relative prices and stresses in the market than previously, and is now an established part of the Bank's architecture defending against future financial crisis. In 2014, the Bank of England launched a new, improved, version of the auction with our help. We continue to advise the Bank of England on the auction's use, and future development. |
| Description | DECC CfD auctions |
| Geographic Reach | National |
| Policy Influence Type | Participation in a guidance/advisory committee |
| Impact | Auctions to set subsidy levels for Renewable Electricity Generation Renewable technologies are subsidised via a "contract for difference", and the strike price is set by an auction. This was originally designed and managed by the Department of Energy and Climate Change (DECC), and has now been transferred to the department of Business, Energy and Industrial Strategy (BEIS). Since the ESRC project start, the Co-Investigator has been in contact with the design team, highlighting problems and potential problems, and suggesting possible improvements. The auction was run in February 2015 and again in April 2017. Discussions continue on future modifications. Back in 2012, the Co-Investigator developed possible design templates with DECC. Several significant aspects of the auction design used follow the original recommendations of the Co-Investigator, but the final design was not developed in tandem with us in the same way as in the Bank of England. |
| URL | https://www.gov.uk/government/publications/contracts-for-difference/contract-for-difference |
| Description | cato institute interest(1) |
| Geographic Reach | North America |
| Policy Influence Type | Citation in other policy documents |
| URL | https://www.cato.org/blog/reforming-last-resort-lending-flexible-open-market-alternative |
| Description | cato institute interest(2) |
| Geographic Reach | North America |
| Policy Influence Type | Citation in other policy documents |
| URL | https://www.cato.org/blog/cowen-central-banks-liquidity-traps |
| Description | iceland's interest in PMA |
| Geographic Reach | Europe |
| Policy Influence Type | Citation in other policy documents |
| URL | http://www.voxeu.org/article/why-iceland-can-now-remove-capital-controls |
| Description | mentions in public policy documents |
| Geographic Reach | North America |
| Policy Influence Type | Citation in other policy documents |
| Impact | influence on financial practitioners + regulators |
| URL | https://www.cdhowe.org/video-and-media-appearances/bank-canada-should-prepare-now-next-financial-cri... |
| Description | mentions in public policy documents |
| Geographic Reach | National |
| Policy Influence Type | Citation in other policy documents |
| Impact | influence on financial practitioners http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2015/q206.pdf |
| Description | (DUALMARKETS) - Duality in Market Design: Theory and Applications |
| Amount | € 1,060,381 (EUR) |
| Funding ID | 949699 |
| Organisation | European Commission |
| Sector | Public |
| Country | European Union (EU) |
| Start | 07/2021 |
| End | 06/2025 |
| Description | ILTR Extension |
| Organisation | Bank of England |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | Advice on auction design: In 2014, the Bank of England launched a new, improved, version of their auctions for indexed long-term repo (ILTR) operations. This auction had been developed with the help of both the Principal Investigator and the Co-Investigator, and using techniques from our research into the geometry of consumer preferences. The 2014 development built on auction design work the Principal Investigator had been undertaking for the Bank of England since 2007. In 2007-8 he invented a new auction design, the original "product-mix auction" for ILTR operations; the then-Governor Mervyn King said "The Bank of England's use of Klemperer auctions in our liquidity insurance operations is a marvellous application of theoretical economics to a practical problem of vital importance to financial markets," (as quoted in the Economist magazine, after the auction was fully implemented in 2010 ). Although the original collaboration began before the ESRC project in question, this key update took place during the award. We continue to advise the Bank of England on the auction's use, and future development. |
| Collaborator Contribution | Our partners in the Bank of England discussed their needs with us and tested implementations of our ideas, for discussion. |
| Impact | These auctions allocate funds more efficiently, and gives the Central Bank better information about relative prices and stresses in the market than previously. They are now an established part of the Bank's architecture for routine use, and defending against future financial crisis. For example, ILTR auctions were used to stabilise markets after the Brexit vote. The 2014 update to the auction increased the variety of collaterals that could be handled, and allowed the total market allocation to respond to bidders' demands. |
| Start Year | 2014 |
| Description | Icelandic Auctions for Offshore Assets |
| Organisation | Central Bank of Iceland |
| Country | Iceland |
| Sector | Public |
| PI Contribution | Discussions with Icelandic government and Icelandic central bank on design of a new Product-Mix Auction to permit blocked offshore Icelandic funds to be spent on one or more of three alternative financial instruments. This design was different from that used by the Bank of England in their Indexed long-term repo operations, because each bidder would always be allocated additional units to spend their entire budget. This auction design was developed in partnership with the Icelandic authorities in the expectation that it would be used in practice. |
| Collaborator Contribution | Discussed needs. |
| Impact | Unfortunately political developments in Iceland meant that this auction did not take place. However, it led to some press and discussion of the auction (see citations in "Impact on Policy, Practice, Patients and the Public) |
| Start Year | 2015 |
| Description | Icelandic Auctions for Offshore Assets |
| Organisation | Government of Iceland |
| Country | Iceland |
| Sector | Public |
| PI Contribution | Discussions with Icelandic government and Icelandic central bank on design of a new Product-Mix Auction to permit blocked offshore Icelandic funds to be spent on one or more of three alternative financial instruments. This design was different from that used by the Bank of England in their Indexed long-term repo operations, because each bidder would always be allocated additional units to spend their entire budget. This auction design was developed in partnership with the Icelandic authorities in the expectation that it would be used in practice. |
| Collaborator Contribution | Discussed needs. |
| Impact | Unfortunately political developments in Iceland meant that this auction did not take place. However, it led to some press and discussion of the auction (see citations in "Impact on Policy, Practice, Patients and the Public) |
| Start Year | 2015 |
| Description | RSPB Turtle Doves Habitat |
| Organisation | Dot Econ |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | Working with Dr Alex Teytelboym (dept of Economics, University of Oxford) we have been developing an auction to allocate subsidies for provision of food and habitat for turtle doves. Our focus has mainly been on exploring how to translate ornithological priorities into auction preferences. |
| Collaborator Contribution | RSPB are running the project and will find and inform potential participants, and monitor the compliance of auction winners. Dot Econ are developing the computer optimisation on the basis of our preference work. EnTrade provide the computer front-end, designed for users in the rural environment. |
| Impact | First round of auction was delayed by covid-19, and opened 15th February 2021. Success in this project may inform post-Brexit policy towards replacing agricultural subsidies with "public money for public goods" Multi-disciplinary: relates to ornithology: knowledge of priorities in habitat provision for turtle doves. |
| Start Year | 2019 |
| Description | RSPB Turtle Doves Habitat |
| Organisation | Entrade |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | Working with Dr Alex Teytelboym (dept of Economics, University of Oxford) we have been developing an auction to allocate subsidies for provision of food and habitat for turtle doves. Our focus has mainly been on exploring how to translate ornithological priorities into auction preferences. |
| Collaborator Contribution | RSPB are running the project and will find and inform potential participants, and monitor the compliance of auction winners. Dot Econ are developing the computer optimisation on the basis of our preference work. EnTrade provide the computer front-end, designed for users in the rural environment. |
| Impact | First round of auction was delayed by covid-19, and opened 15th February 2021. Success in this project may inform post-Brexit policy towards replacing agricultural subsidies with "public money for public goods" Multi-disciplinary: relates to ornithology: knowledge of priorities in habitat provision for turtle doves. |
| Start Year | 2019 |
| Description | RSPB Turtle Doves Habitat |
| Organisation | Royal Society for the Protection of Birds (RSPB) |
| Country | United Kingdom |
| Sector | Charity/Non Profit |
| PI Contribution | Working with Dr Alex Teytelboym (dept of Economics, University of Oxford) we have been developing an auction to allocate subsidies for provision of food and habitat for turtle doves. Our focus has mainly been on exploring how to translate ornithological priorities into auction preferences. |
| Collaborator Contribution | RSPB are running the project and will find and inform potential participants, and monitor the compliance of auction winners. Dot Econ are developing the computer optimisation on the basis of our preference work. EnTrade provide the computer front-end, designed for users in the rural environment. |
| Impact | First round of auction was delayed by covid-19, and opened 15th February 2021. Success in this project may inform post-Brexit policy towards replacing agricultural subsidies with "public money for public goods" Multi-disciplinary: relates to ornithology: knowledge of priorities in habitat provision for turtle doves. |
| Start Year | 2019 |
| Title | Software to Implement Product-Mix Auctions |
| Description | The software implements three different product-mix auctions. The first ("standard version") is a generalised version of the design that formed the basis of an auction that is now regularly used by the Bank of England (implemented with our help). It allows rich preferences on the side of both the buyers and seller, and with options as to how units should be rationed between buyers when bids are tied. The second is an implementation that was designed for the government of Iceland; in this auction, each of a bidder's bids specifies a budget to spend on the bid's "best-value" good, and so some technical aspects are very different. The last version extends the standard version to permit all participants (buyers and seller) to express *any* strong substitute preferences for indivisible goods, and so is technically more advanced than the other mechanisms. It correspondingly has higher complexity. We have online interfaces to allow easy experimentation with these auctions. All this software is available under the Affero General Public Licence |
| Type Of Technology | Software |
| Year Produced | 2018 |
| Open Source License? | Yes |
| Impact | We hope this open-source software will be of use both in academic experiments and practical applications of the auction designs. |
| URL | http://pma.nuff.ox.ac.uk |
| Description | Discussion of auctions with Department of Environment, Food and Rural Affairs |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Policymakers/politicians |
| Results and Impact | Expert panel discussion on how auctions could be integrated into Environmental Land Management. The department intends to change the way payments are made to rural landowners, focussing on "public money for public goods", and auctions potentially have a large role to play in this. We continue to develop these ideas in ongoing discussions with them. (Colleagues Prof. Alexander Teytelboym and Prof. Daniel Quigley are also collaborating in this project). |
| Year(s) Of Engagement Activity | 2019 |