Transfin.
HomeNewsGuidesReadsPodcastsTRANSFIN. EOD
  1. Reads
  2. Deep Dives

Why Paul R Milgrom and Robert B Wilson Won This Year's Economics Nobel Prize

Editor, TRANSFIN.
Oct 19, 2020 5:54 AM 5 min read
Editorial

This year’s Nobel Memorial Prize in Economic Sciences has been awarded to American economists Paul R Milgrom and Robert B Wilson for “improvements to auction theory and inventions of new auction formats”.

The work done by the economists, who are presently professors of humanities and sciences at Stanford University, “have benefited sellers, buyers and taxpayers around the world”, according to the Nobel Committee.

Honourable Mentions

This year’s choices for the coveted award mark the first time in 26 years that game theorists are the recipients (the previous winner was John Nash). They also mark the third time since 2007 that the award has gone to designers of market mechanisms.

Because that’s what Mildrom and Wilson’s work revolves around - using game theory principles to make auctions more efficient for real-world market applications.

To understand the nature of this year’s Nobel-winning idea, it’s important to understand what auction theory is in the first place…

 

Auction Theory for Dummies

If you think about it, auctions address one of the fundamental concepts in economics - finding what a commodity is worth + deciding who gets to buy it.

Auctions are not a new phenomenon. They have been around for centuries. In their modern form, they are used to sell everything from paintings and rare collectibles to livestock, used cars and fine wine.

But from a policy standpoint, auctions serve a broader purpose. They are used to sell drilling rights, radio spectrum, solar energy farms and even the right to pollute vis-a-vis carbon credits.

Because the commodities involved in auctions can be so varied, a one-size-fits-all approach to auction design and management is bound to fail.

This is where “auction theory” comes into play. It is the branch of economics that studies how different players (bidders and auctioneers) behave in an auction market and proposes different ways to conduct auctions so as to optimise the process, ensure more profitability for one party or both parties, and prevent abuse of the system (such as through collusion among bidders).

 

Types of Auctions

Not all auctions are the same. The type we may be acquainted to thanks to Hollywood - the model of a group of bidders gathered in a room and shouting out their bids one after another until a bid is not topped - is called the English auction.

But there are others. If an auction is required to be completed within a certain period of time, it is a Scottish auction. In a Dutch auction, the seller sets a high price for the commodity being sold, and gradually lowers the ask price until one of the bidders makes a bid. If a Dutch auction also stipulates that once the bidding commences no new bidder can join, it becomes a Japanese auction.

Then there are auctions where prospective buyers place their bids in sealed envelopes and hand them to the auctioneer; the one with the highest bid wins and pays the bidding amount. Another variant of this First-price sealed-bid auction involves the party with the highest bid winning - but being required to pay a price equal to the second-highest bid. Predictably, these are called Second-price sealed-bid auctions.

 

The Ideal Auction

Let’s say we live in a perfect world where perfect auctions can take place. In such a utopia:

  • all bidders would be risk-neutral, 
  • the amount of information possessed by all bidders is the same (symmetric information), 
  • each bidder comes to the auction with an informed private valuation of the commodity being sold, 
  • and both sides of the table have simple objectives (for the auctioneers, to make the most profit; and for the bidder, to win the auction by spending the least amount of money).

But we don’t live in a perfect world. Ipso facto, auctions are flawed.

Bidders are not risk-averse, there can always be information asymmetry, not all bidders may have private valuations - or even properly-informed ones - and both sides of the table may have more intentions in mind than money.

And this is where Milgrom and Wilson come into the picture.

 

Telecom Talk

Before the mid-1990s, the US didn’t have a fixed way of selling licenses to electromagnetic spectrum to telecom companies. Sometimes, companies were asked to make presentations of their offers (these were called “beauty contests”); other times, the winners were chosen by lottery.

These methods were both inefficient - Uncle Sam got hardly any greenbacks in return - and prone to abuse - they led to excessive lobbying by vested interests.

In 1994, the US decided to try auctioning telecom spectrum licenses instead. For this, a group of economists - including this year’s two Economics Nobel laureates - were enlisted. They devised a new auction format called a Simultaneous Ascending Multiple-Round Auction (SMRA) which enabled the US government to pocket $617m that year instead of virtually nothing.

 

What is an SMRA?

An SMRA differs from a conventional auction in two main ways:

  1. In an SMRA, the “simultaneous” means that all licenses are sold at the same time in an auction round.
  2. Bidding is conducted in successive rounds instead of being a continuous process.

What does that mean, exactly?

The first difference dissuades buyers (in this case, telcos) from buying frequencies in only some geographies and monopolising them or selling them in turn to larger telcos.

The second difference is an interesting one. Bidding is held in successive and discrete rounds, with the length of each round being predetermined.

After one round closes, results are made public. Now, each bidder knows what the general sentiment is. So they can adjust their bidding strategies accordingly and alter their bid price in the second round.

In an SMRA, bidding continues until only one bidder remains. Sort of a “last bidder standing” situation.

Since 1994, the SMRA technique has been used for a variety of purposes, including auctioning electricity and natural resources. It has also been adopted by countries around the world...including India.

 

The Flaw in the Plan

Now, India’s tryst with this year’s Nobel-winning idea has not particularly been a clear-cut success.

India began using SMRA to sell spectrum licenses in 2010. That year’s sale saw all licenses sold above the reserve price set by the Government.

Since then, however, even as SMRA has been continuously used in every telecom auction, spectrum sales have happened increasingly at reserve price levels.

At the same time, the costs incurred by telcos have skyrocketed. This is both due to high spectrum fees and due to increased competition within the sector. From 2016, with the advent of Reliance Jio and dirt-cheap data rates, the telecom industry has seen tariffs plummet, debt soar and bankruptcies rise. This has led to consolidation such that presently there are only three major private-sector operators in the country.

Another flaw may be setting revenue generation as the goal for the auctioneer - in this case, the Government. The Indian Government often uses telecom spectrum revenue to help finance its deficits. So it has an active interest in setting high spectrum prices.

But this means the telcos shell out large sums of money and incur high levels of debt...which means they have less capital to invest in infrastructure and service quality. In 2019, India had 50 times as many subscribers per MHz of spectrum as did Germany. Therefore, eventually, the consumers - and, inadvertently, the Government - lose out.

Auction theory may finally have its day in the sun thanks to this year’s Economics Nobel laureates. But as we have seen, it still has a lot to fix and uncover. So that’s work cut out for a future Nobel laureate!

FIN.

The cut-throat world of Business and Finance means that there is fresh News everyday. But don't worry, we got you. Subscribe to our Wrap Up Newsletter and get commentaries like the above straight to your inbox.