How to design Nobel Prize Worthy Auctions?

In today's Finshots, we talk about SMRAs and spectrum auctions.

The Story

Auctioning spectrum is always a tricky task. For starters, you have these bands of frequencies — 700MHz band, 800MHz band, 2300 MHz that most people can’t make sense of. So you have to visualize them a bit differently. Maybe think of these bands as highways fashioned using airwaves. That is to say — If you’re a telecommunication company and you wanted to facilitate voice or video calls then you can’t do your job unless you have access to these highways.

And on most occasions, these bands are subdivided across multiple regions. Think Bangalore, NCR, etc. So for instance, if you bought the 700 MHz band in Bangalore, then you’ll have exclusive rights to transmit over these airwaves in the region — no questions asked.

But auctioning spectrum is a relatively new phenomenon. Things were a bit different in the early 1900s. When the US started allocating spectrum to radio companies back in 1930, they had “comparative hearings” or more popularly called as beauty contests. The spectrum was allocated to prospective radio companies for absolutely no cost. All they had to do was appear before a panel and make their case. And the commission would select the applicant who best advanced the public interest. However, this qualitative criterion was a bit dubious. After all, who determines what public interest actually means. And since policymakers are always susceptible to influence, competitors could cry foul alleging favouritism. It was a bit arbitrary.

But then with a surge of new applications (especially from TV broadcasters), policymakers decided to make the process “completely arbitrary” by resorting to lotteries. And while it did eliminate some problems associated with comparative hearings, it brought some new challenges to the fore.

Here’s an excerpt from an article in Priceonomics succinctly summarizing this issue

“As it turns out, selecting license winners at random sometimes meant that the people who ended up with large tracts of one of the nation’s most valuable natural resources were, well, a little random. People, for example, like the actor Ernest Borgnine.

As part of a syndicate of lottery participants, the actor was one of hundreds of thousands of citizens who played the cellular telephony lotto in the mid-1980s. According to an article in Forbes from 1990, Borgnine’s group were one of the lucky ones and took home the right to provide cell service to Yakima, Washington, which they then sold for a handsome profit."

Clearly, lotteries were a suboptimal solution too. And policymakers were beginning to realize that auctioning spectrum for telecom companies was a whole different ball game. After all, telcos were expected to earn billions of dollars in revenues by leveraging these spectrum licenses. The government had to get its cut. So the natural solution here was to look at auctions.

But what kind of auction exactly?

The most simplistic auction would involve selling bands of frequencies one at a time sequentially.

For instance, in March 2000, the Swiss government decided to sell three bands of frequencies across the entire country. Meaning if you had a license to transmit in one of these bands, then you could do it in the whole of Switzerland. The first license was sold for 121 million francs after a flurry of bidding activity. And once this block sold out, bidders had a clear idea of what to expect with the next band (which by the way was more or less similar to the first band). This one sold for 134 million francs — a slight premium, but definitely in line with expectations. However, the biggest block was still up for grabs. This was twice the size of the first two blocks. But when it sold for just 55 million francs, people were puzzled. How on earth did the most valuable bloc sell for such a pittance.

Well…

When participants offer their bid in these kinds of auctions, they always have to think about future bids on subsequent items, since every item might be of interest. If they overpay and nab the first item, then that might effectively render them incapacitated in future auctions. And as more participants drop out by virtue of these miscalculations, there is every possibility that the most valuable bloc might go to the last man standing. As one report notes —"Bidders are likely to regret having purchased early at high prices, or not having purchased early at low prices."

It's like the IPL. When marquee players are put to auction, there is a frenzy of activity with teams trying to outbid each other. But as the player roster begins to run thin, you'll see a bunch of teams folding simply by virtue of running out of money. And if a decent player were put on the auction floor then, you'll likely see a team nab him at a bargain price. It's just the way these auctions work.

Besides, there is another more sinister problem at play here.

Let’s assume that a telco wanted a certain radio band in Karnataka. But they only want it if they can also have broadcast rights in Tamil Nadu. Now it's quite possible that in a sequential auction, the participant might win the Karnataka bloc, while losing out on Tamil Nadu when it goes to auction. This might render the Karnataka block worthless and so it's safe to say sequential actions don’t produce optimal outcomes. So you need something else.

And this is where Robert Wilson and Paul Milgrom come in.

They designed the Simultaneous Multiple-Round Auction (SMRA) to alleviate these problems. In this format, all the items go to auction at the same time. That is to say, the Tamil Nadu bloc and the Karnataka bloc will go to auction at the same time.

So let’s suppose Airtel is interested in Tamil Nadu and they bid 100 crores for this item during the first round. But then we have Jio — a company that’s interested in both Tamil Nadu and Karnataka. So they bid 90 crores on Tamil Nadu and another 95 Crores on Karnataka. At the end of the round, participants can see the highest bid for each market and they can adjust their preferences accordingly.

If at any point, Jio feels it's overpaying for the combination, they can drop out by not submitting a bid in the subsequent rounds. The auction finally ends when a round passes in which no bids are made on any of the items. And while we grossly oversimplified the technicalities here, it’s safe to say that the design of SMRA’s fundamentally changed our understanding of auctions and its eventual application brought in billions of dollars in additional revenue for governments across the globe.

And that's why the duo won the Nobel Prize.

Until next time.

And remember, if you thought this was insightful or if you know anybody who likes auctions or even IPL for that matter, don't forget to share this article on WhatsApp, Twitter, or LinkedIn.