Bid Winners Can Face 'Curse'

Auction Theory Usefulness Has Oily Origin

You won an auction with the high bid?

Too bad.

Even today, auction theory says the highest bidder probably will experience the "Winner's Curse."

It's a concept straight out of the oil industry, and a big reason a number of industries call on auction theory when devising bids.

Auction theory can't tell you what to bid, but it may help you avoid the trap of paying too much.

In fact, "the usefulness of auction theory is the Winner's Curse," said John Morgan, professor of economics in the University of California-Berkeley's Haas School of Business.

A number of bidding strategies exist for different types of auctions.

In the end, they all come down to two rules for winning an auction without overpaying.

The Winner's Curse

The seminal paper on the Winner's Curse was written almost 35 years ago by three employees of Arco — Ed Capen (an AAPG member who has taught several courses for AAPG and wrote articles for the AAPG Treatise book the Business of Exploration), Robert Clapp and William Campbell. The paper was "Competitive Bidding in High Risk Situations" in the Journal of Petroleum Technology, in 1971. That article coined the term "winner's curse."

The researchers were charged with avoiding a repeat of a sale in which Arco bought everthing it bid on and put the company in budgetary straights for a couple of years.

And the reason for those low returns? The authors suggested that the winning bids for leases were too high to allow for a decent profit.

Consider the range of bids in a lease sale, said John Kagel, professor of economics at Ohio State University.

"Suppose those are unbiased estimates. Some are going to be too high, some too low. Some are going to be right in the middle. It's a standard distribution," he explained.

Everyone will estimate a value for the lease and then make a bid below that, leaving room for profit.

"If everybody uses the same rule of thumb — say they all take off 30 percent from their estimate — then the one who's going to win is going to have the highest estimated value," Kagel said.

Some bids will undervalue the lease, and some will overvalue it. Whoever errs the most on the high side will win the auction. In many cases, this also means that the winner is the person who has overestimated the most.

"That's the Winner's Curse," Kagel noted.

The curse is still with us, even after all those years, and even though everyone knows it exists.

Please log in to read the full article

You won an auction with the high bid?

Too bad.

Even today, auction theory says the highest bidder probably will experience the "Winner's Curse."

It's a concept straight out of the oil industry, and a big reason a number of industries call on auction theory when devising bids.

Auction theory can't tell you what to bid, but it may help you avoid the trap of paying too much.

In fact, "the usefulness of auction theory is the Winner's Curse," said John Morgan, professor of economics in the University of California-Berkeley's Haas School of Business.

A number of bidding strategies exist for different types of auctions.

In the end, they all come down to two rules for winning an auction without overpaying.

The Winner's Curse

The seminal paper on the Winner's Curse was written almost 35 years ago by three employees of Arco — Ed Capen (an AAPG member who has taught several courses for AAPG and wrote articles for the AAPG Treatise book the Business of Exploration), Robert Clapp and William Campbell. The paper was "Competitive Bidding in High Risk Situations" in the Journal of Petroleum Technology, in 1971. That article coined the term "winner's curse."

The researchers were charged with avoiding a repeat of a sale in which Arco bought everthing it bid on and put the company in budgetary straights for a couple of years.

And the reason for those low returns? The authors suggested that the winning bids for leases were too high to allow for a decent profit.

Consider the range of bids in a lease sale, said John Kagel, professor of economics at Ohio State University.

"Suppose those are unbiased estimates. Some are going to be too high, some too low. Some are going to be right in the middle. It's a standard distribution," he explained.

Everyone will estimate a value for the lease and then make a bid below that, leaving room for profit.

"If everybody uses the same rule of thumb — say they all take off 30 percent from their estimate — then the one who's going to win is going to have the highest estimated value," Kagel said.

Some bids will undervalue the lease, and some will overvalue it. Whoever errs the most on the high side will win the auction. In many cases, this also means that the winner is the person who has overestimated the most.

"That's the Winner's Curse," Kagel noted.

The curse is still with us, even after all those years, and even though everyone knows it exists.

Morgan cited the stock market's reaction to company acquisitions as a real-world example.

"When a company announces it has succeeded in acquiring another company in a bidding war, the stock price of the company that makes the acquisition suffers negative abnormal profit," Morgan said.

That means its stock price goes down. The market perceives the acquirer must have overpaid, simply because it was the high bidder.

"You want evidence that the Winner's Curse is with us today, and that there's a huge amount of information out there in the fact you've won? This is Exhibit A," Morgan said.

Auction Action?

Google generated plenty of media coverage when it chose an auction to value its initial public offering of shares.

It used a form of Dutch auction, in which more bidders buy in as the share price drops from a high level.

Those first buyers got a bargain, as it turned out. Google's stock price went up by 50 percent within a few weeks.

What can the petroleum industry learn from the Google experience?

"Nothing," agreed Kagel. But the Wall Street Journal wrote that bidders were reading up on the Winner's Curse as they readied their strategies.

Still, it's helpful to understand the main types of auction.

▸ English Auction.

In an English auction, bidders bid against each other as the price rises. When only one bidder remains and no other bidder will raise the bid price, the auction is over.

Notice that all of the bidders know exactly what others are bidding in a typical English auction.

▸ First-Price Sealed-Bid Auction.

In this auction, everyone makes one bid according to each bidder's estimate of value. The highest bid wins and pays the bid price.

No one knows what anyone else is bidding, so the highest bidder isn't aware of winning until the bids are opened. That's "a huge piece of information," Morgan said.

▸ Second-Price Sealed-Bid Auction.

The winner pays the second highest bid price in a second-price auction. Other rules are similar to the first-price auction.

"Knowing you will pay the second bid should force you to bid a higher amount than in a first-price auciton," Capen said. The goal is to maximize net present value.

▸ Dutch Auction.

The general form of the Dutch auction starts with the auctioneer setting a high price and then lowering it until a bidder agrees to pay the price.

You'll find many variations of this idea. Car sales begin with a high list price, and buyers pay progressively lower prices until the lot is cleared.

Strategy Lesson 1

Morgan said almost any oil company can draw on seismic studies, evaluate a lease bloc and arrive at a net present value (NPV) number in a lease sale.

"You might think if you are bidding, a sensible strategy would be (to start with) that NPV and then take an amount below that to allow for a reasonable profit," he said.

But that won't work, because, depending on the opponent's strategy, the most optimistic valuation always wins, he noted.

"That's not good, because there's noise in these seismic studies, and you've just found out that you bid too high," Morgan said.

"The right bidding strategies in these auctions is knowing other people's bidding," he added. "Know thy enemy."

Since you don't know the amount of the other bids in a sealed-bid auction, you have to understand your opponents' bidding practices and motivations.

Luckily for bidders, the petroleum industry has built up a significant history of lease sales and other property auctions, Morgan said.

"The one thing about the oil business is that there's a rich empirical literature about how these auctions work," he noted.

Auction theory says to consider the bidding situation and reduce your bid by an appropriate amount, a practice called "hedging" the bid.

"How much to shade (hedge) depends on how many other bidders there are and what strategies they might employ," Morgan said.

But bidding strategy gets a little quirky in lease sales and other industry auctions, according to Kagel.

You might think you should hedge your bid less when a large number of bidders participate. Having more bidders will force the winner toward full value and reduce the chance for profit.

That makes perfect sense, but it's all wrong.

"In petroleum auctions, after you get more than two or three bidders, you should shade your bid a little more," Kagel said. "That's one of the odd characteristics of mineral rights auctions."

Strategy Lesson 2

Again, consider the range of bids as a standard distribution, Kagel said. With more bidders you have a wider distribution of bids.

That gives you more chance of bidding too high.

To compensate, you should take a little more off your bid price, Kagel explained.

Governments offering mineral leases also call on auction theory to choose the right strategy.

The most basic choice is the type of auction to hold. Timing and the structure of the offer also make a difference, according to Morgan.

"One big issue is the ordering of the lots," he said. "If a small-time guy wins a bid early on, he's out of the picture."

In revealing information about leases in an auction, you might think the leaser should hold something back. When bidders are forced to guess about real value, they might guess too high.

But that's all wrong, too, Kagel said.

"It's the lack of information that creates the opportunity for profit," he said.

"The best thing to do would be to try to provide maximum information about what's under the ground," he added. "That reduces the companies' expected profit."

If everyone follows the right bidding strategy, Kagel explained, they will compensate for lack of information by reducing their bids.

Suppose you are bidding on something that might have a value of $1. If you can't be sure of the value, you will bid well below $1, Kagel noted.

But what if you find out the auction is for a $1 bill? Then all of the bids will tend to rise toward $1, because the exact value is known, he observed.

Can lease auctions be fixed? Probably not, at least not without difficulty — and not by the bidders.

"In a first-price auction, the way that most oil tracts are auctioned, it's not really going to work," Morgan noted.

"The only way that you're really going to affect the price is if you bid the most, but in that case you're on the hook for the tract, and you don't particularly want that," he said.

Auctions combine knowledge, gamesmanship, game theory, expertise, experience and intuition.

In the end, auction theory appears to come down to just two rules for winning an auction without overpaying.

Rule 1: Don't bid too high.

Rule 2: Don't bid too low, either.

You may also be interested in ...