News & Events

Print/PDF this page:

Print Friendly and PDF

Share this page:

A Reverse Auction to Conserve Kilowatts—A Commentary by Ian Ayres ’86

The following commentary was posted on newyorktimes.com on August 5, 2010.

A Reverse Auction to Conserve Kilowatts
By Ian Ayres ’86

The Missouri Department of Natural Resources on July 28th ran what may be the first-ever online reverse auction for energy efficiency grants. The state allocated $3 million of American Recovery and Reinvestment Act funding in a series of three one-hour auctions by having 23 pre-qualified businesses bid on a $/kWh saved basis for projects that were expected to enhance energy efficiency. Grants with fixed dollar amounts were awarded to the bidders who promised the best (conservation) bang for the (grant) buck. For example, two $500,000 grants were awarded to AmerenUE and The Gasket Guy because they (by bidding 3.25 cents/kWh) effectively promised that they would each be able to demonstrate conserving more than 15 million kilowatt hours.

Peter Cramton and I have written before about how auction rules (in particular regarding affirmative action) can affect auction competition. Having enough serious bidders show up to bid is particularly important. In fact, in an article which has become an instant classic, Jeremy Bulow and Paul Klemperer have shown that most sellers would be better off having one more serious bidder show up to bid than having more market power over the existing set of bidders.

One of the most powerful ways that sellers can impact the degree of bidding competition is by reducing the number of items being auctioned. If you are auctioning two widgets with a simultaneous ascending auction, the price is determined by the third-highest bidder. If you auction four widgets, the price of all four widgets will be determined by the fifth-highest bidder.

Years ago after taking my daughter to see Seusical, the Musical, the cast came on stage and auctioned an autographed cap to help the AIDS charity Broadway Cares. The final two bidders might have gone as high as $700 and $750. After selling the cap to the high bidder for $750, the savvy auctioneer (who might have just played the Cat in the Hat) added that the cast had already signed a second hat and would be happy to sell that to the second highest bidder for $700. Even though it was all for a good cause, the high bidder had reason to feel ill-used. If the cast had simultaneously auctioned the two caps simultaneously, the highest two bidders would only have had to outbid the third highest bidder who dropped out substantially earlier.

The auction for four $250,000 grants was interesting, however, because it shows that simultaneously auctioning more things doesn’t always mean a poorer outcome for the seller. The Missouri auction for the four smaller awards was even more competitive with the winners promising to deliver future conservation for only 2.75 cents of grant for every kWh conserved. Even though there were twice as many grants being offered – which would tend to depress the auction price — there were more bidders willing to bid on delivering a smaller total energy savings. Even at this lower price per kilowatt hour, each of these smaller grant recipients were promising to demonstrate only about 9 million kilowatt hours saved. It also may have helped that this auction happened after the larger auction so that bidders who were shut out before realized that they had to bid more aggressively.

Overall, across three different auctions the averaged promised price-effectiveness was 3.97 cents per kilowatt hour of saved energy. This compares favorably “to recent point estimates of the average cost of other utility energy efficiency programs, which range from 4.7 to 13.3 cents per kilowatt-hour.” (Auffhammer, Blumstein, and Fowlie 2008).

Of course, there are many a slip between cup and lip. Promised and realized energy savings may be very different things. The grant guidelines talk a tough game when it comes to “measurement and verification” of grant recipients who must complete their projects by January 31, 2012. But this verification task would be true under any grant allocation system. For now, the great state of Missouri (the only state with two Federal Reserve banks) has shown that competitive auctions are a feasible way to get the most out of our stimulus money. (Now if only the Treasury would run with the reverse auction idea as a way to value troubled banking assets).