Last September, when Treasury Secretary Henry Paulson first described to a startled world his $700 billion plan to deal with the emerging financial crisis, it was with these words: “And we’ll use – you know, we’re working through the processes, but there will be some form of auction process.”
Eight months later, the first auctions of those toxic assets are still a month away – assuming the Public Private Investment Plan, or PPIP, gets off the ground on schedule. Even now, its success is not assured.
Unmistakable, however, is the centrality of auction technology to the twenty-first century – indeed, of economic engineering in general. Newly-created markets resting on “some form of auction process” can be found for commodities of all sorts these days, from electricity and advertising to carbon emission and radio-spectrum rights.
That such new machinery might be equal to the task was anything but obvious fifteen years ago when Federal Communications Commission Chairman Reed Hundt boldly presided over the first successful modern high-tech auction, designed to carve up the emergent national market for broadband personal communication services. Congress had only required that he sell the relevant frequencies, rather than give them away.
Computer software replaced the sealed bids with which the government was accustomed to selling potentially valuable property, such as offshore oil and gas leases. Simultaneous ascending bids on multiple licenses replaced the much less complicated method of open-outcry auction of single properties. An activity rule prevented “sniping” – stealthy observation of others’ behavior, round by round, followed by a crashing big bid at the end.
The auction was a smashing success, and not just because it raised a lot of money for the government. Its bigger pay-off was the way it offered a complicated industry a way to begin working on the same page. Its design protocols quickly became the template for many other auctions in many other industries. And before long, it was clear that new tools were regularly being added, not just auction techniques, but clearinghouse techniques that addressed various market failures. Yet as recently as 2002, when Harvard University professor Alvin Roth wrote “The Economist As Engineer,” there was still something novel about the proposition that economic engineering was becoming close kin to chemical engineering or medicine.
Not any more! Auction technology has blossomed around the world, and called into existence a whole new industry, replete with firms, entrepreneurs, consultants, counter-consultants and even a few gunslingers. When the National Bureau of Economic Research’s Market Design Working Group held its first-ever meeting earlier this month, many of the stars were there, mostly economists, but a handful of computer scientists, as well.
That’s because auction technology has become a feature of life on the World Wide Web so familiar that it is scarcely noticed that competitive bidding determines what you see every single time you make a search, or call up a page that displays advertising. For a while, that was the province of the warriors who wrote the big engines’ code. Today, however, big-name economists advise all the top firms: Harvard’s Susan Athey is Microsoft’s chief economist, Berkeley’s Hal Varian is chief economist for Google, Caltech’s Preston McAfee leads a group at Yahoo. Lawrence Ausubel, of the University of Maryland, is helping ICANN, the Internet’s governing body, organize an auction of new top-level domain names. (Roth’s Game Theory, Experimental Economics and Market Design Page is a cornucopia of information on developments and emerging trends.)
For all their success organizing private markets, economists still often face an uphill struggle when it comes to winning government work. The reason is simple: auctions are all about, as the old saying goes, getting the most feathers out of the goose. The geese in the government flock are all owners or would-owners of something the government used to license at nominal rates or ignore altogether – the right to broadcast on a certain frequency, to harvest timber in a national forest, to send pollutants into the air. They would have paid for the privilege in the past – a relative or two on the payroll, good seats at the baseball game, a bundle of campaign contributions.
But whatever they paid for the privileges they received probably would have borne little relation to the worth of the franchise. That’s precisely the information an auction is designed to elicit, along with the cash. No wonder companies reflexively oppose auctions; auctions cost them cash. So spend freely to avoid them.
That’s why, for example, the US Department of Transportation earlier this month canceled plans to auction landing slots for New York’s three busiest airports. The Bush administration had sought the measure, hoping to cut delays at the chronically congested airports (and, of course, raise some much-needed cash). The airline industry lined up against the proposal, so did Democratic congressmen. Incumbent airlines will continue to profit; frequent travelers will continue to suffer delays.
Similarly, the banking lobby, among the nation’s strongest interest groups, has so far successfully opposed Treasury Department attempts to put up for bid banks’ questionable (now “legacy”) assets. The reason is simple: when the asking price is, say, 90 cents on the dollar and the bid is closer to 40 cents, no manager will willingly take part in an auction that seems certain to lower book values. The PPIP plan is designed to produce artificially high bid prices by leveraging private capital with government guarantees – a mechanism that seems unlikely to produce much of a gain for the taxpayer aside from jolting the system itself back into gear. The alternative would be for the government itself to temporarily take over threatened banks and hire the auctioneer – the dreaded “nationalization.” (For a look back at the design of an off-the-shelf auction that could have been conducted in a matter of days back in October 2008, see “A Troubled Asset Reverse Auction,” by Lawrnce Ausubel and Peter Cramton.)
Even when overwhelming pressures bring about some fundamental change in the regulatory apparatus, you can count on industry to oppose the auction mechanism. That’s the case with the landmark legislation to limit greenhouse gas emissions through a cap-and-trade permit system expected to be taken up by the full Congress in July. President Obama campaigned on a promise to auction the permits. But a coalition of Midwestern and Southern Democrats teamed up to alter the bill, and when its language was released last week it turned out that fully 80 percent of the permits would be given away at first to electricity utilities and their big industrial customers, with the portion of permits to be sold at auction slowly rising to 100 percent by 2030. Consumers will experience higher prices only slightly less slowly, but owners of capital will have twenty years to adjust. The gradual transition is a price the president is said to be willing to pay in order to pass the bill into law. Equality of sacrifice will take a back seat to the goal of capping carbon emissions, a tradeoff noted with special force by journalist Matt Yglesias.
But the historical momentum in this case is clearly on the side of equality. Auctions, especially auctions of government property, are not a tool of the rich, especially when coupled with egalitarian principles of distribution (for instance, the proposition that every citizen should benefit equally when the radio spectrum is sold). As principles of market design become more thoroughly articulated and widely understood, the sphere of governmental discretion will shrink. More and more, politicians will be forced to play by the rules.
For example, serial entrepreneur Paul Milgrom, of Stanford University, who, along with Stanford’s Robert Wilson, played a key role in that first great auction in 1994, last week opened the Palo Alto office of a new firm, Auctionomics, with a proposal to avoid the kinds of manipulation that have sometimes plagued carbon markets in Europe. Allow firms to borrow against their future credits, when liquidity is short, he urged, in order to prevent market “corners.” Such “discounted time shifting” – emitting 95 tons of carbon this year in exchange for a promise to reduce emission by 100 tons next year – could end the manipulative practices that swept California electricity markets in 2000-01, causing blackouts and imposing huge financial tolls on ratepayers.
It’s the beginning of another episode in the eternal cat-and-mouse game between those who seek to exploit the political and market systems and those determined to prevent them, or at least diminish the opportunities for mischief – the latter, increasingly, are economic engineers. There will be many disappointments. But expertise in market design ordinarily will win out.