...include the roles of generalized transactions costs, commercial buying practices, management control systems, the costs of managing and monitoring contractors, and measuring "adequate" competition. Of course, in the United States, "since the mid-1950s it has been federal government policy to rely on the private sector to provide commercial products and services as long as those products or services can be obtained more economically" (Gansler, 2003, p. 42). More recently, Office of Management and Budget Circular A-76 tried to facilitate fairness by streamlining and standardizing competition and promoting trust in the federal purchasing system. (1) Generally, we believe that having multiple bidders makes contracting more effective (e.g., Pallesen, 2003; Natural Academy of Public Administration, 1989), (2) although we also recognize that contracting does not always give government better value for its money (John Ward, 2005). Clearly, understanding how bidding processes work is fundamental to understanding government procurement, but bidding is also integral for the sale of lumber, off-shore oil leases, cellular telephone spectrum rights, government securities, defense products, and recently for the provision of health services like Medicare and Medicaid. Other uses include the sale of rights to a natural monopoly (Demsetz, 1968; Williamson, 1976), the sale of import quotas (Pickford, 1985), the location of noxious facilities (Kunreuther & Kleindorfer, 1986), and the allocation of airport slots (Rassenti, Smith, & Bulfin, 1982).

Most studies have paid less attention to the design and execution of mechanisms by focusing instead on why and when services are contracted out. The purpose of this essay is to identify how both the theoretical development of bidding mechanisms and past empirical studies of these mechanisms' performance depart from more traditional studies of contracting out. This essay does so by building on contemporary models that see the design and execution of bidding systems in terms of a bid-taker and bidders who interact strategically. Specifically, I focus on the mechanisms that a bid-taker can use for the sale or purchase of an asset, good, or service: A bid-taker selling an item receives a number of offers (simultaneously or in succession) by potential purchasers until the bid-taker accepts the highest and final bid; a bid-taker purchasing items (or services) receives offers, accepting the lowest and final bid. Traditional studies argue that bidding is justified only under certain combinations of the transaction and production costs associated with different systems (e.g., Globerman & Vining, 1996). But those studies do not speak to the great variety of ways governments can implement policies requiring bidding and auctions.

Bidding and auctions are defined by how a bid-taker interacts with bidders, and that interaction determines the effectiveness of the bidding process. I draw on the theoretical and empirical development of bidding systems and pay particular attention to experiences from bidding processes that involve the government as a bid-taker. I assess four principles that flow from this theoretical and empirical study: that the credibility of the system is a major determinant of its effectiveness; that governments must consider how bidders can engage in collusive behavior, entry deterrence, and predation; that governments must set correct reserve prices; and that it is difficult to build effective systems when bidders hold different levels and types of information. One starting point is that the bid-taker tries to maximize the benefits of using bidding by considering how bidders will respond to the design of the bidding process. In the context of these principles, I focus on what is unique when governments act as bid-takers and design bidding systems. I also address recent calls for the expanded use of dynamic pricing in government.

One important implication is that we have an incomplete theoretical and empirical understanding of how government officials choose the mechanisms they use for selling and purchasing goods and service. We learn about bidding theoretically, empirically, and experimentally. Yet we do not really understand how those systems have been designed, and because those systems help us understand auctions and bidding generally, our knowledge of auctions and bidding in general is incomplete. Specifically, bidding theory presumes that bid-takers are maximizing expected revenue, and system design is the main tool for meeting that goal (e.g., McAfee & McMillan, 1989). If designers pursue other (political, or even corrupt) objectives then most of the primary lessons of bidding theory are wrongheaded, for the theory is most concerned with revenue maximization. But if governments want contracting out to be effective and efficient, the long-neglected lessons of bidding theory provide a foothold for future studies of how, when, and why public policies fail.

This essay proceeds as follows. The next section describes the theoretical, empirical, and institutional development of this field of study. The section that follows explicates the theoretical and empirical foundations for the four principles described above. Last, I discuss the need for vigorous study of such transaction mechanisms and their use in public policy.

Competitive Bidding and Auctions: Function, Form, and Use

Competitive bidding can take many forms. (3) All bidding procedures and auctions involve sales where an auctioneer (or more generally, a bid-taker) fixes a reserve price, invites bids, and awards the article (or resource or asset) that is being auctioned to the highest bidder. For sales, reserve prices are minimum prices at which the asset will be sold (e.g., for oil leases, the minimum that must be paid for the lease to occur); for purchases, they are maximum prices that the bid-taker will pay (e.g., for defense products, the maximum the bid-taker will pay for any transaction to occur under any circumstances). When firms, governments, and individuals use auctions, they expect potential buyers and/or sellers to compete.

Competitive bidding is often different in practice, but in theory it almost always corresponds exactly to auctions (Rothkopf & Harstad, 1994). Most theoretical development and documentary evidence about competitive bidding come from auctions. The specific auction form (how competitive bidding proceeds) is a fundamental choice in this context because economists usually think of economic problems as having specific structural characteristics (Davis & Holt, 1993, p. 33). Key determinants of the performance of any trading environment are factors like the number of interacting agents (sellers and buyers), the resources those agents bring to the table (their endowments), the information the agents hold, their preferences over the potential outcomes of trades, their costs of participation in trades, and the technologies they bring to the production/trading environment. The rules (or institutions) that govern the trades are also important (Fligstein, 1996; Smith, 1967). "In a loose sense, a market institution specifies the rules that govern the economic interaction; the nature and timing of messages and decision, and the mapping from these messages and decisions to the traders' monetary earnings" (Davis & Holt, p. 33). Thus, the bidding framework (how bidders interact with one another and with the bid-taker) helps determine the outcome of the process.

There is virtually an infinite variety of bidding and auction forms, and it can be difficult to capture the key issues designers must consider when choosing a specific form. However, the choice of an auction type can have important revenue implications for an auctioneer selling an object or a bid-taker making a purchase. A central and basic difference between types is whether bidding involves simultaneous decisions or sequential decisions. For simultaneous decisions, the core question is who proposes the price of the good at auction: the offerer of the good, the bidder, or whether bids and offers are posted simultaneously. For example, posted offer and posted bid auctions differently determine which side of the market can post terms of trade on a nonnegotiable basis. The situation when there is only one offerer, one bidder, and only one unit to be exchanged is like an ultimatum bargaining game (Davis & Holt, 1993, p. 37). (4) Other variants include discriminative, first- and second-price sealed-bid, and clearinghouse auctions. The U.S. government uses discriminative auctions for maximizing revenue from the sale of treasury bills. One example of how theorists have addressed the implications of mechanism design in the case of simultaneous bidding systems is Vickrey's proof of the revenue neutrality of first-price and second-price sealed-bid auctions: The bid-taker can do no better under the first-price auction (no bidder is deterred from bidding their true valuation out of the fear that they will pay too high of a price) (Vickrey, 1961).

Examples of mechanisms that involve sequential decision making by bidders and offerers are offer, bid, and double auctions. Bidders may accept a "standing offer" (the most attractive offer on the table at that moment) and offerers can accept a standing bid; sometimes an "improvement rule" is used so that new bids must be greater than the standing bid (likewise for new offers and the standing offer). In sequential bidding systems, bidders would like to know whether other bidders are continuing to pursue the good, service, or contract at block; systems vary in how they handle the release of that information (examples include English, Japanese, and Dutch auctions) (Harstad & Rothkopf, 2000). Again, the structure of these bidding systems has important implications for the strategies used by bidders. For example, in a Dutch (or open first-price) auction, bidders face a decreasing series of prices called out by the auctioneer and the first bidder who stops the series of called prices is awarded the asset; in open second-price auctions, prices paid for an asset for sale ascend with successive bids and the winner often pays only a marginal amount over the valuation of the second-highest bidder. In both auctions, though, the winning bidder may overestimate the value of the asset and overpay for it, suffering the "winner's curse." Specifically, if all bidders value an item (or contract) the same, but they do not know its market value at the point they submit bids, then each is forced to independently estimate the value of the item before submitting a bid. If the highest bidder wins the item (or contract), given that all bidders value the item the same, what distinguishes their bids is the estimates they made of the item's worth (bids they made under incomplete information). The winner's bid is highest because that bidder constructed the highest estimate. But all bidders value the item (or contract) the same (e.g., a "common value auction"), so the winning bid is based on an overestimate of the item's value, and as a result, the winner has overbid for the item. In contrast, in a simultaneous (sealed-bid) auction the highest bidder wins but only pays the price offered by the next-highest bidder.

How can we classify different types of bidding systems? Some emphasize the ascending or descending nature of auctions as a way of discriminating between the market power of the bidder or bid-taker; as this article does, others emphasize the sequential/simultaneous nature of the exchanges between bid-takers and bidders. More importantly, for many types of bidding systems, the differences may be illusory because many systems provide the exact same revenue to the bid-taker (Vickrey, 1961); this is the well-known Revenue Equivalence (RE) Theorem. (5)

However, the RE Theorem makes key assumptions about the bidding environment. First, bidders are risk-neutral and hold independent signals about the valuation of the asset for sale (or the contract on which they are bidding). Second, and more importantly for this article, the theorem assumes that the bid-taker wants an auction mechanism that produces the greatest revenue (or savings, in the case of contracting out) in this round of bidding. It also assumes that the bid-taker sets an optimal reserve price (a restriction on the range of values), and that no sale occurs below that price. If we relax these assumptions--if these conditions do not hold up in the real world of the auctioneer--then one bidding mechanism does better (in revenue terms) than...

NOTE: All illustrations and photos have been removed from this article.



Looking for additional articles?
Search our database of over 3 million articles.

Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication name or publication date.

About Goliath
Whether you're looking for sales prospects, competitive information, company analysis or best practices in managing your organization, Goliath can help you meet your business needs.

Our extensive business information databases empower business professionals with both the breadth and depth of credible, authoritative information they need to support their business goals. Whether it be strategic planning, sales prospecting, company research or defining management best practices - Goliath is your leading source for accurate information.