«The Effect of Incumbent Bidding in Set-Aside Auctions: An Analysis of Prices in the Closed and Open Segments of FCC Auction 35 Peter Cramton ...»
With the passage of the 1994 amendments to the Communications Act of 1934, Congress instructed the Commission to pursue a combination of objectives when allocating spectrum, including a diversity objective. In particular, Congress required the Commission to adopt spectrum allocation procedures that would create opportunities for entrants in the wireless telecommunications industry (H.R. Rep. No. 111, 103d Cong., 1st Sess., at 254 (1993)). The Commission satisfied that mandate, in part, by setting aside the C and F blocks (two of the six blocks of PCS spectrum) for “entrepreneurs.” A central purpose of Congress’s diversity mandate was to promote greater competition—more carriers in the wireless industry would likely produce lower prices and greater choice for wireless consumers. In addition to using set-asides to promote competition, the Commission placed restrictions on the total amount of spectrum any single carrier could own in a given license area.
Although competition in the mobile telephony sector has increased significantly since the first broadband PCS auction,8 roughly 20 percent the available spectrum was tied up in bankruptcy proceedings for several years (FCC’s Fifth CMRS Report). In the original C-block auction, the FCC offered to finance the winning bids of entrepreneurs (including NextWave’s $4.7 billion of winning bids) with a generous payment schedule. In particular, the FCC allowed winning bidders to make interest-only payments, at a below market interest rate, for six years and payments of principal and interest amortized over the remaining four years of the license term (FCC’s Sixth Report, July 18, 1995). Certain bidders exploited the FCC’s rules by bidding beyond their means, and then declaring bankruptcy when the payments came due. Therefore, rather than adding a new competitor in the wireless sector, the FCC’s policy managed to retain the status quo.
The original C-block auction was an embarrassment for the Commission. Although it felt compelled to devise some form of a diversity program, the Commission would not err in the same direction. Auction 35 was a re-auction of spectrum that was previously auctioned in the C- and Fblock auctions and subsequently confiscated by the Commission when NextWave and other winning bidders defaulted on payment obligations. In Auction 35, the Commission classified the returned spectrum into large markets (cities with over 2.5 million residents) and small/medium markets (cities with less than 2.5 million residents).9 In small/medium license areas, the Commission divided the returned spectrum into three parts: two parts for “entrepreneurs” (the “closed” auction) and one part for incumbents (the “open” auction). In large license areas, the Commission set aside only one-third of the spectrum for entrepreneurs.
To determine an applicant’s eligibility to bid in the closed auction—that is, to determine whether an applicant was genuinely an “entrepreneur”—the Commission adopted a “controlling interest” standard (FCC Public Notice, September 6, 2000). In particular, the controlling interest standard prohibits any applicant who is controlled by a firm with either significant assets (in excess of $500 million) or significant revenues (in excess of $125 million over the past two years) to bid on closed licenses.10 The FCC’s control standard would presumably prohibit the participation by a large, incumbent carrier such as AT&T.11 Although the FCC set aside licenses to “entrepreneurs” who would not have to compete with major phone companies, it revised its rules in August 2000 so that these entrepreneurs could “transfer control of C Block licenses to a non-entrepreneur” (FCC’s Sixth Report, August 29, 2000). The Commission rules provided both for findings of de jure control—more than 50 percent of an applicant’s voting stock is owned by a company—and for findings of de facto control (FCC Public Notice, October 5, 2000). Some incumbent wireless carriers such as AT&T exploited the Commission’s case-by-case process for determining de facto control by forming a new entity, Alaska Native, in which it appeared to own a “non-controlling” interest but in fact it owned a significant share (nearly 80 percent) of the equity. Because the FCC does not, as a matter of policy, screen applicants for eligibility before the auction begins, AT&T effectively gained access to the closed auction. The FCC defends its policy of reviewing eligibility status after the auction ends by suggesting that auctions would never begin if bidders could challenge the eligibility of rivals before the auction.12
2.2. THE CREATION OF ALASKA NATIVE In its effort to qualify for designated entity status prior to Auction 35, Alaska Native represented that an entity known as “Council Tree Alaska Native Wireless, L.L.C.” owned 60.1 percent of its shares, while an entity known as “AT&T Wireless PCS Interests, L.L.C.” owned the other 38.2 percent. AT&T Wireless Interests, however, held debt issued by Alaska Native that was convertible to equity in Alaska Native at its option. Counting this interest as an equity interest, AT&T Wireless Interests in fact owned 79.4 percent of Alaska Native’s shares. AT&T Wireless Interests was wholly owned by an entity known as “AT&T Wireless PCS, LLC,” which in turn was wholly owned by AT&T Wireless, at the time the third largest wireless carrier in the country and a subsidiary of AT&T. Thus, through shell corporations, AT&T owned 79.4 percent of Alaska Native’s shares. AT&T thus had effective control over ownership of Alaska Native.
With these complex corporate and financial artifices, AT&T gained entry to the closed auction and as discussed later in the paper, even obtained bidding credits reserved for very small businesses participating in the open auction. In the aftermath of Auction 35, AT&T has admitted as much. In a series of securities filings, it has disclosed that its ability to conduct its own wireless telecommunications operations depended significantly on Alaska Native obtaining licenses in the closed auction.13 According to company officials, “Alaska Native gives us greater flexibility for bidding on the spectrum that is available. Short of buying Alaska Native outright, AT&T could benefit by negotiating a low-cost ‘roaming’ agreement with the upstart” (Wall Street Journal, January 5, 2000: B1).
Similarly, Council Tree officials have admitted that they had no intention of actually operating a wireless telephone network, and instead view their participation in Alaska Native simply as an “investment.” In this regard, they have a right after five years to require that AT&T or its affiliates purchase for as much as $1 billion the equity interests in Alaska Native previously acquired by them for a small fraction of that amount.14 According to a recent AT&T Wireless securities filing, the acquisition by AT&T Wireless of Council Tree’s minority interest may take place even sooner if AT&T Wireless can satisfy rules governing the transferability of wireless telecommunications licenses.15 As one securities analyst observed shortly after Auction No. 35 had concluded: “The bigger entities put up the capital and the minority entities put up their eligibility” (Bloomberg News, February 27, 2001: 1). Furthermore, AT&T provided $2.6 billion of the $2.9 billion that Alaska Native bid on licenses in Auction 35 representing nearly 90 percent of Alaska Native’s financing.16
3. THE RESULTS OF FCC’S AUCTION 35 This paper begins with a review of AT&T’s impact in the closed auction. Next, the results of the open auction are reviewed.
3.1. ALASKA NATIVE’S IMPACT IN THE CLOSED AUCTION Alaska Native dominated the closed auction, where it bid approximately $2.6 billion to win 29 closed licenses. In the following sections, this paper describes how AT&T’s participation in the closed auction likely harmed legitimate entrepreneurs.
3.1.1. Markets in Which Alaska Native Won the License Alaska Native won 29 closed licenses that were set apart for small entrepreneurs. Because of the company’s substantial financial assets and its backing by AT&T, Alaska Native continued bidding on those closed licenses well beyond the “reservation price” of the qualified DEs.
Consequently, Alaska Native was the dominant bidder in the closed segment of Auction 35. As Table 1 shows, Alaska Native won 29 closed licenses covering 60.3 million people.17 Because the coverage of all licenses in the closed auction was 167.1 million, Alaska Native won approximately 36 percent of the set-aside licenses on a population-weighted basis. Because the total revenues raised in the closed auction were $5.3 billion, Alaska Native won approximately 50 percent of the set-aside licenses on a value-weighted basis (FCC Public Notice, January 19, 2001).
---------------------Place Table 1 here
---------------------By winning closed licenses specifically set aside for legitimate entrepreneurs, Alaska Native undermined the FCC’s diversity objective, which was to allow smaller companies to acquire spectrum and compete against incumbents in the wireless industry.
3.1.2. Markets in Which Alaska Native Was the Last Bidder to Drop Out The FCC auction continues as long as a bidder is willing to bid higher on one or more licenses. As soon as the price of a license reaches or exceeds the reservation price of one of the two remaining bidders, that bidder drops out and the competition for that license ends. Therefore, the last bidder to end the bidding for a license sets the price of the license. By submitting bids on closed licenses that it eventually does not win, Alaska Native inflated the prices for legitimate entrepreneurs that ultimately won those licenses. Assuming AT&T’s presence did not significantly change market structure, if Alaska Native had not participated in the closed auction, the qualified DE with the second-highest reservation price would have set the price of the license in each market.
Table 2 illustrates the results for those markets in which Alaska Native was the last bidder to drop out.
---------------------Place Table 2 here
---------------------As Table 2 shows, Alaska Native set the price in fourteen closed license areas. Alaska Native placed second-to-high bids totaling $296.1 million for licenses in those markets. The table also lists, for each market, the final bid by the last legitimate entrepreneur before Alaska Native to drop out of the auction. Legitimate entrepreneurs placed third-to-high bids totaling $257.7 million in those license areas. By this measure, Alaska Native’s participation in the closed portion of the auction increased prices paid by winning bidders by $38.4 million (equal to $296.1 million less $257.7 million).
3.1.3. Markets in Which Alaska Native Induced a Value-Seeking Firm to Bid in a Second-Tier Market In addition to increasing directly the prices of closed licenses, Alaska Native’s participation in the closed portion of the auction indirectly increased prices for markets in which Alaska Native never bid. In particular, Alaska Native’s participation induced some value-seeking firms to abandon their plans to acquire spectrum in first-tier markets of the closed auction and to start bidding in second-tier markets instead. For example, because of the participation of Alaska Native in first-tier markets, value-seeking firms like Leap, Northcoast, and 3DL were forced to refocus their interests in the second-tier closed markets, which, for those bidders, were not as valuable. An analysis of the bidding reveals that Leap diverted its interests toward San Antonio and Austin (away from Dallas and Houston), Columbus (away from Cincinnati), and Providence (away from Boston). Leap would place less value on these second-tier markets either because they have lower population density than Leap’s preferred markets, or because they do not fit as well with Leap’s current footprint. A press statement from Leap’s chief executive officer released
shortly after the auction further supports the value-seeking strategy employed by Leap:
We used the auction as an efficient, selective way to target the right markets at the right price. Because of our disciplined bidding, we achieved the lowest average price per POP among the ten most active bidders in the auction. This supports our goal to be a wireless carrier with one of the most efficient cost structures in the country (PR Newswire,
Shortly before the auction ended, a Leap senior vice president suggested that “the company will leave the auction with far less spectrum than it wanted because competition from entrepreneurs backed by big companies has driven up prices” (Wall Street Journal, January 5, 2001: B1).
Similarly, Northcoast Communications did not prefer to bid in second-tier closed markets.
Alaska Native’s participation in the auction for closed licenses in first-tier markets inflated the prices in those markets and forced Northcoast to redirect its interests to second-tier spectrum.
Like Leap, Northcoast followed a value-seeking strategy by trying to acquire first-tier licenses at affordable prices in the closed auction. Northcoast’s reluctance to bid in second-tier licenses is evidenced by its late entry in second-tier markets.
Table 3 lists each of the markets in which Northcoast bid, the round in which it placed its first bid, the population of the market, and whether Northcoast ultimately won the license. As Table 3 illustrates, Northcoast placed its initial bid on most second-tier licenses well after round fourteen of the auction.
---------------------Place Table 3 here
---------------------The mean population of the markets in which Northcoast bid before and after round fourteen are computed. The mean population of the markets in which Northcoast bid by round fourteen were more than four times as large as the markets in which Northcoast placed its first bid after round fourteen. The markets in which Northcoast bid before round fourteen were classified as first tier markets by the FCC (population larger than 2.5 million), while the markets in which the company expressed interest after round fourteen were second-tier markets (population less than 2.5 million) (FCC Public Notice, September 6, 2000).17 Such evidence suggests that Northcoast bid for second-tier licenses only after its most desired market became too expensive.