Statement of Chairman Robert Pitofsky and
BP/Amoco, Docket No. C-3868
On December 30, 1998, the Commission published a proposed complaint alleging that this merger would violate Clayton Act § 7, 15 U.S.C. § 18, and FTC Act § 5, 15 U.S.C. § 45, in 30 wholesale gasoline markets and nine light petroleum products terminaling markets in the United States, and accepted a proposed consent order resolving those allegations. The Commission has now accorded final approval to the complaint and consent order.(1) Our colleague, Commissioner Swindle, dissents from that portion of the complaint and consent order that alleges violations and mandates relief in 27 of the wholesale gasoline markets.(2) We write to clarify our view.
At the time the consent agreement was accepted for public comment -- before the merger at issue was consummated -- British Petroleum Company p.l.c. ("BP") and Amoco Corporation ("Amoco") were integrated producers, refiners and marketers of petroleum products, including gasoline, in the United States. Although BP's and Amoco's operations did not overlap in many areas,(3) both were wholesale marketers of gasoline in the southeastern and midwestern United States, i.e., both BP and Amoco sold gasoline to retail gas stations that they might or might not have owned. In these markets, BP was the only firm that could sell "BP"-branded gasoline to retail dealers, and Amoco was the only firm that could sell "Amoco"-branded gasoline to dealers. Therefore, measuring concentration of retail sales by brand was an adequate proxy for measuring concentration in gasoline wholesaling.(4)
In 25 metropolitan area markets, absent the relief secured by the Commission, the combination of BP and Amoco would have resulted in a highly concentrated wholesale gasoline market, and an increase in concentration in an amount that the Department of Justice-FTC Merger Guidelines presume likely to create or enhance market power or facilitate its exercise. Merger Guidelines § 1.51(c).(5) In each of these markets, the top four firms would together have had at least 70% of wholesale sales; in 15 markets, the top four firms would have had more than 80%.(6)
Market shares and concentration levels of this magnitude raise antitrust concern because they suggest that a small number of firms might, after this merger, be able to raise price without losing significant sales to what could well be an insignificant fringe.(7) See, e.g., United States v. Rockford Memorial Corp., 898 F.2d 1278, 1283-84 (7th Cir. 1990). Concerns about collusion or coordination, and consequent price increases to consumers, are more pronounced in markets -- such as gasoline markets -- where (among other factors) the product is homogeneous and prices are generally observable, making it relatively easier for a small number of firms to coordinate and to detect deviation.
Of course, high market concentration is less of a threat to consumers if retailers in the market are likely to switch to new sources of supply in the event of a wholesale price increase. But, we require persuasive evidence that entry would be timely, likely and sufficient to defeat a coordinated price increase. Merger Guidelines § 3. Our colleague concludes that such entry could occur, and is likely to occur, "if there are enough branded retail gasoline stations that could switch and become customers of the new wholesale entrant."(8) We do not disagree with this analysis, but we are unpersuaded by the investigative record here that there is a sufficient likelihood that enough switching would occur to allay our concerns. The history of switching in these markets appears to be more among incumbents than to new entrants, and switching among incumbents (particularly among incumbents with substantial market shares) will not defeat a wholesale price increase by those incumbents. Dealers also would be less likely to switch to fringe suppliers or to new entrants if there are significant reasons for dealers to prefer major brands (particularly major brands that are well-established in a given area), such as the benefit of local marketing or of brand credit card programs. Moreover, dealers might not have an incentive to switch to new entrants to defeat a price increase by their suppliers in which they also may profit.
Instead, we believe that the consent order will make jobbers and open dealers able to switch, and by relieving them of financial penalties that might deter switching to new entrants, make it more likely that they will in fact switch, preventing an increase in concentration that otherwise could well give rise to a substantial risk of higher prices for gasoline in the markets alleged in the complaint. As we noted, our disagreement with our colleague is narrow: whether, in the absence of the relief under the consent order, jobbers and open dealers are sufficiently likely to switch in substantial numbers to protect the ultimate consumers from the risks that otherwise would be associated with highly concentrated gasoline markets. In this case, we believe the investigative record regarding dealer switching is insufficiently compelling to demand that ultimate consumers bear the substantial risk of higher prices for gasoline that may result from these highly concentrated markets.
1. In response to comments received during the comment period, the Commission, with the agreement of BP-Amoco, has made a few modifications to the details of the complaint and order. None of these changes, however, alter the core relief.
2. Commissioner Swindle concurs in the complaint and consent order to the extent they alleges that the merger of BP and Amoco would violate the antitrust laws in the nine terminal markets and in wholesale gasoline markets in Pittsburgh, Pennsylvania, and Cleveland, Toledo and Youngstown, Ohio.
3. For example, to a large extent, Amoco and BP produced and marketed different petrochemical products in the United States. BP produced acetic acid and acrylonitrile in the U.S., but Amoco did not. Similarly, Amoco produced ethylene, propylene, polypropylene, and styrene in the U.S., but BP did not. In the few petrochemical areas where the parties overlapped in the U.S., concentration did not change significantly as a result of the merger.
4. Indeed, brand concentration may understate concentration in the wholesale market, because some branded wholesale sellers also supply unbranded gasoline to unbranded retail stations. The brand concentration statistics used here would not attribute these unbranded sales by branded wholesalers to the branded wholesaler.
5. The Merger Guidelines presume anticompetitive effects when the post-merger Herfindahl-Hirschman Index ("HHI") is over 1800 and there is an increase of more than 100 points. HHI is a statistical index that measures the degree of concentration in a relevant antitrust market. Those metropolitan areas and the changes in HHI would have been: Albany, Georgia (post-merger HHI 3674, increase of 542); Charleston, South Carolina (1865/362); Charlotte, North Carolina (1909/610); Charlottesville, Virginia (2214/278); Clarkesville, Tennessee (1863/492); Cleveland, Ohio (1859/124); Columbia, South Carolina (2257/738); Columbus, Georgia (2194/351); Cumberland, Maryland (2592/161); Dothan, Alabama (2259/235); Fayetteville, North Carolina (2635/795); Florence, Alabama (1959/269); Goldsboro, North Carolina (2133/310); Hattiesburg, Mississippi (2214/281); Jackson, Tennessee (2051/508); Memphis, Tennessee (1948/468); Myrtle Beach, South Carolina (2138/353); Pittsburgh, Pennsylvania (2129/663); Raleigh, North Carolina (2032/535); Rocky Mount, North Carolina (2003/302), Savannah, Georgia (2668/515); Sumter, South Carolina (1920/528); Tallahassee, Florida (2366/794); Toledo, Ohio (2022/351); and Youngstown, Ohio (2540/1043).
6. In addition, in five areas the HHI would have increased substantially (by more than 100 HHI points): Birmingham, Alabama (post-merger HHI 1778, increasing by 273); Mobile, Alabama (1600/160); Athens, Georgia (1654/251); Meridian, Mississippi (1705/359); and Hickory, North Carolina (1782/354). In each of these "moderately concentrated" markets, the top four firms would together have had at least 70% of wholesale sales, and independent unbranded sellers would have had less than 20%.
7. In this case, the Commission examined the gasoline markets in which BP and Amoco competed and alleged antitrust violations in markets with a small number of fringe players, and not in markets where fringe competitors collectively appeared to have significant market presence.
8. We all agree that our concerns about concentration among wholesale sellers of gasoline are not obviated by the asserted fact that retailers can set their own prices for retail gasoline sold at their outlets. The wholesale price of gasoline is plainly the most substantial portion of the dealer's cost, and increases in wholesale prices will likely result in increases in retail prices.