Comments Of The United States Department Of Justice In Response To Notice Of Technical Conference

COMMENTS OF THE UNITED STATES DEPARTMENT OF JUSTICE
IN RESPONSE TO NOTICE OF TECHNICAL CONFERENCE

Communications with respect to this document should be addressed to:

July 30, 1992

COMMENTS OF THE UNITED STATES DEPARTMENT OF JUSTICE
IN RESPONSE TO NOTICE OF TECHNICAL CONFERENCE

On February 26 and 27, 1992, the Commission issued its Notice of Technical Conference to consider generic market based rates for oil pipelines and to examine the potential for streamlining its regulation of the interstate oil industry. The conference was held on April 30, 1992. At the close of the hearing, the Commission invited participants to submit response to questions raised at the hearing.

The Department of Justice ("the Department") strongly supports the Commission's efforts to rely on competition to set rates to the extent allowed by governing statutes. The Commission has implemented market-based prices in the electric and gas industries. A generic reexamination of oil pipeline ratemaking in light of the initiatives in other regulated industries is likely to result in a more rational and efficient regulation of oil pipelines.

In response to legislative proposals to deregulate oil pipelines, in 1986 the Department released a report analyzing the need for continued federal regulation.⁽¹⁾ In recent years the Commission and parties before it have relied upon the 1986 Report to support their positions in individual rate cases. The notice of the conference inquired whether the 1986 Report may be implemented administratively under existing law. The Department submits the following comments in an attempt to clarify the purpose, premises, and methodology underlying the 1986 Report. While the Commission must regulate within the confines of existing law, we believe that the Commission has considerable latitude to implement efficient regulation. While written from the perspective of new legislation, the 1986 Report, in our opinion, contains much economic analysis that could help the Commission increase the efficiency of oil pipeline regulation.

The 1986 Report made three major recommendations. First, it recommended that all crude oil pipelines in the United States, except the Trans-Alaska Pipeline System, be deregulated. Second, it recommended that all but eleven pipelines transporting refined petroleum products (products pipelines) be deregulated. Continued regulation was recommended for five products pipelines, and no conclusion was reached for six others. Third, it recommended that all new crude and products pipelines be unregulated. The pipelines recommended for immediate deregulation by the 1986 Report represented roughly two-thirds of all oil pipelines (measured in terms of barrel-miles) in the lower 48 United States.

The 1986 Report analyzed the desirability and feasibility of deregulation of both rates and access obligations of oil pipelines through adoption of regulatory reform legislation. The 1986 Report employed the general principle that Congress should impose regulation only when the benefits outweigh the costs. The Department used as it standard the effect of deregulation on total economic welfare, i.e., the sum of consumer and producer surplus. Thus the effect of deregulation on pipeline rates was not the primary consideration in the 1986 Report. It is possible that deregulation might raise transportation rates without reducing economic welfare. One example could be the deregulation of a monopoly crude oil pipeline that supplies a local unregulated monopoly refinery. Bilateral negotiation between the refinery and the deregulated pipeline could, by implementing efficient two-part tariffs, prevent any potential efficiency loss resulting from deregulation. The average transportation rate charged by the deregulated pipeline could increase, however. The Department's recommendation were not affected by such wealth transfers resulting from deregulation as long as the deregulation did not reduce economic welfare. It is important to recognize that while the Department used an economic welfare standard in its 1986 Report, it may not necessarily use the same standard to evaluate mergers under the Merger Guidelines.

Another important premise of the 1986 Report is that it considered only two regulatory/deregulatory alternatives: either 1) existing cost-of-service, common carrier regulation, including existing rates, or 2) deregulation with respect to both rates and access provisions. If, contrary to this premise, regulatory reform reduces the costs of regulation, such as by the adoption of incentive regulation and deregulation of rates in competitive markets, or if statutory requirements reduce the benefits of deregulation, such as by the continued imposition of common carrier obligations for competitive pipelines, then the analysis of the desirability of deregulation could change, including changes in the appropriate market concentration thresholds for recommending deregulation. Furthermore, the 1986 Report used existing rates, and not some competitive, or ideally regulated, rates as the benchmark for market delineation and for addressing the issue of how deregulation would affect economic welfare. Any change in the benchmark level of pipeline rates could affect the analysis and recommendations for the need of continued economic regulation of oil pipelines.

Finally, the 1986 Report considered the question of deregulation at the level of an individual pipeline system rather than at the level of the various markets served by the pipeline. The reason for this was that under cost-of-service regulation, the costs of serving one market cannot easily be separated from those serving another; thus, regulation of only certain pipeline markets would raise difficult cost-allocation problems. In particular, partially deregulated pipelines might attempt to shift costs of serving deregulated markets to regulated monopoly markets. To the extent that such strategy were successful, a pipeline would be able to circumvent cost-of-service regulation. If traditional cost-of-service regulation were replaced by an incentive regulation plan (e.g., price-cap regulation, stand-alone cost regulation, or yardstick regulation) that did not tie a pipeline's regulated rates to its accounting costs, it might be feasible to deregulate a pipeline in its competitive markets, while continuing to regulate in non-competitive markets. Market-by-market deregulation would have required, however, investigation to delineate additional geographic markets. Since the 1986 Report envisioned that pipelines, and not markets, be deregulated, the Report did not generally delineate all of a pipeline's markets once it had identified a market that warranted the pipeline's continued regulation.

The rest of these comments describe three types of situations in which the 1986 Report recommended that particular oil pipelines be deregulated. The first consisted of situations in which market outcomes are reasonably efficient, particularly in light of the cost of regulation. Such situations included low market concentration and an absence of an effect on consumer prices. The second consisted of situations in which market power could exist in the oil industry, but the regulation of oil pipelines, without regulation of other sectors of the oil industry, would be ineffective. An example would be the case of a crude oil pipeline that is the only input source for a local monopoly refinery. Rate regulation of the pipeline alone may be inadequate to mitigate any market power that the pipeline and refinery may possess jointly. The third is a situation in which the continued regulation of one pipeline may be rendered unnecessary by the continued regulation of another.

I. Indicators of Efficient Market Outcomes

1. Low Market Concentration
   1. Overview

A pipeline may participate in a number of origin and destination markets. In origin markets, from which the pipeline transport oil, pipeline monopsony power is the potential competitive concern. In destination markets, into which the pipeline transports oil, monopoly power is the concern.

Despite their natural monopoly characteristics, oil pipelines often face competition from other pipelines. As a result of market growth and changes in regional patterns of supply and demand, new pipelines have been constructed to compete with existing pipelines.⁽²⁾ Moreover, since petroleum is largely a homogeneous commodity, a pipeline running from origin A to destination B faces not only competition from other pipelines and other modes of transportation from A to B but also source competition from any pipeline or other facility either out of A or into B. In its origin market, A, the pipeline faces competition from any pipeline or water transportation out of market A. It also competes with any other market participants that process or consume petroleum at A. A crude oil pipeline, thus, competes in an origin market with local oil refineries, and a products pipeline competes in an origin market with local product consumption. In its destination market, B, the pipeline faces competition from any pipeline or water transportation into market B. It also competes with any other market participants that supply petroleum at B. A crude oil pipeline, thus, competes in a destination market with local crude production, and a products pipeline competes in a destination market with local oil refineries. Source competition can often protect a shipper who is tied to a particular pipeline: it could be unprofitable for the pipeline to raise its shipper's rates, if that resulted in the pipeline losing volumes to a competing petroleum source.

The first step in computing market concentration is the delineation of product and geographic markets for oil pipelines. The basic principles that the Department uses to delineate antitrust market boundaries are set forth in the Merger Guidelines.⁽³⁾ They define a market as a product and an area such that a hypothetical monopolist over that product in that area would profitably impose a "small but significant and nontransitory" increase in price. The profitability of a rate increase by a hypothetical pipeline monopolist--and thus the geographic scope of origin and destination markets--depends not only on the specific meaning given to a "small but significant and nontransitory increase in price" but also on factors such as the spatial distribution of consumption (or, where relevant, production), the profit margin of the pipeline, and the cost of local trucking. While the 1986 Report used the Economic Areas of the U.S. Department of Commerce's Bureau of Economic Analysis ("BEAs") as approximate markets for the purpose of organizing pipeline data, it did not consider BEAs to be relevant geographic markets. Throughout its discussions of individual product pipelines, the Report recognized that trucking from sources outside the local BEA was feasible in some cases.⁽⁴⁾

Having delineated markets, the Department computed market concentration. In pipeline markets characterized by low levels of market concentration, the 1986 Report concluded that any economic welfare loss from market power by deregulated pipelines was unlikely to exceed the costs of regulation. The 1986 Report recommended that a pipeline be deregulated if the Herfindahl-Hirschman Index (HHI) of market concentration is 2500 or less in all of its major origin and destination markets. An HHI of 2500 corresponds to a market with four equal-sized firms. The 1986 Report maintained that competition among four equal-sized deregulated firms in any origin or destination market is likely to be more efficient than traditional pipeline regulation in view of the fact that cost-of-service, common carrier regulation imposes significant direct and indirect costs.⁽⁵⁾ If regulatory reform reduces the costs of regulation, such as by the adoption of incentive regulation and deregulation of rates in competitive markets, or if statutory requirements reduce the benefits of deregulation, such as by the continued imposition of common carrier obligations for competitive pipelines, then the appropriate market concentration thresholds for recommending deregulation could change.

1. 2. Computing Concentration of Throughput Capacity

The 1986 Report computed HHIs initially based on pipeline throughput capacity. Under the Report's implicit assumption that terminal capacity can be quickly expanded without significantly increasing unit costs, throughput capacity is the appropriate measure of a pipeline's ability to serve a market. The computation of a market capacity HHI for pipelines is complicated, however, by the possibility of significant surplus pipeline capacity, which can arise for two reasons. First, a pipeline may simply be operating at less than full utilization. The second, and more fundamental, cause of surplus capacity arises from a peculiar characteristic of the pipeline industry. The throughput capacity of a pipeline at a particular city serves not only that city but also any cities further down the pipeline. If a city is located near the origin of a major long-distance products pipeline, the capacity of the pipeline may greatly exceed the city's demand. The presence of such additional capacity can greatly bias upward an HHI based on throughput capacity.⁽⁶⁾ This points up the fact that concentration measures are only a first step in the assessment of the competitiveness of markets.⁽⁷⁾

The 1986 Report addressed the potential upward bias in the HHI by adjusting for surplus capacity. Its method was to allocate total market volume among individual firms so as to minimize a local market volume HHI, subject to the constraint that no firm's assigned volume exceeded its capacity. For example, consider a market in which local product consumption equals 20 MBD. Suppose the market is served by four independent pipelines A, B, C, and D, with throughput capacities of 20, 15, 10, and 2 MBD, respectively. The unadjusted capacity HHI is 3305. The assignment of market consumption that minimizes a volume HHI subject to the capacity constraints is to assign D its maximum volume of 2 MBD, and to split the remaining 18 MBD equally among A, B, and C. The resulting HHI adjusted for surplus capacity is 2800. The adjustment for surplus capacity essentially converts the concentration index from the HHI to an index based on the number of pipelines in a market, except that small pipelines are not given full weight.

The 1986 Report acknowledged that its surplus capacity adjustment may understate concentration in some cases. The problem with the adjustment is that, once the smallest pipeline, D, is assigned an output equal to its capacity, it no longer has any capacity to react to a noncompetitive output restriction. The three other pipelines, A, B, and C, could exercise market power by themselves. While there may be a better way to adjust for surplus capacity in a market, it is unlikely that any of the 1986 Report's recommendations, which were made with respect to an entire pipeline, were dependent on its treatment of surplus capacity.

Any adjustment for surplus capacity for a single market may understate concentration if the surplus capacity is being utilized in other concentrated markets. Therefore, concentration in larger areas comprising several pipeline markets may have to be computed. For example, if 1) each of a small group of pipelines serve each of a group of destination markets, 2) the pipelines serve no other markets, 3) the markets have no other source of supply, and 4) the pipelines have unequal throughput capacities and operate at full capacity, then no adjustment for surplus capacity in any individual market would be warranted--one would have to compute a single HHI for the entire group of markets. The 1986 Report recognized this problem, and calculated multi-market HHIs in its discussion of the Capline and Buckeye pipelines.

1. 2. The Absence of an Effect on Consumer Prices

The 1986 Report used a total economic welfare standard in its analysis of deregulation of oil pipelines. The Report went on to find, however, that deregulation will not generally reduce total welfare unless it lowers the welfare of ultimate consumers, by reducing product consumption. In theory, the exercise of market power by deregulated pipelines could reduce economic welfare by inefficiently reducing not only product consumption, but also oil production and refinery utilization. An example in which pipeline market power could reduce oil production without harming consumers would be a crude oil pipeline that is a monopsonist in a crude production area but competes at its destination with other domestic and imported crude oil. An example in which pipeline market power could reduce refinery utilization without harming consumers would be a crude oil pipeline that is the only source of crude for a number of refineries that compete with numerous products pipelines. In both examples, there is no effect on consumer prices. As a remedy for potential upstream welfare losses such as these, the 1986 Report found that market mechanisms are more efficient than regulation.

The most significant such market mechanism is vertical integration, which is very common in the oil pipeline industry.⁽⁸⁾ Vertical integration ensures that a shipper faces a price of transportation equal to marginal cost. If, for example, each owner's equity share of a monopsony joint venture crude oil pipeline equals at all times its share of upstream crude oil production, there will be no upstream welfare loss, regardless of the pipeline's rate.

The 1986 Report found that market mechanisms other than vertical integration can prevent welfare losses that are purely upstream. For example, the concentration of firms engaged in crude production or refining in any pipeline market is likely to be high. In such cases, negotiations between such firms and oil pipelines may likely yield outcomes that approach the maximization of joint profit. As long as it does not affect consumer prices, the joint profit-maximization by oil pipelines and producers or refiners will be socially efficient. It is unlikely that traditional regulation is more efficient than negotiation as a remedy for potential upstream welfare losses.

The 1986 Report discussed not only product destination markets but also crude origin, crude destination, and product origin markets. In many cases the Report found that such markets were competitive. In other, the 1986 Report found that market forces are more efficient than regulation as a remedy for potential upstream welfare loses that do not affect consumer prices. These include, for example, crude origin markets in central Michigan, the Williston Basin, and the Four Corners area.⁽⁹⁾

II. Indicators that Regulation of Pipelines Alone Will Be Ineffective

There may be instances of potential market power in the oil industry that would not be lessened by the regulation of oil pipelines alone. In light of the costs of regulation on society, the 1986 Report recommended deregulation in some cases where regulation of oil pipelines would be ineffective.

1. 1. Market Power in an Unregulated Sector of the Vertical Supply Chain

Pipelines are the only sector of the oil industry subject to price regulation. If there is an unregulated sector of the vertical supply chain characterized by substantial market power, for example, a monopoly crude producer or a monopoly refinery, regulation of the pipeline serving such monopolist will not by itself mitigate the effects of market power on consumers.

As a possible example, consider Lakehead pipeline. Lakehead is the U.S. segment of a trans-Canadian crude pipeline system from Edmonton to Montreal. The Lakehead segment begins at the North Dakota border and runs through the Great Lakes area to Michigan before running into Ontario. In the Duluth-Superior area, Lakehead supplies the Murphy refinery. Lakehead is the only source of crude oil for Murphy, which is the only refinery in the Duluth-Superior area. The Lakehead-Murphy joint entity may or may not possess market power in the supply of petroleum products in the Duluth-Superior area.⁽¹⁰⁾ But even if it did, regulation of Lakehead alone would not improve efficiency. Any market power possessed by the Lakehead-Murphy joint entity could be exercised alone by the unregulated Murphy refinery. Thus, there is no benefit from the regulation of Lakehead in the Duluth-Superior market.

In theory, one might argue that regulation of Lakehead may prevent potential problems of successive monopoly.⁽¹¹⁾ Lakehead and Murphy can avoid any problem of successive monopoly, however, by negotiating to maximize joint profit. There is little reason to expect regulation to be more efficient than such negotiation.⁽¹²⁾ Alternatively, one might argue that if refining were a competitive activity due to easy entry, then an unregulated Murphy refinery could not exercise market power, and thus deregulation of Lakehead could result in welfare-reducing increases in consumer prices in Duluth. Refining is unlikely to be a highly competitive activity, however, because entry is likely to be difficult. The entry conditions for refineries resemble those for pipelines. There are significant economies of scale, immobile assets, environmental start-up costs, etc.

1. 2. Regulatory Evasion by Vertically Integrated Joint Venture Pipelines

A significant proportion of oil pipelines are vertically integrated joint ventures. Pipeline economies of scale result in the construction of a single pipeline to serve an area, and efficiencies of vertical integration result in vertically integrated joint ventures. As discussed above, vertical integration may help prevent upstream welfare losses. However, vertical integration also can render pipeline regulation ineffectual. This can occur with a joint-venture pipeline, owned by its shippers, that is a bottleneck facility with substantial market power. The vertically integrated shipper-owners can evade existing pipeline regulation and gain monopoly profit by simply setting the capacity of the pipeline at its monopoly level.⁽¹³⁾ The shipper-owners would then gain the monopoly profit at some unregulated sector of the oil industry. Regulating the pipeline's rates would have no effect on consumer prices or industry profit.

The 1986 Report did not use the likelihood for regulatory evasion by vertically integrated joint venture pipelines as a justification for deregulation of any particular pipeline.⁽¹⁴⁾ One might argue that to the extent the correlation between shipments and pipeline ownership is not perfect, regulation might have some beneficial effect under certain conditions. That is, for any shippers that are shipping on pipeline capacity owned by others, the transportation cost for the incremental barrel is the pipeline rate, and not the marginal cost of pipeline transportation. Pipeline regulation should constrain this transportation cost below the monopoly level. Moreover, given the regulated rate, the existence of transportation revenues from non-owner shippers may create a profit incentive for the joint venture pipeline to expand its capacity beyond the monopoly level.⁽¹⁵⁾ Furthermore, one might argue that there may be solutions to regulatory evasion such as vertical divestiture,⁽¹⁶⁾ competitive rules joint ventures,⁽¹⁷⁾ or regulation of other sectors of the oil industry. Such solutions, however, would require government intervention that would impose its own costs on society. In sum, the issue of regulation of vertically integrated joint ventures is very complex. As a practical matter, vertical integration may make unnecessary the regulation of pipeline market power, but it can also thwart attempts at regulation.

III. Deregulation of a Pipeline Conditioned on Continued Regulation of Another

Another basis used by the 1986 Report for the deregulation of a pipeline is the continued regulation of a competing pipeline with sufficient excess capacity. Consider two competing pipelines A and B. Suppose A will remain regulated because it can exercise market power. If A has sufficient excess capacity, it will be unprofitable for B to increase market price by raising its own rate. Thus, B can be safely deregulated.

As an example, the 1986 study discusses Colonial and Plantation as a products pipeline duopoly in the Southeast. Colonial has four times the throughput capacity of Plantation. In the mid-1980s Colonial had sufficient excess capacity to accommodate the entire throughput of Plantation.⁽¹⁸⁾ Furthermore, a significant proportion of Colonial capacity serves New York and other Northeast ports, where Colonial competes with water transportation and local refining. Colonial's capacity to New York could be considered additional excess capacity to the Southeast markets, since Colonial shipments to New York could be diverted to the Southeast and be replaced at New York by water shipments. This has given Colonial significant excess capacity to the Southeast. Such regulated excess capacity of Colonial could serve as a "competitive fringe"; it would likely be unprofitable for a deregulated Plantation to increase market price by raising its rate since Plantation would lose volume to Colonial. Thus, the 1986 Report recommended deregulation of Plantation.

IV. Concluding Comment

These comments have laid out the basic methodology underlying the Department's 1986 Report. The emphasis of these comments has been on the technical issues raised by the Report. The Department believes that fundamental principles of economics coupled with its experience in analyzing numerous and diverse industries strongly indicate that a substantial number of oil pipelines in the U.S. could be deregulated without any harm to consumers and with potentially substantial gains to the economy as a whole.

July 30, 1992

FOOTNOTES

1. Oil Pipeline Deregulation, Report of the U.S. Department of Justice (May 1986) (hereafter cited as 1986 Report). See also Charles Untiet, "The Economics of Oil Pipeline Deregulation: A Review and Extension of the DOJ Report," U.S. Department of Justice, Antitrust Division, Economic Analysis G roup Discussion Paper, EAG 87-3, May 22, 1987. The 1986 Report did not analyze pipelines for natural gas liquids and anhydrous ammonia. In 1989, in response to a request by the Senate Committee on Appropriations, The Department prepared a report on competition in anydrous ammonia, natural gas liquids, and propane pipelines. The 1989 report did not make any final recommendations with regard to deregulation of particular pipelines.

2. The historical development of the Plantation and Colonial pipelines is worth noting in this regard. Plantation was original built in 1941 as a 12-inch diameter line running from Baton Rouge to Greensboro, N.C., with throughput capacity of 60 thousands of barrels per day (MBD). In 1950, an 18-inch diameter, 100 MBD capacity parallel trunk line was added. In 1963-64, Colonial began service with a 36-inch diameter, 792 MBD capacity main trunk line from Houston to New York. In 1967-71, Plantation added a 30-inch diameter parallel trunk line and a 14-inch diameter extension from Greensboro to Washington, D.C.; today the 18-inch and 30-inch diameter Plantation trunk lines have a combined through capacity of 560 MBD. In 1971-80 Colonial added a 36- to 40-inch diameter, parallel main trunk, giving it a total throughput capacity of 2280 out of Houston and 2184 MBD out of Baton Rouge. Arthur M. Johnson, Petroleum Pipelines and Public Policy, 1906-1959, at 320, 364 (1967); George S. Wolbert, Jr., U.S. Oil Pipe Lines, 65 (1979); and Colonial Pipeline Company, "The Quest for Excellence, 1962-1987" (1987).

3. Horizontal Merger Guidelines, U.S. Department of Justice and the Federal Trade Commission, April 2, 1992.

4. In formulating its recommendations for the destination markets of products pipelines, the 1986 Report treated a pipeline (or group of pipelines in the same location) as having no other competitors in its (their) relevant market if there were none within 125 miles. This could correspond to a threshold price increase of roughly one cent per gallon in the case of motor gasoline and distillate. See Charles Untiet, "The Economics of Oil Pipeline Deregulation: A Review and Extension of the DOJ Report," U.S. Department of Justice, Antitrust Division, Economic Analysis G roup Discussion Paper, EAG 87-3, May 22, 1987. In the case of jet fuel, to the extent that its spatial distribution of consumption is more concentrated and its cost of local trucking is higher than that of motor gasoline and distillate, and to the extent that deregulated pipelines may price discriminate against shippers of jet fuel, the geographic scope of pipeline markets for jet fuel could be smaller than that for motor gasoline and distillate.

5. A study of NERA sponsored by a group of independent refiners concerned about deregulation also argues that the 2500 threshold is appropriate. National Economic Research Associates Inc., Competition in Oil Pipeline Markets: A Structural Analysis 94-95 (1983). The proposed 2500 HHI threshold exceeds the 1800 HHI threshold that defines "highly concentrated" markets for the purposes of merger assessment. See Horizontal Merger Guidelines, at 16. Since deregulation and the competitive assessment of mergers raise potentially different policy issues, the two thresholds need not coincide.

6. To illustrate, consider the Explorer products pipeline from the Gulf Coast to Tulsa, St. Louis, and Chicago. In the Tulsa area, Explorer competes with three local refineries. The capacities (in MBD) of the four supply facilities in the Tulsa area are: Explorer--380, Conoco--140, Sun--85, and Sinclair--50. The capacity HHI for the Tulsa area thus is 4049. However, consumption in the Tulsa BEA is estimated to be only 55 MBD. Each of the four Tulsa facilities can virtually satisfy local demand by itself. In competing to supply the Tulsa market, the four pipelines are of equal importance, so they should be assigned equal market shares, yielding an HHI of 2500. The inclusion of Explorer's Chicago-bound capacity greatly overstates concentration in Tulsa.

7. See Horizontal Merger Guidelines at 18.

8. For example, in 1984, 13 major integrated oil companies owned 72% of interstate oil pipelines measured in terms of barrel-miles. American Petroleum Institute, Market Shares and Individual Company Data for U.S. Energy Markets: 1950-84, at 93 (1985). The 13 oil companies are Sohio-BP, Chevron, Exxon, Amoco, Texaco, ARCO, Mobil, Shell, Phillips, US Steel (Marathon), Dupont (Conoco), Unocal, and Sun. These companies' combined share of interstate crude oil pipelines alone is likely to be higher than 72% because the largest independent pipelines (e.g., Texas Eastern, Williams, MAPCO, Buckeye, Southern Pacific, Kaneb, and Calnev) are primarily products pipelines.

9. See 1986 Report's discussion of Lakehead at 106-08, of Portal and Butte at 109-12, and of Texas-New Mexico at 141-42.

10. Lakehead-Murphy may compete in the Duluth-Superior refined products market with the Williams product pipeline to Duluth and with the potential for refined product receipts at the Lake Superior port of Duluth-Superior.

11. If a downstream monopolist B faces a monopoly input price from an upstream monopolist A, the resulting economic welfare could be less than if A and B merged to form a vertically-integrated monopolist. See, e.g., Frederick R. Warren-Boulton, Vertical Control of Markets 51-55 (1978).

12. Welfare losses from successive monopoly may result in the case of a large upstream monopolist supplying numerous regional downstream monopolists. It may be costly for the upstream monopolist to seek to maximize the joint profit with each of the downstream monopolists. In the oil pipeline industry, however, contractual resolution of successive monopoly distortion is unlikely to be a significant problem because the number of downstream refiners tends to be small.

13. The FERC does not regulate the capacity decisions of oil pipelines.

14. The 1986 Report did note, however, that in the case of Colonial, a joint venture pipeline recommended for continued regulation, the regulatory evasion criterion does not apply due to significant non-owner shipments on Colonial.

15. Robert J. Reynolds and Thomas C. Spavins, "Natural Monopoly and Regulatory Evasion," paper presented at the 1979 meetings of the Eastern Economic Association, Boston, May 10, 1979.

16. See Testimony of John H. Shenefield, Assistant Attorney General, Antitrust Division, U.S. Department of Justice, Before the Subcomm. On Antitrust and Monopoly of the Senate Comm. On the Judiciary, 95^th Cong., 2d Sess. (June 28, 1978), reprinted in Edward J. Mitchell, ed., Oil Pipelines and Public Policy: Analysis of Proposals for Industry Reform and Reorganization 191 (1979); Donald L. Flexner, Deputy Assistant Attorney General, Antitrust Division, U.S. Department of Justice, "Oil Pipelines: The Case for Divestiture," in Mitchell, ed., supra, at 3-13.

17. Competitive rules have been proposed as an alternative to rate regulation for joint venture pipelines. Lucinda M. Lewis and Robert J. Reynolds, "Appraising Alternatives to Regulation for Natural Monopolies," reprinted in Edward J. Mitchell, ed., Oil Pipelines and Public Policy: Analysis of Proposals for Industry Reform and Reorganization 135 (1979); Frederick R. Warren-Boulton and John R. Woodbury, "The Design and Evaluation of Competitive Rules Joint Ventures for Mergers and Natural Monopolies," (1990); and Dan Alger, "Competitive Joint Ventures for Natural Gas Pipelines," (1991).

18. FERC Form 6s for Colonial and Plantation, 1983-85.