PG&E originally intended that Line 401 would transport Canadian gas only to Southern California. When Southern California demand did not fill the pipeline, PG&E looked to Northern California markets. Today Line 401 offers gas transportation service from the California-Oregon border at Malin, Oregon, to Southern California at Kern River Station, the southern terminus, and to Northern California at intermediate points. Coupled with downstream pipeline systems operated by PG&E, SoCalGas, and SDG&E, Line 401 can serve end users in most of California.
The connecting distribution systems operate largely without constraints or bottlenecks. The same is not true for transmission-level alternatives to Line 401. PG&E’s Line 400 parallels Line 401 from Malin to the Antioch terminal. Line 400 has lower embedded costs and lower rates than Line 401. Demand for Line 400 service, driven by Canadian gas supply prices that are lower than competing Southwest U.S. supply prices, almost always exceeds the capacity of Line 400. Correspondingly, interstate pipelines that deliver gas from the Southwest into California now operate at low capacity factors. With Line 400 generally operating full, Line 401 competes directly with Southwest interstate pipelines. California gas supplies do not have the capacity to alter the basic features of this competition.
Marketers now dominate gas sales to noncore end users in PG&E’s service territory. End users are generally concerned with burnertip prices, not gas supply basins or transportation routes. Among noncore customers, only PG&E’s utility electric generation (UEG) department and a few large end users actively purchase gas at supply basins, then arrange for transportation service.
Demand in excess of capacity on Line 400 has led to market responses that vex market participants. In D.94-12-061, issued December 21, 1994, the Commission ordered an RPCA scheme at Malin that allocates to noncore shippers the available pipeline capacity on Lines 400 and 401. The adopted scheme is based on end-use priorities, and continues a "crossover ban" previously ordered by the Commission as an essential element of incremental ratemaking for the new pipeline. Under the crossover ban, quantities of gas transported anywhere on the PGT portion of the expansion project are subject to incremental Line 401 rates in California. Marketers have responded to RPCA rules and the crossover ban by transferring ownership of gas packages upstream from Malin, by direct sales or exchange agreements, and by overnominating daily deliveries into Line 400. There is no consensus among the parties or among pipeline customers on how to resolve RPCA problems.
In its market assessment report, PG&E concludes that regional gas markets are competitive and are becoming increasingly integrated. [ Exhibit 207, Chapter 3C.] According to PG&E, an economic link exists between Canadian and Southwest supply basins, despite their geographic separation. Price changes in Canada or the Southwest are transmitted to the other region through competitive interactions in California, which is the contested consuming market.
Other parties discuss more specific market features in their market assessment reports, which are attached to September 20, 1995, post-workshop comments. Amoco, PGT, and Wild Goose recite problems with the crossover ban, the existing RPCA scheme, overnominations at Malin, and peculiar market rules. CanWest reminds the Commission that gas supplies are developed in British Columbia as well as Alberta, Canada. CIPA notes that PG&E still holds a monopoly on most intrastate transportation service within its service territory. El Paso believes that PG&E has a conflict of interest in operation of Line 401, and that ratepayers are harmed by the crossover ban. PG&E and Edison claim that Canadian competition has lowered overall gas prices in California, despite market problems.
PG&E sets prices for as-available service on Line 401 based on competitive alternatives at Topock, Arizona, the principal receipt point for Southwest gas that enters California. In review of the Gas Accord and other issues in this proceeding, we should examine PG&E’s market power, now and under the Gas Accord and other future ratemaking scenarios. We define market power as the ability to sustain revenues, through increased prices or sales, above competitive levels for a significant period of time.
There is much information in the record about PG&E’s market behavior, but we will endorse no single measure of market power. Instead, we begin by looking at five characteristics of PG&E’s participation in gas transportation markets: (1) sufficiency of supply and transportation alternatives, (2) assured sales, (3) the Herfindahl-Hirschman Index (HHI), (4) mitigation and regulation effects, and (5) geographic constraints. PG&E asserts that it has little market power because it cannot sustain control over gas prices at Topock. Whether that single statement is true or not, we must take a broader view of possible market power. PG&E holds virtual monopoly power over intrastate transportation in Northern California.
PG&E claims that it acts as a price follower when it sets Line 401 rates because PG&E has no ability to control market prices. According to PG&E, SoCalGas is the price leader at Topock. PG&E recites several supply alternatives for noncore end users: Southwest gas transported on the El Paso, Kern River Gas Transmission Company, and Transwestern pipelines; California gas; and gas withdrawn from storage. However, PG&E sets Line 401 prices based on only one of those alternatives--El Paso deliveries to Topock. This competition between only two supply sources suggests that PG&E might have significant market power.
On the other hand, the capacity of Line 401 is less than the difference between total interstate capacity into California and typical total demand. There is sufficient overall pipeline capacity that PG&E is assured of only limited sales of Line 401 capacity. By itself, this factor indicates that PG&E might not have significant market power.
The HHI is a measure of market concentration frequently used to assess competitive effects of mergers and acquisitions. The index does not predict anti-competitive behavior by a firm, but is a measure of the number of active participants in a market. For example, the HHI for interstate transportation of Southwest gas into California during 1995 was approximately 0.44, indicating 2.3 effective competitors in that limited market. [ Recorded 1995 data taken from the "1996 California Gas Report," p. 19. At the border, SoCalGas transported 63.5%, PG&E transported 14.5%, and nonutilities transported 21.6% of Southwest gas delivered to California. The calculated HHI assumes four or five nonutilities, and includes Mojave pipeline gas. The number of effective competitors is the inverse of the HHI.] Looking only at this measure, we would conclude that SoCalGas and PG&E are dominant players at Topock. [ Issues relating to market power for SoCalGas will be examined more closely in A.96-10-038, the merger application of Pacific Enterprises and Enova.]
Market power can be mitigated by regulation, but individual circumstances must be reviewed carefully. Regulation now has little impact on price competition between Line 401 and PG&E’s Line 300, which delivers gas from Topock to PG&E’s service territory. The lower limit for Line 401 prices is the cost of original system backbone facilities plus $0.02 per decatherm (Dth). [ D.94 - 02 - 042, 53 CPUC2d 215, 239 (1994).] This leaves PG&E much latitude for discounting below the tariff rate of approximately $0.48/Dth. Service on Line 300 is sold at tariff rates; delivered gas costs are determined by upstream costs of Southwest gas and interstate pipeline service to the border. Incremental interstate service is typically over the El Paso pipeline using capacity that is under contract to PG&E but is not used by PG&E customers. PG&E sells that excess capacity under its capacity brokering program. PG&E sets minimum bids for brokered capacity, but claims that actual prices are often negotiated downward to rates lower than the posted minimums. Commission regulation includes reasonableness review of the negotiated transactions, as part of this proceeding, but such retrospective review has little effect on PG&E’s market power. Taken as a whole, there seems to be little regulatory mitigation of PG&E’s potential market power at Topock.
In times when gas markets were isolated and regional, geographic constraints enhanced utility market power. Today we share PG&E’s expectation that national gas markets will become increasingly integrated. Nonetheless, geographical factors have led to the emergence of Malin and Topock as the two principal entry points for transportation of gas into California. To a certain extent, geography has caused the present constraint on Line 400. We cannot simply find that increasing market integration prevents PG&E from exercising market power.
We draw no firm conclusions about PG&E’s market power from the above simple measures of market behavior. We must dig deeper. In doing so, we should keep in mind the relationships among gas supply, transportation, and distribution costs. Currently, procurement costs are roughly $2.20/Dth, and local transmission and distribution costs are in the range of $0.75/Dth for noncore customers to $2.65/Dth for core customers, exclusive of public purpose and balancing account charges. By comparison, Line 401 firm service tariff rates are approximately $0.48/Dth, and as-available service is discounted below that. Interstate pipeline costs for Southwest gas are scarcely above variable costs, in the neighborhood of $0.10/Dth. The transportation rates disputed in this proceeding are important, but they are only a small fraction of burnertip gas costs. Therefore, the effects of gas transportation ratemaking on supply competition and California’s pipeline infrastructure are crucial to our deliberations.
PG&E and El Paso provide the best evidence on utility market power. PG&E makes many arguments about competition and pipeline markets, but they can be reduced to six principles. First, according to PG&E, markets are workably competitive if actual prices are substantially the same as prices that would result from full competition. No single party holds the power to control prices in the market. Second, PG&E cannot control prices or flows of gas at the California border, specifically at Topock or Malin. Third, theoretically, the existence of two market participants produces competition because one party can undercut prices that are set artificially high by the other party. In this way PG&E and SoCalGas compete against each other for sale of brokered interstate capacity into Topock. Fourth, gas supply competition in Alberta and burnertip competition in the end use market in California eliminate the possibility of market power in the transportation corridor between the two locations. Fifth, increased supply costs in Alberta caused by increased gas demand in California--enabled by construction of the expansion project by PG&E and PGT--are mitigated by consequent increased drilling and production in the supply basin. Sixth, overall gas cost reductions achieved in California subsume customer costs for new pipeline capacity. PG&E claims that California gas costs have dropped by $1.3 billion in the two years since Line 401 has gone into service, and costs in PG&E’s service territory have dropped by more than $500 million.
El Paso concludes that PG&E does have market power at Topock. El Paso believes the gas transportation market there fits the "dominant firm/competitive fringe" model. One or several firms are dominant price setters in the market, and other, smaller players operate within the fringe of the price-setting behavior of the dominant firms. In this case, SoCalGas and PG&E are the dominant firms. According to El Paso, these circumstances inevitably lead PG&E to use its market power in setting Line 401 prices. The effectiveness of PG&E’s pricing strategy confirms that PG&E holds market power. El Paso believes that PG&E’s minimum bids for brokered capacity held on the El Paso pipeline allow PG&E to control Topock prices and thereby control market rates for Line 401 capacity. El Paso criticizes PG&E’s calculation of gas cost savings since Line 401 went into service, claiming that the observed cost reductions are due to factors like lower Canadian and San Juan basin supply prices and lower upstream pipeline costs. Most of PG&E’s calculated cost savings began at least one year after Line 401 went into service. El Paso believes that PG&E’s expansion project has caused at least $289 million in excess pipeline demand charges.
We will not make a finding of fact that the transportation market at Topock follows the dominant firm/competitive fringe model strictly, but in our judgment that model is the best description of market dynamics there. PG&E’s theoretical model of two-party competition is too limited. SoCalGas and PG&E control dominant shares of incoming interstate capacity, at least until their various contracts with interstate pipelines expire. Several factors give the utilities incentives to exercise price leadership at Topock. The market is concentrated, interstate pipeline capacity is in part substitutable, pipeline cost functions are similar, there are barriers to market entry, and overall demand for capacity is relatively inelastic. Price leadership is not necessarily collusive, but it gives SoCalGas and PG&E the opportunity to coordinate their behavior in ways that can lead to higher than competitive prices.
We do not endorse PG&E’s theory that supply basin competition and burnertip competition are sufficient to preclude market power in the transportation corridor between Canada and California. Because there are few supply alternatives to Canadian gas, and transportation costs are not large relative to fundamental supply price differences between Canada and the Southwest, PG&E may hold enough market power to limit end user access to the supply price benefits of Canadian gas.
Considering all the evidence before us, we find that PG&E does hold market power at Topock and within California. PG&E may not be able to control gas prices at Topock, but to a substantial degree it can control flows through Topock and can sustain flows and therefore revenues on Line 401.
Footnotes are bracketed and in blue