Opinion for the Court filed by Circuit Judge TATEL.
TATEL, Circuit Judge.
Three years ago, in Interstate Natural Gas Ass’n of America v. FERC, 285 F.3d 18, 29 (D.C.Cir.2002) (INGAA), we resolved several matters relating to the Federal Energy Regulatory Commission’s efforts to “increase flexibility and competition in the natural gas industry,” but remanded two issues to the Commission for further consideration. The first issue is whether FERC’s pre-granted abandonment scheme “appropriately balance[s] the protection of captive customers with the furtherance of market values.” Id. at 52. Twice the Commission imposed caps on the contract length existing customers must bid to retain service rights in the face of competing bids from new customers. Twice we rejected FERC’s justification for the cap length selected and remanded the issue for further explanation. In the order now before us the Commission removed the cap altogether. The second issue involves FERC’s decision to allow shippers to make what are known as “forwardhaul” and “backhaul” deliveries of gas “to a single point in an amount greater than the shipper’s contracted for capacity at” that point. Id. at 40. On remand, FERC again defended, though in greater detail, its decision to allow these transactions.
Various petitioners now challenge the new order. Finding that the Commission engaged in reasoned decision making with respect to both issues, we deny the petitions for review.
I.
We begin with the matching-term cap. In INGAA we held that section 7(b) of the Natural Gas Act (NGA), 15 U.S.C. § 717f(b), protects long-term pipeline capacity holders by prohibiting “ ‘natural gas companies]’ from ceasing to provide service to their existing customers” when their contracts expire “unless, after ‘due hearing,’ FERC finds ‘that the present or future public convenience or necessity permit such abandonment.’ ” INGAA, 285 F.3d at 51 (quoting 15 U.S.C. § 717f(b)). For twenty years the Commission has struggled to streamline the regulatory process for contract termination by allowing pre-approved abandonment while meeting its obligations under § 7(b) to protect customers from pipeline market power. See id. at 50-51.
In United Distribution Cos. v. FERC, 88 F.3d 1105 (D.C.Cir.1996) (UDC), we reviewed FERC’s regulation providing captive shippers with a right of first refusal (ROFR) which allowed such shippers to avoid pre-approved abandonment and extend their contracts if they matched the rate and duration — up to a twenty-year cap — offered by competing bidders. We approved “the basic structure of the right-of-first-refusal mechanism,” finding that it “provides the protections from pipeline market power required for pre-granted abandonment under § 7.” Id. at 1140. At the same time, we remanded the twenty-year cap, in part because we saw no reasoned explanation of how it would adequately meet the Commission’s § 7(b) obligation to protect captive customers from the exercise of pipeline market power. Id. at 1140-41.
On remand, FERC adopted a five-year cap. Pipeline Service Obligations and Revisions to Regulations Governing Self-Implementing Transportation Under Pari 281 of the Commission’s Regulations, 78 F.E.R.C. ¶ 61,186, at 61,774, 1997 WL 81264 (1997). We vacated and again remanded, citing (1) FERC’s unsupported decision to rely on the median contract length; (2) FERC’s concerns that the cap would “result[ ] in a bias toward short-term contracts,” fostering an “imbalance of risks between pipelines and existing ship[179]*179pers” and raising “the overall cost of pipeline transportation”; and (3) FERC’s earlier suggestion that “elimination of the cap would foster efficient competition.” IN-GAA, 285 F.3d at 52-53 (internal quotations marks omitted). We also noted that the Commission provided neither “an affirmative explanation for the selection of five years, nor a response to its own or the pipelines’ objections.” Id. at 53.
“[Wjhether a term-matching cap must be required as part of the ROFR,” FERC explained, “turns on whether [it] is necessary to protect the existing long-term shipper from the pipeline’s exercise of market power.” Regulation of Short-Term Natural Gas Transportation Services, and Regulation of Interstate Natural Gas Transportation Services, 101 F.E.R.C. ¶ 61,127, at 61,522, 2002 WL 31974225 (2002) (“Order on Remand”) (citing UDC, 88 F.3d at 1140), aff'd, 106 F.E.R.C. ¶ 61,088 (2004) (“Order on Rehearing”). According to the Commission, “[m]arket power is exercised through the withholding of capacity to create an artificial scarcity, thereby raising prices.” Id. at 61,521. Finding that existing regulations adequately control the exercise of market power, FERC abolished the cap altogether. Id. at 61,519.
In finding the term cap unnecessary, FERC employed the reasoning we sustained in Process Gas Consumers Group v. FERC, 292 F.3d 831 (D.C.Cir.2002) (“PGC II ”), a recent decision in a parallel line of cases addressing term-matching caps for new, rather than existing, customers. In PGC II, we rejected challenges to FERC’s removal of a twenty-year cap for new shippers, accepting the Commission’s determination that “existing regulatory controls already limit [pipelines’] market power, thereby minimizfing any] danger that the pipeline will withhold ... capacity from the market to create [the] artificial scarcity necessary to force shippers to bid for supercompetitive contract terms.” Id. at 836 (alteration and omission in original) (internal quotation marks omitted). In the order at issue here, the Commission cited several “existing controls” it believed would protect existing shippers from any pipeline exercise of market power: (1) traditional cost-of-service rate regulation; (2) pipelines’ obligation to offer to sell all existing capacity; (3) the Commission’s complaint process, Order on Remand, 101 F.E.R.C. at 61,521; Order on Rehearing, 106 F.E.R.C. at 61,298; (4) the opportunity customers have to offset the cost of unwanted capacity through capacity release, Order on Rehearing, 106 F.E.R.C. at 61,-304; and (5) the constraints on contract conditions imposed by the pro forma tariff reviewed by FERC, id. at 61,303. Given these regulatory controls, FERC concluded that pipelines could exert market power only by refusing to “build additional capacity when demand requires it.” Order on Remand, 101 F.E.R.C. at 61,521. According to the Commission, however, “pipelines would have a greater incentive to build new capacity” since pipelines could only increase profits by investing “in additional facilities to serve the increased demand. Moreover, if [pipelines] did refuse to build new capacity, the shippers could file a complaint.” Id.
FERC acknowledged that unlike the new customers at issue in PGC II, existing shippers enjoy statutory protection from pipeline market power. Id. at 61,522. The Commission explained, however, that the ROFR adequately protects existing customers by “ensuring] that, if the existing customer is willing to pay the maximum approved rate and match the contract term of a rival bidder, the pipeline may not abandon service to that customer.” Id.
[180]*180Represented by petitioner American Gas Association, a collection of existing shippers now challenge FERC’s elimination of the matching-term cap, charging that the Commission reversed “ten years of practice” and failed to provide “coherent support for the conclusion that the Commission need no longer concern itself with the exercise of pipeline monopoly power upon contract expiration.” Pet’rs’ Opening Br. 24. For such claims, our standard of review is both well established and highly deferential. We will “uphold FERC’s factual findings if supported by substantial evidence and ... endorse its orders so long as they are based on reasoned decision making.” Texaco, Inc. v. FERC, 148 F.3d 1091, 1095 (D.C.Cir.1998). In addition, we must ensure that FERC has adequately responded to our concerns in IN-GAA. PGC II, 292. F.3d at 836. As we have already recognized, however, because selection of the matching cap duration must be “somewhat arbitrary,” we “defer to the Commission’s expertise if it provides substantial evidence to support its choice and responds to substantial criticisms of that figure.” UDC, 88 F.3d at 1141 n. 45.
In INGAA, we evaluated the evidence upon which FERC relied to determine appropriate term caps, examining the data it used as well as its explanation linking those data to the cap length selected. See 285 F.3d at 51-53; see also Process Gas Consumers (Group v. FERC, 177 F.3d 995, 1003 (D.C.Cir.1999) (“PGC I”) (asking the same questions). We do the same here.
As to the first inquiry, petitioners argue that FERC ignored available evidence — indeed, that it failed to look at any data at all. FERC responds that the only existing evidence came from regulated markets and that such evidence reveals nothing at all about the duration of pipeline contracts in competitive markets. Order on Rehearing, 106 F.E.R.C. at 61,305. FERC is correct. As we explained in PGC I, the relevant question for assessing a term cap is “whether it will prevent the [pipelines] from compelling shippers to offer the pipeline longer contracts than they would in a competitive market.” PGC I, 177 F.3d at 1003 (emphasis added).
Even were we to find the available data useful, under our second inquiry — the linkage between the evidence before the Commission and the cap length selected— FERC’s analysis of the evidence is reasonable. First, FERC found that the evidence gathered both before and after the imposition of the five-year cap demonstrates that “the five-year term-matching cap has not led to shorter contract terms than would otherwise occur,” and that “customers have generally been able to negotiate contract terms of less than five years and do not desire longer contracts because of the risk that their needs will change.” Order on Remand, 101 F.E.R.C. at 61,523. Second, FERC explained that the evidence of stable contract duration indicates that regulation without a term cap fulfills its obligations to the natural gas market. With this market-based distributive policy, FERC effectively responded to its own concern, cited in INGAA, “that the cap would foster an imbalance of risks between pipelines and existing shippers,” adversely affecting allocation of both cost and capacity. IN-GAA, 285 F.3d at 53 (internal quotation marks omitted); Order on Remand, 101 F.E.R.C. at 61,522. We therefore agree with FERC that since the “matching cap is not necessary to limit the exercise of market power by the pipelines,” there is “no justification for distorting the bidding process and not allocating scarce pipeline capacity to the shipper placing the highest value on obtaining” it. Order on Rehearing, 106 F.E.R.C. at 61,300.
Having satisfied INGAA’s evidentiary requirements, FERC must still demon[181]*181strate it has adequately responded to petitioners’ “substantial criticisms.” INGAA, 285 F.3d at 52 (quoting UDC, 88 F.3d at 1141 n. 45). Petitioners assert that PGC II’s holding that existing regulations are sufficient to protect against abuses of market power has no applicability here because pipelines can exercise greater market power over existing customers than over new customers. While in the context of new shippers “market power can be exercised only by withholding capacity in order to create artificial scarcity,” petitioners argue, “the exercise of pipeline monopoly power vis-a-vis existing customers lies in a pipeline’s ability to threaten termination of service to customers who require essential pipeline services.” Pet’rs’ Opening Br. 34-35. Where shippers have no viable alternatives for service, contract term becomes a substitute for price, undermining rate regulation established under NGA sections 4 and 5. Id. at 44.
Although we decided PGC II in a context where the Commission owed no extra duty to protect shippers from pipeline market power, PGC II, 292 F.3d at 838 (noting the difference), our reasoning there applies equally to existing customers. As repeatedly quoted by the Commission, PGC II found that “shippers may at times bid up contract length” but that such bidding “likely reflects not an exercise of [pipelines’] market power, but rather competition for scarce capacity.” 292 F.3d at 837. Petitioners are therefore incorrect when they assert that pipelines inevitably exercise market power merely because scarcity deprives existing customers of alternatives or forces them to bid higher terms than they desire. See Pet’rs’ Opening Br. 41-45; see also PGC I, 177 F.3d at 1003 (noting pipelines should be prevented “from compelling shippers to offer the pipeline longer contracts than they would in a competitive market”). Rather, to demonstrate that pipelines exercise market power, petitioners must show that the pipelines manipulate that scarcity by intentionally withholding capacity to drive up the value of bids or to extract special advantages or concessions. Order on Rehearing, 106 F.E.R.C. at 61,-301-03. Having found captive shippers fully protected from such manipulation by existing regulatory controls, see supra at 5, FERC concluded not only that retaining the cap was unnecessary, but also that it would interfere with allocation of capacity to shippers who value it most and simply protect captive customers “from actual competition.” Opp’n Br. 40.
Petitioners next argue that FERC failed to address the impact of scarcity on historical captive customers either (1) forced to compete with new electric generation customers for capacity or (2) held captive by virtue of retail access programs. As to the first point, petitioners contend that removal of the term cap, which allows new customers to bid up contract term, conflicts with FERC’s policy requiring new shippers to pay higher rates reflecting the cost of new construction. Pet’rs’ Opening Br. 48-49. This argument is without merit. Petitioners presented no evidence that existing ROFR shippers competing to renew capacity on a fully subscribed pipeline with different vintages in capacity are not also subject to incremental pricing. See Regulation of Short-Term Natural Gas Transportation Services and Regulation of Interstate Natural Gas Transportation Services, F.E.R.C. Stats. & Regs. [Reg. Preambles 1996-2000](CCH) ¶ 31,099, at 31,635-36 (“Order No. 637-A”), reh’g, F.E.R.C. Stats. & Regs. [Reg. Preambles 1996-2000](CCH) ¶ 31,091 (“Order No. 637”), order den. reh’g, 92 F.E.R.C. ¶ 61,602 (2000) (“Order No. 637-B”).
For them second point, petitioners argue that in balancing consumer and pipeline risks, the Commission failed to account for [182]*182the greater risks faced by local distribution companies (LDCs) bound by retail access programs that threaten their market share while simultaneously obligating them to serve as suppliers of last resort. Pet’rs’ Opening Br. 50-54. Given this predicament and the lack of alternative sources of capacity, petitioners contend that the Commission’s rule will allow customers “with greater buying power to bid up the pipeline’s firm, long-term services.” Id. at 53. As a result, LDCs will be forced to “enter into long-term pipeline contracts now to serve markets they may or may not serve in the future.” Id. at 54.
This argument depends on two false assumptions: (1) that existing regulations without a term cap leave pipeline market power unregulated and (2) that § 7(b) obligates FERC to guarantee shippers the ability to renew their contracts indefinitely rather than simply provide them the opportunity to do so. PGC II makes clear that a market is not unregulated just because shippers have no alternative sources of service. 292 F.3d at 837. In INGAA, moreover, we accepted as valid FERC’s concern about creating “an imbalance of risks between pipelines and existing shippers, allowing shippers indefinite control over pipelines’ capacity, but giving the pipelines no corresponding protection.” 285 F.3d at 53 (internal quotation marks omitted).
As for petitioners’ concern that uncertainties about future markets require extra protection, we have made clear that “LDCs are no different from other industry participants in that they will have to evaluate future risks in determining how much capacity to reserve.” UDC, 88 F.3d at 1140-41 n. 44. The costs versus the benefits of renewal are, we explained, financial determinations all companies, including LDCs, must make. Id. Having considered these risks, FERC concluded that the possibility that LDCs may end up with unneeded capacity does not outweigh the benefit of allocating scarce capacity to parties valuing it the most. Order on Rehearing, 106 F.E.R.C. at 61,300. Moreover, because FERC’s capacity release program allows LDCs to market capacity they retain but cannot use, LDCs may “mitigate any business harm that might occur ... from elimination of the term matching cap.” Id.
In sum, since FERC must protect existing shippers from market power, not from competition, and given its conclusion that existing regulations protect against the exercise of market power, the Commission reasonably concluded that the ROFR gives existing shippers the competitive advantage that § 7(b) requires while allowing for the most efficient allocation of capacity. We will therefore deny the petitions for review on this issue.
II.
The second issue before us has its origins in Order No. 636, in which FERC adopted a segmentation policy and expanded flexible point rights for shippers with firm service. Pipeline Service Obligations and Revision to Regulations Governing Self-Implementing Transportation Under Part 28U of the Commission’s Regulations, F.E.R.C. Stats. & Regs. [Reg. Preambles 1991-1996](CCH) ¶ 30,939, reh’g granted & denied in part, F.E.R.C. Stats. Regs. [Reg. Preambles 1991-1996KCCH) ¶ 30,-950 (“Order No. 636-A”), order on reh’g, 61 F.E.R.C. ¶ 61,272 (1992) (“Order No. 636-B”). These changes allowed shippers to segment their capacity and use any receipt and delivery point within the zone for which they pay reservation charges. Through shipper ability to release excess firm capacity, FERC also created a secondary market for firm capacity in competition with pipeline interruptible service. [183]*183With this new market, shippers acquired increased control over the capacity for which they pay. In addition to allowing segmentation, the Commission allowed shippers to engage in “baekhaul/forwardhaul transactions” to the same delivery point. In INGAA, we upheld segmentation and flexible point rights as a general matter, 285 F.3d at 38-40, but remanded for further explanation FERC’s decision to allow baekhaul/forwardhaul segmented transactions. Id. at 41.
INGAA describes baekhaul/forwardhaul transactions as follows:
Suppose a pipeline runs from A to B to C, and has 10,000 dekatherms of daily capacity, all of which is contracted for from A to C and of which X holds 1000. X’s mai'ket at C declines, and X would like to ship only to B and to release the 1000 in B-C capacity. X learns of ... Y, who has a right to 1000 dekatherms at C and would like to sell it at B.
Id. at 40. X then forwardhauls 1000 dekatherms to B and releases the B-C portion to Y, who backhauls 1000 dekatherms to B. The result is 2000 dekatherms at point B, 1000 dekatherms in excess of the amount X contracted to have pass through that point.
On remand, FERC defended its decision to allow baekhaul/forwardhaul transactions. Order on Remand, 101 F.E.R.C. ¶ 61,127. Various pipelines, represented by Tennessee Gas Pipeline Company, now challenge that decision. They argue that FERC’s policy effectuated an increase in shippers’ delivery point entitlements and the services pipelines are required to provide, thereby modifying existing contracts. Those contract modifications cannot stand, they argue, because FERC failed to make findings under either the Mobile-Sierra heightened public interest standard or NGA § 7. See generally MCI Telecomms. Corp. v. FCC, 822 F.2d 80, 87 (D.C.Cir. 1987) (describing the Mobile-Sierra standard); 15 U.S.C. § 717f(a). We disagree.
To begin with, contrary to petitioners’ argument, nothing in INGAA “found that the Commission’s policy will modify contracts.” Pet’rs’ Opening Br. 61. Noting the difference between contract modification and operational feasibility, INGAA remanded the baekhaul/forwardhaul issue to the Commission “so that it can more clearly confront the question of whether this aspect of the orders can stand without additional findings.” INGAA, 285 F.3d at 41. This language does not direct FERC to make Mobile-Sierra or § 7 findings. Instead, it directs the Commission to “more clearly confront the question of whether” additional findings are necessary. Id. (emphasis added). This left FERC free to determine “whether” its baekhaul/forwardhaul policy modified pipeline contracts, and if so to make the necessary Mobile-Sierra or § 7 findings. Examining the issue, FERC concluded that the policy did not modify the contracts, and we agree.
FERC’s conclusion rests on the difference between firm and guaranteed service. As the Commission explained, its
flexible point policy distinguishes between primary points and secondary points. Firm contracts between pipelines and their shippers typically provide that the pipeline will transport up to a specified contract demand from a primary receipt point or points listed in the contract to a primary delivery point or points listed in the contract. This provision specifies a shipper’s guaranteed right to firm service.
Order on Rehearing, 106 F.E.R.C. at 61,-305 (emphasis added) (footnotes omitted). In contrast to primary firm service, “an out-of-path, secondary firm transaction”— the usual categorization of a backhaul— “receives a lower scheduling priority than [184]*184primary firm service.” Id. at 61,308. Since backhauls are secondary transactions rather than guaranteed service, back-hauling shippers utilize the delivery point on a secondary basis — a basis not covered by the contract. Id.
Even if it is true, as petitioners contend, that the contract specifies the maximum daily quantity of gas parties agree to transport on a firm basis and that released capacity delivered on a secondary basis constitutes firm service, see Order No. 636-A, F.E.R.C. Stats. & Regs, at 30,583 (noting that secondary rights are firm rights that are subordinate to primary rights but superior to interruptible rights for scheduling and curtailment purposes), petitioners have not challenged FERC’s contention that given its segmentation and flexible point policy, shippers make use of two types of firm service, primary and secondary, with only the former amounting to 'guaranteed service. See Order on Rehearing, 106 F.E.R.C. at 61,308. Therefore, secondary transactions — firm but not guaranteed — are not covered by the contracts. Id. Because the terms of primary service for which the parties have bargained remain unchanged, FERC’s decision does not modify contracts, even if it affects them.
Petitioners also contend that the back-haul/forwardhaul policy allows shippers to get “more service than they are paying for.”- Id. at 61,309. As FERC explained, however,
it is the Commission’s policy that a shipper may use all of the points in a zone for which it is paying on a secondary basis precisely because the shipper must pay the costs of the entire zone.... The shipper is getting no more than what it pays for. The pipeline, for its part, has fully allocated its costs and is collecting those costs from its shippers.
Id. at 61,310 (footnotes omitted). The Commission further explained that since backhaul/forwardhaul transactions do not alter pipelines’ certificated service levels or specified service entitlements by changing the quantity provisions of their transportation contracts, the policy merely changes the terms of an existing service. Id. at 61,313. Finally, FERC pointed out that should a backhaul/forwardhaul transaction cause the pipeline to lose significant revenue, “then a pipeline is permitted to file a new rate case in which more of its costs would be allocated to firm service.” Id. at 61,310.
Because FERC’s backhaul/forwardhaul policy does not abrogate pipeline contracts, the Commission had no obligation to make either Mobile-Sierm or § 7 findings. Instead, to justify its new policy, the Commission needed to comply only with NGA § 5, 15 U.S.C. § 717d. Under NGA § 5, before replacing an existing rate or tariff with a new one, the Commission must demonstrate by substantial evidence that the existing rate or tariff has become unjust or unreasonable, and that the proposed rate is both just and reasonable. 15 U.S.C. § 717d; W. Res., Inc. v. FERC, 9 F.3d 1568, 1579-80 (D.C.Cir.1993).
FERC found that since backhaul/forwardhauls represent a type of segmented transaction, failure to permit them is unjust and unreasonable for the same reasons the Commission gave in Order No. 637 and that we accepted in INGAA, i.e., “it restricts efficient use of capacity without adequate justification.” Order on Remand, 101 F.E.R.C. at 61,529; INGAA, 285 F.3d at 38 (citing Order No. 637, at 31,304). At the same time, FERC found that permitting these transactions is “just and reasonable because it creates additional supply alternatives for shippers and enhances competition on the pipeline’s sys[185]*185tem ...[,] because it provides the kind of flexibility that pipelines enjoyed prior to Order No. 636 and because it will assist in creating more competition in the transportation market.” Order on Rehearing, 106 F.E.R.C. at 61,307.
“[I]t is within the scope of the agency’s expertise to make ... a prediction about the market it regulates, and a reasonable prediction deserves our deference ....” Envtl. Action, Inc. v. FERC, 939 F.2d 1057, 1064 (D.C.Cir.1991). Petitioners have given us no reason to second guess FERC’s conclusion that the benefits of segmentation and flexible point policy apply equally to backhaul/forwardhaul transactions.
III.
On remand from INGAA, FERC reevaluated both the term cap and the back-haul/forwardhaul issues and gave satisfactory explanations for its decisions. The petitions for review are denied.
So ordered.