Opinion for the Court filed by Senior Circuit Judge SILBERMAN.
SILBERMAN, Senior Circuit Judge:The Postal Service offers a host of different products for consumers and businesses. By statute, the Postal Service’s products are divided into “competitive” products, which are subject to a statutory price floor, and “market-dominant” products, which are subject to a statutory price ceiling. In this case, the Postal Service applied to the Postal Regulatory Commission to have one of the Service’s products, the “round-trip mailer,” used in the DVD-by-mail industry, classified as competitive rather than market-dominant. The Commission denied the request, and the Postal Service petitioned for review. We deny the petition.
I.
The round-trip mailer is a product used by intervenors Netflix and GameFly whereby customers receive a DVD disc by mail that they can return with an enclosed envelope. Netflix, a provider of movies, is the dominant purchaser of this product, with nearly 97 percent of purchases.1 It is provided only by the Postal Service; Fed Ex and UPS do not offer anything similar. Customers, to be sure, can also rent DVDs at various physical locations, such as storefronts and kiosks, but these alternatives are declining. Movies can also be delivered to customers through streaming directly to televisions or computers, and that is a booming market in which Netflix, which also offers streaming, meets a number of direct competitors.
Netflix customers are, in huge numbers, leaving the DVD-by-mail service. After peaking in 2010, DVD-by-mail declined 14 percent, and then another 41, 21, and 22 percent, year over year. Meanwhile, streaming service continues to expand, growing 26, 25, and 19 percent, year over year, from 2012 to 2014. Nevertheless, a significant core continues to use the round-trip mailer service, in part because the content available through the round-trip mailer is more extensive than that available through streaming and, in part, for difficulties in reception with streaming. Indeed, nearly two-thirds of all DVD-by-mail customers also subscribe to streaming. Importantly, although the round-trip mailer is a declining product, it remains quite profitable to Netflix, with a profit margin of nearly 50 percent. That makes it considerably more profitable than streaming.
In 2013, the Postal Service formally requested that the Commission remove the round-trip mailer from the market-dominant list, where it is subject to a statutory price ceiling, and place it on the competitive product list, where it would be subject to only a statutory price floor. The request was opposed by intervenors Netflix and GameFly, as well as by a Commission-appointed public representative. Market-dominant products obviously include those that are an actual monopoly, but the governing statute describes market-dominant products more broadly as those “in the sale of which the Postal Service exercises sufficient market power.” See 39 U.S.C. § 3621-3634.
As noted, the Commission denied the request. Applying antitrust concepts, it defined the relevant market as the physical delivery of DVDs by mail, and ruled that the Postal Service, as the only entity producing the round-trip mailer, exercised sufficient power within that market to justify a market-dominant classification.
II.
Section 39 U.S.C. § 3642(b)(1), the governing statute’s definition of a market-*488dominant product, includes all products meeting the following description:
[E]ach product in the sale of which the Postal Service exercises sufficient market power that it can effectively set the price of such product substantially above costs, raise prices significantly, decrease quality, or decrease output, without risk of losing significant levels of business to other firms offering similar products, (emphasis added)
It would seem that a permissible reading of the statute would foreclose the Postal Service’s position because arguably there really are no products in the marketplace “similar” to the round-trip mailer. But the Commission does not rely on that interpretation of the language. Instead, the Commission and all parties agree that the statute should be read as incorporating general concepts of federal antitrust law.2
In that regard, the Postal Service contends that the Commission erred in its definition of the market. First, it made too much of the difference between DVD-by-mail and streaming services, whereas antitrust law has traditionally recognized that even differentiated products, provided they are reasonable substitutes for one another, may be included in the same market. In this case, the precipitous drop in DVD-by-mail customers, which has attended the concurrent rise in streaming services, strongly suggests, argues the Postal Service, that a substantial substitution is taking place, at the consumer level, between the two forms of delivery. That substitution is an indication that whatever the differences between DVD-by-mail and streaming, customers see them as similar enough that they ought be considered in the same market. And as DVD-by-mail service is in the same market as streaming services downstream, the Service argues that streaming services have the effect of disciplining the price the Postal Service can charge for DVD-by-mail service upstream, thus mitigating any market power.
Secondly, the Postal Service argues that the Commission relied too heavily on the “core” group of DVD-by-mail consumers who do not use streaming services. While any product will have core consumers who refuse to switch products even in the face of price increases, the Commission, we are told, overemphasized their numbers or loyalty. This misconception led to an unduly narrow definition of the downstream market, failing to take account of the constraints faced by the Postal Service upstream in setting its price for the round-trip mailer.
In the alternative, the Postal Service offers two arguments apart from market definition. Even if the definition of the relevant market was not unreasonable, the Commission still failed to consider that Netflix, as a near-monopsonist in purchasing the round-trip mailer upstream, can exercise strong countervailing buyer power. As such, it would limit the Postal Service’s pricing discretion. And the Commission did not consider that the market, defined as either DVD-by-mail or distribution of content generally, is in a constant state of technological evolution. Such change provides an intrinsic limitation on the exercise of market power.
The Commission’s arguments dovetail with its considerations set forth in its order. In defending its definition of the relevant market as limited to DVD-by-*489mail, the Commission emphasizes evidence that DVD-by-mail provides a different range of material, and services a different base of consumers, than streaming and other distribution channels. Regarding substitutability, it is noteworthy that many customers subscribe to both DVD-by-mail and streaming services. It further disputes the Postal Service’s assertion that the simultaneous decline in DVD-by-mail subscriptions and increase in streaming subscriptions suggests a high enough sub-stitutability to justify a unified market, because the shift toward streaming may have nothing to do with price sensitivity of DVD-by-mail customers; it may just as well result from broader market trends unrelated to price-based substitution.
The Commission stresses that the Postal Service bore the burden in the present case. It was obliged to present evidence demonstrating that, despite the fact that the Postal Service is the only entity providing the round-trip mailer for DVD-by-mail services, it nevertheless lacked market power. That evidence, particularly in the form of price elasticity data, was lacking. Thus, the Commission argues it was reasonable to conclude that the Postal Service did not meet its burden.
Finally, the Commission contends that even if there is downstream competition between delivery by mail and delivery by other means (such as streaming and kiosks), the Postal Service still failed to show that it would not be in a position to simply capture more of Netflix and Ga-meFly’s profits upstream. Even if increased prices resulting from a new classification could not be passed on by Netflix and GameFly because of rugged downstream competition, the removal of a price ceiling would allow the Postal Service to use its position as the sole provider of the round-trip mailer to capture a larger portion of the intervenors’ profits.
[[Image here]]
Petitioner’s case essentially rests on the proposition that the proliferation of competitors in the downstream market (direct sales to consumers) restricts the capacity of the Postal Service to raise prices to Netflix and GameFly. Therefore, the Postal Service contends, streaming services and DVD-by-mail services compete with each other and, under antitrust law, should be treated as a single market. We should note at the outset of our analysis that we do not sit as an appellate court reviewing an antitrust case from the district court. Rather we are reviewing an administrative agency decision under the APA—which means the agency is entitled to deference. Even if we were to disagree with the Commission’s economic analysis, we are obliged to affirm if the agency was at least reasonable.3 See City of Los Angeles v. Dept. of Transp., 165 F.3d 972, 977 (D.C.Cir.1999) (“[W]e do not sit as a panel of referees on a professional economics journal, but as a panel of generalist judges obliged to defer to a reasonable judgment by an agency acting pursuant to congres-sionally delegated authority.”).
But we think the Commission was more than just reasonable; its decision seems rather compelling. As the Commission found, the petitioner produced no evidence to establish that a significant increase in the price of DVD-by-mail—the cost to Netflix and GameFly in the upstream market—would result in a significant shift by Netflix or GameFly away from purchasing DVD-by-mail services upstream. In other words, there was no evidence presented to *490establish at what point an increase in the price the Service charged would cause Netflix and GameFly to look elsewhere for distribution.
There is a fundamental logical flaw in the Service’s argument. The Service enjoys market power in the (upstream) distribution market regardless of conditions in the (downstream) content market because it does not face any competition in the distribution market. Whether because of economic or legal hurdles, no other business offers or seems reasonably poised to offer a competing distribution channel that Netflix could readily take advantage of. If the Postal Service were to raise the price of its round trip mailer to Netflix and GameFly, there are only two possibilities. First, it may be the downstream market is not sufficiently competitive, and thus Net-flix and GameFly could pass on the price increase to consumers without consequence. Second, alternatively, it may be the downstream market is competitive, and thus Netflix and GameFly could not pass on the price increase to consumers without losing them. In the first instance, the Postal Service’s market power upstream coexists with Netflix and GameFly’s market power downstream. They would share in the supracompetitive price. In the second instance, although Netflix and GameFly lack power downstream, the Service’s power upstream remains, and puts it in a position to extract more of the profit margin of its captive suppliers. In either case, the Postal Service employs market power.4
There remain petitioner’s two other arguments, even assuming it can be thought to have market power. First, that Net-flix—after all, a market behemoth—has sufficient economic clout to counter petitioner’s market power. But that theory simply doesn’t apply in this case because Netflix and GameFly have no alternative means to transport DVDs by mail, so there is no way that Netflix could exercise any bargaining leverage vis-a-vis the Postal Service. It should be obvious that as long as the DVD-by-mail business is quite profitable, Netflix has no rational economic course but to pay the market-dominant piper. As to the second alternative argument, regarding the speed of technological change, we have said that market power analysis need account for only potential “substitutes that constrain pricing in the reasonably foreseeable future.” United States v. Microsoft Corp., 253 F.3d 34, 53-54 (D.C.Cir.2001). The Commission was not, therefore, unreasonable to hold that the potential technological evolution suggested by the Service was too speculative to condition its market power analysis here.
For the foregoing reasons, we deny the petition for review.
So ordered.
. GameFly provides video games on disc.
. A statutory price cap, indexed to inflation, is meant to assure that the Postal Service does not abuse a dominant position, see U.S. Postal Serv. v. Postal Regulatory Comm'n, 785 F.3d 740, 744 (D.C.Cir.2015). This is not unlike rate regulation for other state-granted monopolies. The language of 39 U.S.C. § 3642 has been consistently taken to reflect basic antitrust concerns with market power.
. In other words, not arbitrary and capricious, see Gundersen Lutheran Med. Ctr., Inc. v. Sebelius, 666 F.3d 1335, 1337 (D.C.Cir.2011); Blumenthal v. FERC, 613 F.3d 1142, 1146-47 (D.C.Cir.2010).
. Court precedent, and antitrust doctrine generally, recognize that market power may be exercised upstream, even where there is a competitive market downstream. See Coal Exporters Association v. United States, 745 F.2d 76, 84-85 (D.C.Cir.1984); 2B Areeda et. al., Antitrust Law ¶ 564 (4th ed.2014). This effect is recognized in several other antitrust inquiries that involve market power, see Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co., Inc., 549 U.S. 312, 127 S.Ct. 1069, 166 L.Ed.2d 911 (2007) (finding buyer power in upstream market for inputs, despite lack of seller power in national downstream market for outputs); Mandeville Island Farms v. American Crystal Sugar Co., 334 U.S. 219, 68 S.Ct. 996, 92 L.Ed. 1328 (1948) (finding market power upstream, despite lack of power downstream); Einer Elhauge, United States Antitrust Law and Economics, 172, 263 (2d ed.2011); Herbert Hovenkamp, Federal Antitrust Policy, § 8.10 (4th ed.2011).