Just How Much Antitrust Risk Is There in a Netflix - Warner Brothers Merger?

Less than I thought, less than you've read about, and less still if Paramount is the buyer. Either will probably get sued anyway. And win. Read why.

Merry Christmas and Happy Holidays COTM readers! I hope you had a wonderful break and Santa brought you everything you wanted.

We’re back to take on the potential deal that has the antitrust world talking as the year closes. The saga between Netflix and Paramount’s rival attempts to purchase Warner Brothers Discovery.

I won’t recount all of the bidding here. You can go elsewhere for that. Here’s a recent update. But the basics are relatively simple:

• Warner Brothers Discovery (WBD) has recommended its shareholders accept Netflix bid over rival Paramount

• Paramount is not out of the game, and its hostile bid was for the entire WBD business, including the TV and cable networks that would be split off under the Netflix bid. Some reports place the value of the parts of WBD Netflix would leave behind at about $5/share (i.e. Netflix slightly outbid Paramount, but that is subject to some debate and interpretation and not especially important for our discussion about antitrust risks)

A comparison of the rival bids look something like this:

Now, while the competitive saga between Netflix and Paramount has been a fun and entertaining watch with all of the interesting optics one would expect out of a Hollywood deal, I’m here for the antitrust analysis. And I presume, COTM subscribers, that you are too.

What got me interested in tracking this deal was not the optics, or even the $100 billion price tag on the deal. A few things got me interested enough to write here:

• First, I’m interested in the products! I subscribe to Netflix, I’m writing this on a Sunday and about to lock in and watch Landman on Paramount.

• Second, with those large breakup fees and the stakes here, either deal looks like it would be headed to litigation of the DOJ Antitrust Division challenges. And when we get antitrust issues in front of federal judges, the case is not decided by politics, or on X, or even popular opinion, but rather by antitrust analysis. That’s fun.

• Third, when I looked for serious analysis of the relative antitrust risks of the competing deals I could not find any.

What I could find was what appeared to be a widely shared, if not shallow and reflexive, view that the Netflix deal would be illegal. Tim Wu has never met a merger he liked but here’s one example:

Ditto Liz Warren:

Stoller remains in character and declares while waving hands wildly about that both deals would be illegal.

OK, you are not surprised to read that Lina Khan or Matt Stoller or Tim Wu don’t like a merger. The best we get for analysis is Tim Wu’s claim that the Netflix deal will trigger the PNB presumption — a subject upon which COTM readers are well educated. As a reminder to COTM readers – the PNB presumption is a rebuttable presumption of illegality that courts apply to mergers challenged under the Clayton Act when the government proves that post-merger shares are greater than 30 percent.

I mean, I’m the first person to say invocation of the PNB presumption in federal court – if successful – is a pretty big deal in terms of the potential success of a merger challenge. But that is because without the PNB presumption the government nearly always loses. So yes, the government’s chance for success is much higher when it invokes the presumption because it gives the government at least some chance to win. But invoking the PNB presumption is not some sort of binary operator giving the DOJ a 100% win rate.

In most of the merger challenges the government LOSES it also invokes the PNB presumption. Invoking the PNB Structural Presumption is sort of like getting your driver’s license before a car race. Yes, you have done enough to get behind the wheel of the car. But a lot of things can still go wrong. And you can absolutely still lose the race. And the government often does. When it comes to substantive analysis of the actual antitrust risks here — we are not left with much. That sounds like a great topic to take on. So here we are.

And to close out the year, as always, please subscribe, upgrade to paid, and send to your friends to do the same:

With that, let’s get started.

What Kind of Antitrust Animal is the Netflix-WBD Deal?

The Netflix-WBD deal has both horizontal and vertical aspects. Both Netflix and WBD are distributors who are vertically integrated into content to different degrees. Netflix started out in distribution, eventually evolving into the streaming service you see today in 2007. Netflix, of course, has made a substantial investment into producing content as well. WBD also began in distribution and quickly got into the production game, making their own movies by 1908, and growing to become one of the five major Hollywood Studios by the 1920s. Stratechery has a great description of the business. The story he tells is a simple (and correct) one about how the internet turned the world upside down. Content was king in a world where distribution was constrained by the cost of running movie theaters, the number of seats, etc. But digital distribution and aggregation matters more in the new world.

So the deal is horizontal in the sense that Netflix and WBD are rival streaming services, i.e. a reduction in the number of streaming services might result in higher streaming prices. One can also argue that the reduction in streaming services will harm upstream producers. I could see the Khan/Kanter DOJ interested in pursuing this “protect the producers” theory. But I find it unlikely to come from the Trump DOJ. Perhaps the California AG. But I think it is a weak theory. It is not shocking that the Writers Guild and actors like Ben Stiller have raised this objection. But the problem they have is that “big movies in theaters” have been dying for decades for reasons that have nothing to do with consolidation. Antitrust is many things – but it takes consumer tastes as given and it is not the right tool for saving a dying industry. For an excellent take on this foreclosure see ICLE’s Eric Fruits at Truth on the Market here:

The proposed transaction also involves a vertical aspect in that a distributor of content (Netflix) is acquiring a supplier (WBD). With that comes the theoretical concern that Netflix might foreclose rival SVOD networks from access to WBD content. For my money, the primary antitrust theory of harm here is the horizontal objection, i.e. the merger will reduce competitive rivalry in the SVOD market and harm subscribers to those services. I’ll (mostly) focus there because that is where the more salient antitrust risk lives.

So what kind of antitrust animal is it? What is the source of the antitrust risk? And what kinds of arguments, evidence, and analysis will shape that realization of that risk in court?

One thing every antitrust observer from casual follower to expert agrees upon: market definition is really important! And it is. But market definition battles come in different flavors. Yes, the government is always arguing for a narrower market definition — both to increase market shares. But remember, market shares are supposed to be useful in antitrust only to the extent they are a proxy for intensity of competition. And as we know, sometimes competition can be very intense when a firm has a high share. In fact, intensity of competition from a firm can be the reason it earns a high share! Courts know this too. And so ultimately engage in a competitive effects analysis when analyzing a merger, i.e. they ask “what will happen to competition after the merger compared to the world with no merger at all?” So the other reason market definition really matters is that (from the government perspective) it eliminates more (ideally, distant) sources of competition from that analysis.

To give a simple example: if Starbucks merges with Peet’s Coffee the government might want to define the market to exclude coffee carts and Mountain Dew from the market not just to drive up the market shares but also to head off the merging parties from arguing that those sources of competition would prevent a post-merger price increase.

I’ve been working on antitrust analysis of mergers for a very long time. As a lawyer and economist both for and against the government, as an FTC Commissioner, and writing about them as an academic. In that experience you tend to see some patterns about different archetypes of deals and antitrust profiles. I won’t recount all of those archetypes here – it would be a great piece to write one day (I’ll put it on the list) – but here are a few that are relevant here just to simplify the following discussion.

• Antitrust Profile #1: Market Definition Doesn’t Matter, the Battle is Over Efficiencies or Entry

  • Think national carrier telephone deals; these deals easily invoke the PNB presumption and there is not an intense fight over market definition (i.e. both sides loosely agree) but a big fight on effects

• Antitrust Profile #2: Market Definition is Knife’s Edge Issue

  • Think FTC v. Staples-Office Depot (1997). The government argued the market was “consumable office supplies sold through office superstores” and under that market post-merger shares would be in 70s and 100% in some markets. The superstores argued the market was “everywhere you can buy office supplies” and that the market share was about 4%. If the FTC won on market definition, it would not only get the PNB presumption, but ALSO win on competitive effects (and vice versa). But it all turned on market definition.
    

• Antitrust Profile #3: Market Definition Matters, But Just for the Competitive Effects Storytelling

  • Here, both the government and the merging parties have versions of the relevant market that generate market shares over 30% In other words, the PNB presumption is triggered no matter what. But the merging parties offer up a much broader conception of competition. Why? Not to head off the PNB presumption. They don’t care about that. To persuade the judge that they ought to look at all of these other firms and sources of rivalry when they think about the fundamental antitrust question: will THIS merger reduce competition, increase competition, or neither?

  • Think here of something like the FTC’s recent loss in Surmodics. There, the merging parties explicitly lost on PNB but made headway in showing the relevance of a variety of present and future sources of competition – losing the PNB battle to win the competitive effects war.

  • You can also think of most vertical merger challenges as fitting this profile. Sort of. Of course, vertical mergers do not affect market shares and so the PNB presumption is technically not in play. But you will see merging parties fight vigorously on market definition for the purpose of broadening the sources of competition the judge evaluates when assessing the potential impact of the transaction on competition.

Of course, there are other antitrust risk profiles: the quasi-failing firm defense (“the acquired firm is weak and not relevant to competitive effects moving forward”), the “upside-down” market definition fight where the merging parties argue for markets sufficiently narrow to show the parties do not actually compete, and a handful of others. But those are playbook pages that are not really relevant today.

Which of these is Netflix-WBD? To my eye, it is #2. It is definitely not #1. And definitely not #3. The market definition battle straddles the PNB presumption. But – in the backdrop of antitrust analysis declaring the merger a no-brainer Clayton Act violation the wedge between DOJ and Netflix arguments regarding the post-merger market shares is really relatively small. We will talk more about that.

If you take one thing from today’s COTM have it be this: even in the DOJ’s presumptive market definition involving Streaming Video on Demand (“SVOD”) and NOTHING ELSE, the DOJ’s barely above the PNB presumption threshold at 30 percent. I’m going to spend the rest of COTM explaining whether that matters (it does) and why (it implies only weak anticompetitive effects for the merging parties to offset with other arguments).

TL;DR version: everything has to go right for the DOJ to win a merger challenge that is barely over the presumption: (1) it has to convince the court the SVOD market is correct; (2) it has to convince the court that whatever efficiencies the deal generates are not enough to offset a relatively weak presumption.

Far from the slam dunk the Hipster / Anti-Monopolist / NB caucus is selling, I think even spotting the DOJ (or State AG) their preferred market definition, this is a transaction that barely gets over the PNB presumption hurdle. Merging parties litigate and win cases like this quite often. The PNB Presumption matters. A lot. But it is not dispositive. And when it is only weakly triggered (i.e. shares just above 30) courts are very open to competitive effects stories that dispel it.

That’s the punchline for the framing here. So let’s talk a little bit about the specific risks involved in the market definition fight in a potential Netflix-WBD deal. We’ll then turn to a handful of other observations and a comparison with the relative risks if Paramount wins the bid.

Everything Changes if YouTube (or TikTok) Are in the Relevant Market. Are They?

So yes, Netflix-WBD is mostly about market definition too. Don’t get me wrong, there will be other important fights in a prospective DOJ v. Netflix litigation. In particular, the vertical aspects of the deal could give way to serious efficiencies and cost reductions that federal courts have become accustomed to granting (see e.g., AT&T / Time Warner; Tempur Sealy / Mattress Firm).

But if the DOJ does not get over the market definition hurdle the game will be immediately over in favor of the parties. So let’s talk about that. Let’s start easy. What does the competitive world look like before the merger if the market is as the DOJ wants it, i.e. SVOD services only. Here are the 2024 data from Statista. You can get slightly different shares depending upon the data service that you use. But this gives the general picture.

These data project post-merger SVOD shares at about 34%.

Some sources come out a little bit lower. Some come out a little bit higher depending upon how they measure subscriptions. So let’s not take the Statista shares without a good serving of salt. But that said – you get the sense of the ballpark of post-merger shares in conditions most favorable to the DOJ. The best case scenario for the DOJ involves post-merger shares that are probably over the PNB 30 percent trigger depending upon measurement issues and, critically, depending upon excluding all other sources of competition.

Feeling comforted about the DOJ’s case? I wouldn’t be.

But what is the alternative story? The alternative story is YouTube is a close enough competitor in the SVOD market to be included. There are others (e.g., TikTok). But let’s just focus on YouTube for a moment. Why does YouTube matter? Because by some accounts they are Netflix’s closest competitor.

How could YouTube be Netflix’s closest competitor? Well, the market definition question is, put simply, whether if a hypothetical monopolist of all SVOD services raised prices by a small but significant amount would the price increase be profitable? Or would consumers switch in sufficient numbers to the next closest substitute outside the SVOD market? If price increase would be unprofitable because consumers see the substitute as close enough to switch then regulators and courts are told to include that next substitute in the market. OK. So what is the argument that YouTube is a close enough competitor to Netflix?

Netflix certainly sees YouTube as a primal threat to its business. Netflix is competing for the viewing time and attention of customers at the end of the day. YouTube has endless content. It may be less long-lived than Netflix content (i.e. YouTube content is mostly created by users) but when users switch from Netflix to YouTube the opportunity cost to Netflix is just as significant as a consumer switching to another streaming service. As Stratchery notes, the long term competitive threat to Netflix comes precisely from the risk that YouTube has lowered barriers to entry in content creation:

That qualitative analysis is exactly the framing that Netflix will present to the court. That competition among SVOD-only does not capture the relevant rivalry the antitrust laws protect. They will also point out that YouTube is winning that rivalry and earning a greater share over its SVOD rivals over time.

But let’s return to the more technical “hypothetical monopolist test” inquiry for a moment. The above graph gives a very general sense of relative switching over time favoring YouTube – but that isn’t exactly what the HMT envisions. To test whether a rival’s product is a close enough competitor the HMT asks the specific question about whether customers would switch in response to a price increase in sufficient numbers to render the price increase unprofitable. Sometimes that question is more conceptual than quantitative (e.g. when goods are zero-priced we might think about responses to degradation in quality). Sometimes we can measure using natural experiments in the real world.

Obviously we do not have access to sufficiently granular data to run these experiments and present them here. But the merging parties’ economists do! And will! And what can we expect to see when they do? I think the recent FTC v. Meta loss is a great model for the types of evidence we can expect to see in a potentially litigated DOJ v. Netflix challenge. And not a comforting one for antitrust hawks.

The market definition debate had similar contours though the FTC challenged the Facebook-Instagram acquisition under a monopolization theory under Section 2 of the Sherman Act. The market definition question was whether TikTok and YouTube (who the FTC excluded from its narrow relevant market) was a close enough competitor. While everyone has hand-waving arguments about the functional similarities and differences in the product experiences, antitrust points us to objective economic analyses about consumer behavior. Specifically, Meta relied on evidence from a number of natural experiments to empirically identify precisely how close a competitor YouTube (and TikTok) were. Here’s just one example from Judge Boasberg’s decision discussing experiments conducted by Meta expert economist John List in which users were paid to use Facebook less or not at all:

The court found these experiments especially probative:

These weren’t the only type of evidence Meta put before the court or that the court evaluated in assessing the closeness of competition between Facebook-Instagram and rivals excluded from the FTC market definition such as YouTube and TikTok. Meta put forward evidence of customer substitution in response to a Meta outage in 2021, a ban of TikTok in India, the U.S. TikTok shutdown, and a 2018 YouTube outage. The upshot of that evidence was not just that YouTube and TikTok compete with Meta, but that the magnitude of the substitution was sufficient that the court was persuaded those products belonged in the market and were, in fact, the closest substitutes for Meta’s apps. And down went the FTC’s case:

What are the lessons from FTC v. Meta for a potential DOJ or California case against Netflix and WBD? What do they tell us about antitrust risks?

The first and most important lesson is that courts are willing and able to expand the government’s preferred market definition when the evidence warrants it. And that is true even when the products are not identical. It is absolutely true that YouTube and Netflix have different business models and a different strategy in regards to content production. But antitrust asks us not to compare the similarity and differences in the vibes offered by the products. It asks whether consumers and users of the products view them as sufficient close substitutes as a proxy for understanding whether the two companies competitively constrain one another. That is the fundamental question. The answer in Meta was that YouTube and TikTok were competitive constraints upon one another and the consumer switching data backed that up.

Here, the intuition is the same. It is not difficult to understand that YouTube is an already large and growing competitive threat to Netflix. Courts are likely to understand that. Market definition is not about formal line drawing but about understanding competitive rivalry. That favors the parties. Not the DOJ.

What counterarguments are there to including YouTube in the market and limiting it to SVOD? I have not seen anything compelling. The best Tim Wu can muster in his NY Times op-ed to destroy what cannot even be called a strawman version of the market definition argument. Wu argues, well, you have to read it yourself:

Yes, apples and oranges need not be close enough substitutes to constrain each others’ prices just because they are fruit. Same for Porsche and Honda. OK. The law is not, in fact, that stupid. Professor Wu is a smart guy making a not-so-smart argument. But what about the ACTUAL question, Professor Wu? Are these products actually close enough substitutes based upon consumer switching behavior? That takes data to answer. I do not have it all. And I do not pretend otherwise. But the competitive threat that YouTube poses for Netflix and YouTube’s growth very much suggest that Netflix is concerned about losing users to YouTube.

By the way, do you want to know who else thought that FTC’s case in Meta should involve a narrow market definition that was ultimately blown up by persuasive empirical evidence? You guessed it. So maybe take the advice on market definition analysis with a shovel full of salt and a shot of whiskey.

You can bet your last dollar that Netflix / WBD will put together the same kind of evidence concerning substitution to YouTube and closeness of competition with YouTube and others.

One difference between the FTC’s loss to Meta and the hypothetical DOJ / California case against Netflix - WBD is that the FTC’s margin for error was larger than the DOJ’s. The FTC just needed to prove that Meta was a monopolist in any relevant market and alleged Meta had a dominant share in its narrowly pled market. The DOJ, as discussed, has a BEST CASE scenario in which the PNB presumption is barely met in an SVOD-only market. If anything goes wrong for the DOJ or California (or any State AG) in terms of market definition, the PNB presumption is gone and with it any chance for the government to win.

Recall the primary lesson for the day: the government cannot win without the PNB presumption, but the parties can if it is successfully invoked. It hurts the merging parties if it is invoked — but it is not dispositive. And it is at its weakest in cases where it is barely triggered. Here, the government cannot give itself any breathing room. The best case scenario is that it pleads and proves a market share in the low 30s and barely makes the presumption.

Let’s sum up our assessment of the antitrust risks of a Netflix-WBD deal:

• Obviously, the litigation risk turns primarily upon market definition

• More interestingly, and contrary to most commentary I’ve seen discussing the deal, the DOJ/California barely can invoke the PNB presumption even assuming the best case scenario for market definition. Put simply, the government needs everything to go right to win.

• Everything usually does not go right! Namely, the DOJ is probably going to lose on market definition because YouTube should be in the market for any judge interested in assessing (as the law requires) competitive rivalry in the future after the merger

• Even if the DOJ wins on market definition, the presumption of competitive harm will be weak and potentially rebutted by evidence of efficiencies from the vertical aspects of the deal.

• I would not be very confident in a government win were I the DOJ or a State AG looking to intervene in this deal.

What About Paramount-WBD? What Would Be Different?

There are three primary differences in the relative antitrust risk profile if Paramount wins the day. First, the government loses the PNB presumption. I’ve tried to explain the best I can that this is a big deal for both the government and the merging parties but a MUCH bigger deal for the government. The government simply does not win merger cases without the presumption with very few and far between exceptions. The post-merger SVOD shares – that is, the narrowest market one can really conceive of here with a straight face – are definitely below 30 percent. The government will not get the presumption. It will not win a challenge on horizontal overlap grounds in the SVOD market. I argued above that the antitrust risk of the Netflix acquisition is lower than commentators (and perhaps, the market) believe it to be. That is true. But the relative risk of a successful SVOD market based challenge is materially lower if Paramount is the buyer.

The second interesting difference is that both Paramount and Warner Brothers are also competitors as major film studios with theatrical releases. I won’t engage in a lengthy discussion of that here. As discussed above, of course major movie producers would rather have more major film studies to pitch for theatrical releases than less. But a major government intervention on behalf of Hollywood producers and actors seems unlikely on both economic and political grounds. Further, from a pure antitrust perspective, the overarching fact that theatrical releases are on the decline for unrelated reasons makes the antitrust arguments in favor of blocking the consolidation a bit of a hail mary.

A third interesting difference is, well, Makan Delrahim is the new Chief Legal Office at Paramount. As the WSJ points out, former DOJ Assistant Attorney General Delrahim is pretty well versed in these battles. If I were picking someone to move the chess pieces on the board for a company in a major deal that involves politics, antitrust risk and analysis, as well as strategy, Makan would be one of my first calls. And not just because he’s a fellow UCLA Bruin.

So What Will Actually Happen?

I don’t know. I’m a market-oriented guy and the prediction markets tend to slightly favor Netflix closing the deal over Paramount. But it is closer than you might think if you just read the headlines.

The point of today’s COTM is NOT to pick winners or make predictions. But to make a contribution to understanding the relative antitrust risks of the deal — and hopefully fill a bit of a gap by doing so, if not add some complexity and nuance to what is out there. But I am comfortable making a prediction or two based upon what I know:

• Someone is going to sue. It may or may not be the DOJ. But the State AG’s are active and interested in antitrust as ever. I could very much imagine the California AG bringing a claim on behalf of the movie producers on the theater side whether or not the feds do so and regardless of the buyer. It might be good politics even if it is bad antitrust.

• An SVOD claim is pretty likely to lose in federal court, whoever brings the case.

We will keep our eyes open here. I’m sure there will be more to talk about! I’ll be back next week with something new I’m working on already. As always, subscribe, upgrade to paid, and send to friends. Heck, message me and tell me what I got wrong here or what you would like to see me cover next or sometime in 2026.

Until then, Happy New Year!