Merger Arb Update (HPE/Juniper)

Upcoming/recent developments… The DOJ filed a lawsuit last week seeking to enjoin the proposed merger. The
case was assigned yesterday to Northern California Federal District Court Judge P. Casey Pitts, a 2022 Biden
appointee. After graduating from Yale law school in 2008, Pitts clerked for 9th Circuit Court Judge Stephen Reinhardt
and subsequently spent most of his legal career practicing labor law at Altshuler Berzon LLP. To our knowledge, he
has not presided over an antitrust case, although while in private practice he successfully defended the Writers
Guild of America in a Section 1 Sherman Act case. Pitts is considered to be a progressive Democrat and appears to
be a favorite of the Alliance for Justice.

Our outlook… We believe the DOJ is likely to prevail in litigation, but not by a wide margin. The DOJ’s “enterprise-
grade WLAN solutions” market definition appears well supported, and per the Complaint, the merger will cause an
undue level of concentration (3000 + HHHI and 250 + change in concentration) in this market. The DOJ will
therefore be entitled to a strong HHI-based presumption of harm if the court sides with it on market definition. In
addition, the DOJ has a seemingly compelling competitive effects story. The Complaint cites several examples of
head-to-head competition between HPE and Juniper yielding benefits to customers in the form of lower prices,
which implies an upward price effect if the merger is consummated. However, the DOJ also indirectly acknowledges
that the post-merger entity and Cisco face competition from several other vendors, some of which appear to be
significant competitors as well, including Arista, CommScope, Extreme Networks, and Fortinet (second tier vendors).
And we believe the DOJ has a somewhat thin argument regarding the inability of these second tier vendors to
compete with Cisco and post-merger HPE. As such, while we favor DOJ at this early stage of litigation, this is a
winnable case for the companies.

* Based on our research and the evidence cited in the Complaint, it may be hard to dispute that “enterprise-
grade WLAN solutions” is distinct from “consumer-grade WLAN solutions.” Notably, the companies may
have acknowledged this market definition. HPE and Juniper apparently submitted information to the U.K.
CMA last year drawing distinctions between WLAN equipment used by enterprises and WLAN equipment
utilized by small and medium-size enterprises (SME) (see paragraph 27 of the CMA Decision). According to
the CMA, the companies apparently stated that larger customers required more advanced features, while
SMEs typically had less complex needs. The Complaint does not cite this evidence or documents where the
companies utilize the term “enterprise-grade WLAN solutions,” but it does identify evidence suggesting it
has distinct uses, characteristics, and customers (all of which are “Brown Shoe practical indica” used by
courts to define a relevant market). The DOJ’s market definition is not airtight, however. It is important to
note that the CMA did not uncover evidence of significant differences in the functionality or technical
specifications of WLAN equipment used by large enterprises and SMEs (see paragraph 33 of the CMA
Decision). We will therefore have to confirm whether the DOJ and IDC are effectively lumping SME WLAN
equipment into the same market as “enterprise-grade WLAN solutions,” which would mean these differences do not make a difference for purposes of obtaining a presumption of harm. However, if “consumer-grade WLAN solutions” are effectively the same as SME WLAN equipment, the DOJ will have to
combat evidence that there are not significant differences between enterprise-grade and consumer-grade
wireless solutions.

* We expect the DOJ will be able to support its market definition, which would confer it with the
fundamental advantage plaintiffs have enjoyed in litigating merger cases, namely a strong presumption of
antitrust harm. As we have discussed in prior merger cases, over the last 25 years U.S. antitrust agencies
have enjoyed a near perfect track record of success in enjoining mergers where they have been able to
establish a relevant market with a high level of post-merger market concentration. The main exceptions are
cases in which the defendants litigated a robust fix, which may not be feasible in this case (more on that
topic below).

* The DOJ cites IDC market share estimates showing that Cisco, HPE, and Juniper have a collective share of
70%. While we do not have IDC data, we understand from our research that Cisco accounts for the lion’s
share of that percentage, while HPE has market share somewhere in the mid-to-high teens and Juniper
somewhat less than 10%. While Juniper’s sub-10% market share estimate does not appear alarming, the
DOJ’s evidence, as well as our own research, suggest that this share estimate understates Juniper’s actual
competitive significance. According to industry participants that we interviewed last year, Juniper wins a
much larger percentage of contracts that are put out for bid than its market share would otherwise imply
(i.e., it is winning well more than 10% of contracts put out for bid). The Complaint also cites evidence
suggesting Juniper is playing an outsized or disruptive role in the market since it has pressured the second
largest player, HPE, to offer significant discounts (over 70% in at least one case), despite its small size. In
response to this competition, the Complaint notes that HPE developed the “Beat Mist” campaign, which
sought to garner competitive intelligence on Mist products and better train sales associates to compete with
Juniper.

* The DOJ’s decision to define a market around a product rather than a category of customers (an approach
which would have likely implicated a higher level of market concentration) presents some risks for its case
against the deal. A 70% share implies that there are a number of other competitors. The DOJ argues that
second-tier players accounting for the other 30% of the enterprise-grade WLAN market could not reposition
themselves and expand their share. The DOJ claims that other WLAN vendors will be at a disadvantage in
competing with the merged entity and Cisco because they have smaller sales forces and support
organizations and lack certain components necessary to “develop reputations for reliable services.” The
Complaint also cites underdeveloped distribution networks as an additional obstacle for smaller players to
compete. These barriers are more modest than we typically observe in horizontal merger cases.
Consequently, the defendants could mount a compelling case that second-tier vendors like Extreme
Networks, CommScope, Ubiquity, and Arista can clear these hurdles and expand/reposition themselves
within 1-2 years to better compete with Cisco and post-merger HPE.

* The defendants’ path to victory may be in getting testimony from those second-tier players that they will
be able to compete for larger accounts and fill the competitive void left by Juniper. The companies’ case
would be bolstered even further if the bidding data shows that second-tier vendors were finalists on
contract bids and won large enterprise accounts in some cases. In this regard, Extreme Networks has
publicly disclosed that it has won a few large enterprise accounts in recent years. If these smaller players
show up as finalists or winners for bids on large enterprises, the court might conclude DOJ is overstating the
competitive harm of the merger. Oracle/PeopleSoft successfully litigated against the DOJ in 2004, in part
because it was able to amass evidence showing that smaller enterprise software vendors were getting
traction with large enterprise accounts. For what it is worth, that case was also litigated in the Federal
District Court of Northern California.

* We are much more dubious of the DOJ’s coordinated effects claims. It is very difficult to prove up a
coordinated effects case that does not involve commodity products with transparent pricing. Without more
visibility into prices, it will likely be hard for DOJ to convince the court that Cisco and HPE-post-merger could
detect deviations from coordinated conduct. This argument strikes us as a “throw in” claim and perhaps an
effort to argue that smaller enterprise-grade WLAN vendors will behave less competitively post-merger if
competitive pressure from Juniper is eliminated.

* We remain skeptical of a settlement. It is our expectation that incoming DOJ antitrust chief Gail Slater will
support this case and will likely be weary of any proposed remedy that falls short of a divestiture of HPE or
Juniper’s WLAN business (assuming they are severable, which is still an open question in our view). Our
research so far indicates that HPE’s WLAN solutions business may account for somewhere in the range of
$1-1.5 billion in revenue per year, and Juniper WLAN products generated around $400-$500 million in 2023.
While a divestiture of Juniper’s assets in this market may be immaterial from a revenue perspective, the
acquisition of Juniper’s Mist AI technology is a large part of the rationale for the deal, so it seems unlikely
HPE would consider divesting that business. It is possible that the companies could arrange a divestiture
comprised of a mixture of assets of both companies and include the Aruba brand as opposed to Mist, but
such a fix may be deemed overly complex by the Department.

Watch for these developments… The companies’ Answer to the Complaint will likely be filed in the coming weeks.
We also expect that the court will soon schedule a conference to discuss scheduling matters and the trial date.
Once the schedule is set, we will be eagerly anticipating the filing of the witness list (which may be not be available
until shortly before trial starts). A witness list will be confirmatory as to whether the DOJ has succeeded in getting
customers to testify against the deal.