Revisiting Philadelphia National Bank: The Structural Presumption and the Future of Merger Review

WEBVTT
1
00:00:01.590 --> 00:00:06.209
Welcome to FedSoc Forums, a Federalist Society podcast. I'm Elizabeth Slattery,
2
00:00:06.360 --> 00:00:09.880
Vice President of Practice Groups. Be sure to subscribe wherever
3
00:00:09.919 --> 00:00:13.419
you get your podcasts so you won't miss an episode. Now,
4
00:00:13.480 --> 00:00:15.380
please enjoy today's conversation.
5
00:00:17.719 --> 00:00:22.320
Hello, everyone, and welcome to FedSoc Forums. Today, September 15th, 2026,
6
00:00:22.320 --> 00:00:27.699
we're delighted to host a conversation entitled Revisiting Philadelphia National Bank,
7
00:00:27.859 --> 00:00:31.980
the Structural Presumption and the Future of Merger Review. My
8
00:00:32.020 --> 00:00:34.299
name is Emma Blakemore, and I'm an Associate Director of
9
00:00:34.369 --> 00:00:37.869
Practice Groups at the Federal Society. As always, please note
10
00:00:37.909 --> 00:00:40.409
that all expressions of opinion are those of the experts
11
00:00:40.469 --> 00:00:43.719
on today's call, as the Federal Society takes no position
12
00:00:43.850 --> 00:00:48.079
on particular legal or public policy issues. We're very pleased
13
00:00:48.119 --> 00:00:50.359
to welcome Michael Murray as our moderator today.
14
00:00:50.859 --> 00:00:51.100
Mr.
15
00:00:51.140 --> 00:00:54.130
Murray is co-chair of the antitrust and competition practice at
16
00:00:54.149 --> 00:00:57.789
Paul Hastings and represents clients in high stakes mergers and
17
00:00:57.829 --> 00:01:03.750
acquisitions and antitrust litigation and investigations. Based in the firm's Washington, D.C. office,
18
00:01:04.150 --> 00:01:07.719
Michael is Chambers ranked and brings significant government antitrust and competition.
19
00:01:07.719 --> 00:01:12.000
competition experience, having most recently served as the principal deputy
20
00:01:12.060 --> 00:01:15.420
assistant attorney general in the antitrust division at the Department
21
00:01:15.439 --> 00:01:19.090
of Justice, where he oversaw the work of approximately 350
22
00:01:19.090 --> 00:01:23.890
lawyers and 50 economists. In that capacity, Michael supervised DOJ's
23
00:01:23.930 --> 00:01:29.260
merger investigations, conduct matters, and litigation for several years. I
24
00:01:29.280 --> 00:01:32.079
will leave it there, but one last note throughout the panel.
25
00:01:32.459 --> 00:01:34.659
If you have any questions, we ask that you please
26
00:01:34.700 --> 00:01:37.920
submit them through the question and answer feature below, not
27
00:01:37.939 --> 00:01:39.909
the chat, so that we will have access to them
28
00:01:39.950 --> 00:01:42.269
when we get to that portion of the webinar. We
29
00:01:42.290 --> 00:01:44.670
do ask that questions submitted there both pertain to this
30
00:01:44.730 --> 00:01:47.750
discussion and end with a question mark. And with that,
31
00:01:47.810 --> 00:01:49.909
thank you for being with us today, and I will
32
00:01:49.950 --> 00:01:52.730
turn things over to our speakers to start the conversation.
33
00:01:53.209 --> 00:01:54.810
Thank you, Mr. Murray, for moderating.
34
00:01:55.769 --> 00:01:58.209
Thank you, Emma, and I'm pleased to be here today.
35
00:01:58.269 --> 00:02:00.859
This is a really all-star panel that we have with
36
00:02:00.900 --> 00:02:05.340
some of the foremost thinkers and scholars and practitioners in
37
00:02:05.799 --> 00:02:10.430
antitrust law today, with decades of experience discussing the question
38
00:02:10.469 --> 00:02:14.490
that we're going to discuss today. I'll do brief introductions,
39
00:02:14.509 --> 00:02:16.889
a quick level set on what we're talking about, and
40
00:02:16.930 --> 00:02:19.169
then jump right in so that we can hear from
41
00:02:19.930 --> 00:02:22.960
our panelists. First, Debbie Feinstein is one of the leading
42
00:02:23.000 --> 00:02:26.000
antitrust lawyers in the country. She's a partner in Arnold
43
00:02:26.039 --> 00:02:28.759
and Porter, and she served as the director of the
44
00:02:28.800 --> 00:02:33.080
Bureau of Competition at the FTC from 2013 to 2017.
45
00:02:33.080 --> 00:02:38.389
We also have Judge Ginsburg, who is a senior judge
46
00:02:38.469 --> 00:02:38.969
on the D.C.
47
00:02:39.009 --> 00:02:39.370
Circuit.
48
00:02:39.689 --> 00:02:42.469
He previously served as assistant attorney.
49
00:02:42.289 --> 00:02:44.889
General of the antitrust division and is, of course, an
50
00:02:44.930 --> 00:02:48.169
accomplished scholar and professor at multiple universities.
51
00:02:49.000 --> 00:02:51.879
We have Herbert Hovenkamp, who's the James G.
52
00:02:51.919 --> 00:02:54.800
Dinan University Professor at the University of Pennsylvania Law School.
53
00:02:55.560 --> 00:02:58.469
The New York Times called him the Dean of Antitrust Law,
54
00:02:58.810 --> 00:03:02.229
and he's authored numerous case books and treatises on antitrust law.
55
00:03:02.849 --> 00:03:06.129
We also have Professor Daniel Sokol, who's the Carolyn Craig
56
00:03:06.169 --> 00:03:09.009
Franklin Chair in Law and a Professor of Law and
57
00:03:09.050 --> 00:03:11.770
Business at the USC Gold School of Law and Marshall
58
00:03:11.789 --> 00:03:15.229
School of Business in the Marketing Department. Last but not least,
59
00:03:15.250 --> 00:03:17.449
we have Rahul Rao, who is a partner at White
60
00:03:17.530 --> 00:03:19.870
& Case. He served as a deputy director at the
61
00:03:19.949 --> 00:03:24.120
FTC from 2022 to 2025 and also in the Washington
62
00:03:24.129 --> 00:03:29.460
State AG's office from 2018 to 2022. I'd also like
63
00:03:29.500 --> 00:03:32.719
to thank Svetlana Gans for helping to organize this panel.
64
00:03:33.620 --> 00:03:36.400
So today we're talking about the Philadelphia National Bank case.
65
00:03:37.099 --> 00:03:41.870
It's a 1963 Supreme Court case that establishes a structural presumption.
66
00:03:42.430 --> 00:03:44.610
If a merger creates a single entity controlling 30% or
67
00:03:44.650 --> 00:03:48.849
more of the market, it's legally presumed to lessen competition
68
00:03:48.900 --> 00:03:50.840
in violation of Section 7 of.
69
00:03:50.819 --> 00:03:51.400
The Clayton Act.
70
00:03:51.979 --> 00:03:55.360
Federal and state enforcers have been challenging mergers, relying on
71
00:03:55.379 --> 00:04:00.840
this presumption for quite some time since the decision in 1963,
72
00:04:00.840 --> 00:04:02.960
and certain state attorneys generals have also relied on it,
73
00:04:02.960 --> 00:04:07.120
perhaps most recently in the Paramount case.
74
00:04:07.139 --> 00:04:09.520
The 2023 guidelines also cite the decision.
75
00:04:11.030 --> 00:04:15.110
Positively and rely on it in developing the guidelines.
76
00:04:14.689 --> 00:04:17.089
That came out in 2023.
77
00:04:17.089 --> 00:04:19.889
Today, we're going to talk about several topics related to
78
00:04:19.930 --> 00:04:22.569
the case. The first is whether it's the right approach
79
00:04:22.610 --> 00:04:27.800
or not, whether economic developments have changed, or whether Philadelphia
80
00:04:27.819 --> 00:04:30.360
National Bank, as originally conceived or some version.
81
00:04:30.079 --> 00:04:32.660
Of it, is the right approach for antitrust law.
82
00:04:33.379 --> 00:04:36.459
Second, we'll talk about what courts and agencies should do
83
00:04:36.660 --> 00:04:40.009
in light of the foregoing discussion. And third, for those
84
00:04:40.050 --> 00:04:43.129
private practitioners on the line, we'll talk about counseling private parties,
85
00:04:44.170 --> 00:04:47.430
given the foregoing discussion. And with that, I'll turn it
86
00:04:47.509 --> 00:04:51.850
over first for the first question, which is whether Philadelphia
87
00:04:51.870 --> 00:04:55.009
National Bank or a version of Philadelphia National Bank is
88
00:04:55.290 --> 00:04:56.370
the right approach.
89
00:04:56.810 --> 00:05:00.790
I'll start with Professor Hovenkamp. You've recently published a.
90
00:05:02.269 --> 00:05:07.209
Post on your sub-stack entitled The Rehabilitation of Philadelphia National Bank,
91
00:05:08.000 --> 00:05:10.720
in which you state, among other things, that recent scholarship
92
00:05:11.389 --> 00:05:15.480
provides compelling evidence of a correlation between increases in concentration
93
00:05:15.980 --> 00:05:16.800
that result.
94
00:05:16.519 --> 00:05:18.899
From a merger and higher product prices.
95
00:05:19.420 --> 00:05:21.279
I'm wondering if you could unpack that for a little bit,
96
00:05:21.300 --> 00:05:25.430
unpack that for us for a little bit and describe
97
00:05:25.490 --> 00:05:27.529
your recent thinking on Philadelphia National Bank.
98
00:05:28.449 --> 00:05:29.430
Yeah, thank you.
99
00:05:29.509 --> 00:05:31.769
And thank you very much for inviting me to this panel.
100
00:05:33.290 --> 00:05:37.990
Philadelphia Bank came only one year after the Supreme Court's
101
00:05:38.009 --> 00:05:42.839
decision in Brown Shoe, but it dramatically turned the direction
102
00:05:42.879 --> 00:05:46.240
of merger policy. Brown Shoe had actually condemned a merger
103
00:05:46.980 --> 00:05:51.490
because it resulted in lower prices that undersold smaller rivals.
104
00:05:52.129 --> 00:05:54.430
And Philadelphia Bank got.
105
00:05:55.959 --> 00:06:01.310
shifted course to focus on what the then prevailing Harvard
106
00:06:01.350 --> 00:06:05.629
School regarded as the true effects of higher concentration, which
107
00:06:05.670 --> 00:06:10.370
were higher profits and prices, stagnant markets, and so on.
108
00:06:10.829 --> 00:06:15.769
And to do that, the Philadelphia Bank decision actually developed
109
00:06:15.860 --> 00:06:17.459
two different presumptions.
110
00:06:17.740 --> 00:06:18.779
I'll just quote it here.
111
00:06:18.800 --> 00:06:23.120
The court spoke of a merger that, quote, produces a
112
00:06:23.279 --> 00:06:28.100
firm controlling an undue percentage of the relevant market and
113
00:06:28.180 --> 00:06:31.779
results in a significant increase in the concentration of.
114
00:06:31.800 --> 00:06:32.980
Firms in that market.
115
00:06:33.560 --> 00:06:37.930
So one was the absolute level of concentration, and the
116
00:06:37.990 --> 00:06:40.930
court held that the 30% market share in the case
117
00:06:40.970 --> 00:06:50.639
satisfied that. The other one was the significant increase in concentration.
118
00:06:52.120 --> 00:06:55.019
The court didn't use the HHI, which was not in
119
00:06:55.100 --> 00:07:00.100
common use among judges and lawyers at that time, but
120
00:07:00.139 --> 00:07:04.430
it did use the two-firm concentration ratio and found a
121
00:07:04.490 --> 00:07:06.949
significant increase.
122
00:07:07.209 --> 00:07:11.649
Now, what's happened to those two different presumptions? Number one.
123
00:07:13.019 --> 00:07:17.410
The 30 percent presumption, even though it's widely associated with
124
00:07:17.430 --> 00:07:21.209
the Philadelphia bank case, never really got a lot of traction.
125
00:07:22.949 --> 00:07:26.170
Judge Ginsburg has written about it, and I think he
126
00:07:26.209 --> 00:07:29.600
and I are in agreement about that. The 30 percent
127
00:07:30.069 --> 00:07:33.680
presumption was not mentioned in previous editions of the merger
128
00:07:33.699 --> 00:07:36.459
guidelines prior to 2023.
129
00:07:36.459 --> 00:07:39.240
When it was revived and, uh, it didn't get a
130
00:07:39.259 --> 00:07:40.540
lot of traction in the course.
131
00:07:41.100 --> 00:07:44.819
On the other hand, the requirement of a significant increase
132
00:07:44.870 --> 00:07:50.790
in concentration has, it's mentioned in every single edition of
133
00:07:50.829 --> 00:07:56.350
the merger guidelines, different numbers and different guidelines, but nevertheless,
134
00:07:56.389 --> 00:08:04.069
it's always been an important, uh, important feature. Uh, and, uh, Then,
135
00:08:06.069 --> 00:08:08.560
roughly three years ago, we've now had a couple of
136
00:08:08.639 --> 00:08:12.980
very important papers, both of them published in the American
137
00:08:13.000 --> 00:08:17.319
Economics Review, which is one of the top five American
138
00:08:18.439 --> 00:08:24.329
economics journals, who have shown pretty persuasively that there are
139
00:08:24.430 --> 00:08:29.370
high correlations between Not with the absolute level of concentration.
140
00:08:29.589 --> 00:08:35.830
So this is not an imprimatur on the 30% rule,
141
00:08:36.330 --> 00:08:40.710
but rather a high degree of concentration between price increases
142
00:08:41.250 --> 00:08:44.000
and the increase or the delta increase.
143
00:08:45.029 --> 00:08:46.460
Of, of concentration.
144
00:08:46.480 --> 00:08:51.000
Uh, what Naki and Winston in the first paper concluded, uh,
145
00:08:52.519 --> 00:08:58.000
was that, uh, the current, the then existing merger guidelines,
146
00:08:58.039 --> 00:09:03.929
those were 2010 were too, uh, too conservative because, uh,
147
00:09:04.429 --> 00:09:09.289
their concentration numbers did not reach far enough. Uh, and
148
00:09:09.549 --> 00:09:14.860
they have more or less concluded that, uh, an HHI
149
00:09:15.019 --> 00:09:21.759
increase of roughly 100 points would offset efficiency gains in
150
00:09:21.779 --> 00:09:25.490
the 2% to 3% range. The question as they pose
151
00:09:25.549 --> 00:09:31.129
it is, how much efficiency gain is necessary to offset
152
00:09:31.149 --> 00:09:37.139
a particular increase in the HHI. And that's the number
153
00:09:37.200 --> 00:09:39.960
they came up with. And then in this more recent
154
00:09:40.019 --> 00:09:43.460
study that just came out in the American Economic Review
155
00:09:44.620 --> 00:09:48.850
by Bacharia and a group of his co-authors, they looked
156
00:09:48.879 --> 00:09:52.649
at a fairly significant number of mergers, nearly 100 mergers
157
00:09:52.750 --> 00:09:56.990
in fairly highly concentrated markets. And they also found that
158
00:09:57.009 --> 00:10:03.740
there was really no good correlation between price increases and
159
00:10:03.759 --> 00:10:07.679
the absolute level of concentration, but that there was a
160
00:10:07.860 --> 00:10:12.700
pretty good correlation between price increases and the change in
161
00:10:12.759 --> 00:10:19.139
increases in concentration. So I regard this as a important,
162
00:10:19.240 --> 00:10:25.809
but yet rather early stage development. Most merger retrospective studies
163
00:10:26.250 --> 00:10:31.909
have concentrated mainly on absolute levels of concentration. I think
164
00:10:31.929 --> 00:10:33.669
we need to spend a little bit more time in
165
00:10:33.690 --> 00:10:37.299
the future looking at changes in concentration. You can get
166
00:10:37.340 --> 00:10:41.299
that from data and some of the retrospective studies, but
167
00:10:41.320 --> 00:10:44.480
it has the potential to simplify merger litigation. If you
168
00:10:44.500 --> 00:10:47.750
can start focusing on, you have to define a market
169
00:10:47.769 --> 00:10:52.210
and compute market shares, but if you can focus on
170
00:10:52.289 --> 00:10:57.279
the increase in concentration as between the merging parties, we
171
00:10:57.320 --> 00:10:57.879
might get.
172
00:10:57.779 --> 00:10:59.639
A reliable rule.
173
00:11:00.779 --> 00:11:05.100
That equates that change with predicted price increases.
174
00:11:05.519 --> 00:11:07.539
Okay, now I know I've gone a long time, but
175
00:11:07.600 --> 00:11:09.000
just one concluding sentence.
176
00:11:09.570 --> 00:11:12.909
All of this is premised on the idea that the
177
00:11:12.970 --> 00:11:19.309
principal concern of merger policy is mergers that threaten price increases.
178
00:11:20.120 --> 00:11:22.750
If we want to challenge mergers for some other reason,
179
00:11:22.809 --> 00:11:27.269
and we should, mergers that threaten to reduce innovation or
180
00:11:27.330 --> 00:11:31.330
something like that, then these studies don't speak as directly
181
00:11:31.409 --> 00:11:33.049
to those types of conclusions.
182
00:11:34.429 --> 00:11:36.629
Thank you. Thank you, Professor.
183
00:11:37.110 --> 00:11:40.870
There was a lot in there to further think about.
184
00:11:40.929 --> 00:11:43.909
And first, I'd like to turn to Professor Sokol. I
185
00:11:43.929 --> 00:11:47.750
know that several years ago, you wrote that your view
186
00:11:47.830 --> 00:11:51.110
is that Philadelphia National Bank should evolve. And I'm curious
187
00:11:51.149 --> 00:11:52.830
if you think that some of these recent studies that
188
00:11:52.850 --> 00:11:55.889
Professor Hovenkamp mentioned, and maybe other developments as well, are
189
00:11:55.909 --> 00:11:56.850
sort of consistent with.
190
00:11:56.789 --> 00:12:00.830
That view, and if you'd like to react to that. Sure.
191
00:12:01.750 --> 00:12:02.789
So first of all, thank you.
192
00:12:03.049 --> 00:12:04.990
This is actually the very first time I've been a
193
00:12:05.049 --> 00:12:08.009
speaker at a Federalist Society event, so that's very exciting.
194
00:12:08.769 --> 00:12:11.129
A few thoughts.
195
00:12:11.350 --> 00:12:15.980
First of all, if Philadelphia National Bank were to come
196
00:12:16.019 --> 00:12:18.559
up today in front of the court, I think we
197
00:12:18.600 --> 00:12:21.740
could agree it would not come out the same way.
198
00:12:22.080 --> 00:12:24.159
There's a lot more that we know, and.
199
00:12:25.779 --> 00:12:28.220
Some of the things that animated judges in the 1960s
200
00:12:28.340 --> 00:12:31.879
is not what we think about today. Do I think
201
00:12:31.899 --> 00:12:35.139
that there's some aspects of it that are still very useful? Actually,
202
00:12:35.179 --> 00:12:37.759
like Herb, I think the answer is yes. Are there
203
00:12:37.779 --> 00:12:40.000
things that still need to evolve? I think the answer
204
00:12:40.039 --> 00:12:45.840
is also yes. So I'll give a defense to having
205
00:12:45.879 --> 00:12:48.620
a structural presumption of some sort. It's a good starting
206
00:12:48.720 --> 00:12:53.799
point for analysis. Unfortunately, in the 60s, it was a
207
00:12:53.840 --> 00:12:55.600
starting and end point for.
208
00:12:55.519 --> 00:12:59.460
The most part. So if we think about.
209
00:13:01.029 --> 00:13:02.769
What does a presumption tell us? It gives us a
210
00:13:02.769 --> 00:13:08.600
little bit of a roadmap for what's safe, what might
211
00:13:08.629 --> 00:13:14.360
be contested, and what might be highly contested, high risk.
212
00:13:15.080 --> 00:13:18.460
And if we think from a risk matrix type perspective,
213
00:13:18.759 --> 00:13:23.059
I think business people want to have clarity. That said,
214
00:13:24.139 --> 00:13:27.320
does the 30% make sense? No.
215
00:13:27.919 --> 00:13:28.840
Artificially low.
216
00:13:30.590 --> 00:13:34.309
All of this also presumes that in the second part
217
00:13:34.350 --> 00:13:37.110
of the presumption, in terms of what the parties show, entry,
218
00:13:37.149 --> 00:13:39.230
for example, that I think we have a lot of
219
00:13:39.269 --> 00:13:43.870
clarity on. I actually feel that, unfortunately, we still don't
220
00:13:43.889 --> 00:13:46.429
pay enough attention to efficiencies. Now, there are some times
221
00:13:46.450 --> 00:13:51.730
where emerging parties might invent efficiencies, but it seems to me,
222
00:13:52.490 --> 00:14:00.929
borne out by literally since the 185 different efficiency studies
223
00:14:01.009 --> 00:14:09.200
across FT50 journals plus strong IO journals like Rand or JIE.
224
00:14:10.159 --> 00:14:12.100
There's a lot more that we know about.
225
00:14:12.500 --> 00:14:15.419
How we measure efficiencies, how do we think about efficiencies
226
00:14:16.299 --> 00:14:17.409
that hasn't.
227
00:14:17.129 --> 00:14:20.090
Really been embraced by the agencies or the.
228
00:14:20.090 --> 00:14:24.740
Judges and because the kind of cases that involve efficiencies
229
00:14:24.799 --> 00:14:27.899
are so rare that this should be the most important thing.
230
00:14:27.919 --> 00:14:32.379
Those are the deals that basically go through. So we
231
00:14:32.419 --> 00:14:35.620
don't have a good sense of how to conceptualize efficiencies.
232
00:14:35.659 --> 00:14:39.309
And I don't think that the 2023 merger guidelines did
233
00:14:39.330 --> 00:14:41.370
a good job with that. It certainly was not in
234
00:14:41.409 --> 00:14:45.450
line with what we know, but it was aspirational based
235
00:14:45.570 --> 00:14:50.750
on the prior administration. Where I think Philadelphia National Bank did,
236
00:14:58.620 --> 00:15:02.460
increase for related reasons. We know also there's another strand
237
00:15:02.500 --> 00:15:07.360
of literature most recently covered by Ali Rokoglu and Carl
238
00:15:07.379 --> 00:15:12.529
Shapiro saying we see rising concentration. It's still unclear what
239
00:15:12.590 --> 00:15:15.490
that means. The concentration itself is not the problem.
240
00:15:15.509 --> 00:15:16.470
It's what happens.
241
00:15:17.009 --> 00:15:22.070
I think as Herb rightly said, in more concentrated markets
242
00:15:22.090 --> 00:15:24.850
with the other kinds of things that might happen, particularly
243
00:15:24.889 --> 00:15:27.610
on the price side. Herb is also right that this
244
00:15:27.669 --> 00:15:32.250
is not a particularly great proxy for measuring innovation-related effects.
245
00:15:32.789 --> 00:15:35.919
So does it need to evolve? I think yes. I
246
00:15:35.960 --> 00:15:39.340
think we've already seen evolution over the years.
247
00:15:39.799 --> 00:15:43.620
I think Baker Hughes is the most important evolution of
248
00:15:43.679 --> 00:15:45.179
Philadelphia National Bank.
249
00:15:45.240 --> 00:15:45.389
And I.
250
00:15:47.639 --> 00:15:50.549
And I'd like to see further evolution that reflects a
251
00:15:50.590 --> 00:15:54.090
little more business and economic reality.
252
00:15:56.889 --> 00:15:58.289
But I also want to make very.
253
00:15:58.190 --> 00:16:02.570
Clear, without this kind of presumption, where we set the presumption,
254
00:16:02.590 --> 00:16:04.750
I want to leave aside. But without presumption, it's hard
255
00:16:04.769 --> 00:16:07.970
for judges. It's hard for the agencies. It's hard for
256
00:16:07.990 --> 00:16:15.419
the agencies to really bring cases. when there's an asymmetry
257
00:16:15.500 --> 00:16:20.039
in terms of what they know and the resources that
258
00:16:20.080 --> 00:16:22.440
they have relative to the emerging parties. And I think
259
00:16:22.799 --> 00:16:26.450
from that standpoint, Philadelphia National Bank has been a success.
260
00:16:27.370 --> 00:16:29.110
Now it's just paring it back to sort of what
261
00:16:29.269 --> 00:16:30.049
it can do well.
262
00:16:35.250 --> 00:16:37.159
Thanks, Professor Sokol. That's a great question.
263
00:16:38.730 --> 00:16:40.769
A good tour, I think, of some of what has
264
00:16:40.809 --> 00:16:44.100
happened in the past few decades. Judge Ginsburg, I know
265
00:16:44.120 --> 00:16:48.259
that you've been a long-term critic of Philadelphia National Bank.
266
00:16:48.580 --> 00:16:51.779
The article that caught my attention was Philadelphia National Bank,
267
00:16:51.840 --> 00:16:55.769
Bad Economics, Bad Law, Good Riddance, which was eloquently titled.
268
00:16:56.370 --> 00:16:59.190
I'm curious, what do you think of the recent scholarship?
269
00:16:59.950 --> 00:17:03.629
And Professor Sokol, a comment just at the end that
270
00:17:03.690 --> 00:17:06.099
Philadelphia National Bank has been a success.
271
00:17:10.980 --> 00:17:16.309
The 30% presumption is a product of its times, and
272
00:17:16.349 --> 00:17:23.329
particularly a derivative of the structured conduct performance paradigm. Economists
273
00:17:23.809 --> 00:17:28.630
no longer go to graduate school to learn the structured
274
00:17:28.670 --> 00:17:30.599
conduct performance paradigm.
275
00:17:30.630 --> 00:17:31.940
It's an historical relic.
276
00:17:33.190 --> 00:17:37.509
And I was surprised to hear Professor Hovenkamp say that
277
00:17:37.569 --> 00:17:41.450
it never really got much traction because it didn't show
278
00:17:41.640 --> 00:17:47.480
up in the merger guidelines until recently. But it's been
279
00:17:47.519 --> 00:17:54.829
invoked 122 times in cases brought by the agencies and
280
00:17:55.049 --> 00:18:01.549
dispelled only 16 times. It's been a very powerful aspect.
281
00:18:03.079 --> 00:18:04.200
Of merger litigation.
282
00:18:07.039 --> 00:18:13.990
From 82 through 92 guidelines, 97, 2010, there was essentially
283
00:18:14.029 --> 00:18:19.569
a bipartisan agreement that shares and concentration matter less and
284
00:18:19.609 --> 00:18:21.470
direct effects matter more.
285
00:18:22.680 --> 00:18:24.079
And from that.
286
00:18:24.009 --> 00:18:28.970
Consensus emerged empirical metrics that really made the consensus possible.
287
00:18:29.589 --> 00:18:37.779
Empirical metrics like gross upward pressure, pricing pressure, diversion ratios,
288
00:18:39.039 --> 00:18:44.059
greater attention to natural experiments, where that's possible, but this
289
00:18:44.339 --> 00:18:50.369
bipartisan convention was abandoned in the 2023 guidelines, which reverted
290
00:18:50.390 --> 00:18:57.109
to the structural thresholds. Professor Sokol said, does 30% make sense?
291
00:18:57.349 --> 00:18:58.380
No, he said.
292
00:18:58.539 --> 00:19:02.019
Well, if it doesn't make sense, let's think about what
293
00:19:02.059 --> 00:19:06.259
we can do that does make sense. I think this
294
00:19:06.279 --> 00:19:10.000
is an important point that bears repeating from time to time.
295
00:19:11.420 --> 00:19:15.299
The P & B presumption is a rule of law.
296
00:19:15.619 --> 00:19:19.650
It's not just a processing guideline for the agencies, as
297
00:19:19.690 --> 00:19:21.789
to which maybe it would make some sense in terms
298
00:19:21.829 --> 00:19:25.859
of how they allocate their investigative resources. But it's a
299
00:19:25.900 --> 00:19:29.619
rule of law. And the agencies seem to have invited
300
00:19:29.680 --> 00:19:32.789
the courts and everyone else to sort of confuse the two.
301
00:19:34.420 --> 00:19:39.079
PI briefs commonly cite the presumption and offer evidence of
302
00:19:39.140 --> 00:19:43.829
market shares and say, now we've made our prima facie
303
00:19:43.869 --> 00:19:49.869
case and quite commonly succeed with that in a PI context.
304
00:19:51.779 --> 00:19:58.859
So the P & B presumption is share-based, OK? And
305
00:19:59.579 --> 00:20:03.369
the share of the merged company, as Professor Hovenkamp brought out,
306
00:20:03.809 --> 00:20:07.130
talking about undue concentration for that combined share.
307
00:20:09.380 --> 00:20:11.000
Just keep in mind what it's saying.
308
00:20:11.700 --> 00:20:16.240
Observing that a post-merger share is going to exceed 30%
309
00:20:16.240 --> 00:20:19.779
implies that it's more likely than not, the civil preponderance standard,
310
00:20:19.799 --> 00:20:25.029
that it's anti-competitive. And it just isn't empirical evidence to
311
00:20:25.049 --> 00:20:30.150
support that. We have better techniques, the agencies in particular
312
00:20:30.190 --> 00:20:34.769
have better techniques and a better understanding of the world
313
00:20:35.190 --> 00:20:39.450
than that implies. So we get this rigid judicial shortcut
314
00:20:39.650 --> 00:20:44.890
that inappropriately shifts the burden of production from where it belongs,
315
00:20:45.410 --> 00:20:48.089
with the agencies and the private plaintiffs, to the defendants.
316
00:20:50.299 --> 00:20:53.500
By the time the defendants are allowed to rebut, the
317
00:20:53.539 --> 00:20:57.680
court has already been told that they're more likely than
318
00:20:57.700 --> 00:21:01.779
not going to be to lose. Just think about this
319
00:21:01.819 --> 00:21:07.519
situation without the presumption. The agencies bring in their best evidence,
320
00:21:08.960 --> 00:21:15.220
market definition, shares, HHIs, documents, guppies, unilateral effects, efficiencies, whatever
321
00:21:15.259 --> 00:21:18.640
they think is relevant to their case. And the defendant
322
00:21:18.660 --> 00:21:21.849
responds in kind. And the court weighs the totality of
323
00:21:21.890 --> 00:21:25.950
that evidence. Without the presumption, the party with the best
324
00:21:26.049 --> 00:21:26.869
evidence wins.
325
00:21:26.950 --> 00:21:29.049
No one gets an early advantage.
326
00:21:29.640 --> 00:21:30.460
In the contest.
327
00:21:31.720 --> 00:21:35.099
And the idea that plaintiffs typically show up in court
328
00:21:36.819 --> 00:21:40.769
and the agencies with more evidence than shares in HHIs
329
00:21:40.849 --> 00:21:45.509
is really not helpful. They get the presumption regardless of
330
00:21:45.549 --> 00:21:49.880
the evidence that they actually produce. So it just seems
331
00:21:49.900 --> 00:21:54.180
to me backward in the most obvious ways to continue
332
00:21:54.200 --> 00:21:57.019
to rely on this. Now, her, Professor Hovenkamp has introduced
333
00:21:57.900 --> 00:22:02.700
an interesting alternative based on some research that is provisional,
334
00:22:02.740 --> 00:22:06.160
let's say, apart from conclusive at this point. And I
335
00:22:06.180 --> 00:22:08.369
have no problem with seeing that. In fact, I would
336
00:22:08.390 --> 00:22:11.089
like to see that explored further in terms of what
337
00:22:11.109 --> 00:22:14.430
the empirics actually show us. It may prove useful and
338
00:22:14.450 --> 00:22:19.309
divert our attention to the delta in HHIs if that
339
00:22:19.509 --> 00:22:23.400
proves to be more indicative. But a presumption is an
340
00:22:23.460 --> 00:22:26.240
entirely different animal, and it's one that should be really
341
00:22:26.339 --> 00:22:29.579
foreign to litigation over mergers.
342
00:22:31.829 --> 00:22:33.150
Thanks, Judge Ginsburg.
343
00:22:33.170 --> 00:22:36.380
I think you've introduced a couple additional topics that I
344
00:22:36.440 --> 00:22:39.880
suspect Debbie and Rahul want to react to.
345
00:22:41.019 --> 00:22:42.980
Well, maybe we'll start with Debbie first.
346
00:22:43.259 --> 00:22:46.119
One of the things that I think Professor Hobenkamp mentioned
347
00:22:46.160 --> 00:22:49.089
was You know, what is the goal of merger policy?
348
00:22:49.420 --> 00:22:51.950
How does this debate relate to that? And I think
349
00:22:52.190 --> 00:22:54.329
that another topic that we've been talking about is whether
350
00:22:54.349 --> 00:22:58.670
there's parts of Philadelphia National Bank that should be kept.
351
00:22:58.690 --> 00:23:00.440
Other parts perhaps should evolve.
352
00:23:01.140 --> 00:23:04.200
I'm curious, based on your experience in private practice and government,
353
00:23:04.559 --> 00:23:07.720
how you sort of react to both of those two
354
00:23:07.759 --> 00:23:09.980
ideas and, frankly, anything else that was said by our
355
00:23:10.049 --> 00:23:11.250
esteemed panelists today.
356
00:23:11.710 --> 00:23:11.990
Thanks.
357
00:23:12.009 --> 00:23:13.109
Well, it's a pleasure to be here.
358
00:23:14.190 --> 00:23:17.359
So I did, like Danny, read the two cases that
359
00:23:17.400 --> 00:23:20.940
I read this morning were Philadelphia National Bank.
360
00:23:20.980 --> 00:23:24.160
And Baker Hughes, because I do think they sort of
361
00:23:25.049 --> 00:23:26.230
make the two points.
362
00:23:26.250 --> 00:23:27.029
So I would say this.
363
00:23:27.230 --> 00:23:31.049
So first, neither at the FTC nor in private practice
364
00:23:31.089 --> 00:23:34.410
do I utter the phrase Philadelphia National Bank or presumption
365
00:23:34.559 --> 00:23:35.970
or 30 percent.
366
00:23:35.579 --> 00:23:36.319
To most clients.
367
00:23:37.680 --> 00:23:38.019
Right.
368
00:23:38.119 --> 00:23:42.140
Because we all know that most deals with a 30
369
00:23:42.140 --> 00:23:45.400
percent market share don't get looked at, don't get a
370
00:23:45.440 --> 00:23:46.420
second request.
371
00:23:46.980 --> 00:23:47.640
Aren't issues.
372
00:23:47.660 --> 00:23:53.680
You know, usually it's much higher market shares than that.
373
00:23:53.819 --> 00:23:56.839
It is true that deals under 30 percent rarely get blocked.
374
00:23:56.880 --> 00:23:59.059
That is something that when we talk about 30 percent,
375
00:23:59.099 --> 00:24:00.420
it's almost a safe harbor, right?
376
00:24:01.460 --> 00:24:03.890
I think the two examples to the contrary that I
377
00:24:03.910 --> 00:24:04.269
can think.
378
00:24:04.250 --> 00:24:07.910
Of are Hewlett Packard Juniper and H & R Block.
379
00:24:08.210 --> 00:24:12.000
Those were both eminently three to two mergers once you
380
00:24:12.059 --> 00:24:14.240
agreed on the market definition, which I'll get to in
381
00:24:14.259 --> 00:24:14.539
a minute.
382
00:24:14.779 --> 00:24:15.549
They were three to two.
383
00:24:15.960 --> 00:24:18.269
So the fact that the market share for the two
384
00:24:18.349 --> 00:24:21.950
smaller companies, Baby Food's another example of, you know, courts
385
00:24:21.990 --> 00:24:24.730
don't like the argument that, well, the two smaller guys
386
00:24:24.789 --> 00:24:26.900
in the three to two market have to merge to
387
00:24:26.940 --> 00:24:29.920
compete more aggressively against the big guy. That's just not
388
00:24:29.940 --> 00:24:33.579
going to play almost no matter what the market shares are.
389
00:24:33.640 --> 00:24:37.069
So those are the rare examples where deals under 30
390
00:24:37.069 --> 00:24:38.170
percent get blocked.
391
00:24:38.650 --> 00:24:40.509
Deals over 30%.
392
00:24:40.509 --> 00:24:42.529
I mean, like I said, very few of them make
393
00:24:42.569 --> 00:24:44.970
it out of the government to begin with, but those
394
00:24:45.009 --> 00:24:48.490
that do, very mixed bag. And one of the reasons
395
00:24:48.549 --> 00:24:52.109
it's a very mixed bag is because they're so often
396
00:24:52.509 --> 00:24:57.109
the 30% depends on the government's market definition, which is
397
00:24:57.269 --> 00:24:59.009
often something that can be debated and.
398
00:24:59.589 --> 00:25:01.130
The government's.
399
00:25:01.970 --> 00:25:07.369
Method of calculating those market shares and often courts will
400
00:25:07.410 --> 00:25:10.849
say i don't buy the market definition i don't buy
401
00:25:10.890 --> 00:25:14.809
the calculation they don't even have to get to overcoming
402
00:25:14.849 --> 00:25:18.490
the presumption because they're just looking at the whole merger
403
00:25:18.529 --> 00:25:20.730
and the landscape and the market and the shares a
404
00:25:21.349 --> 00:25:29.099
completely different way um So how important is the Philadelphia
405
00:25:29.130 --> 00:25:30.200
National Bank presumption?
406
00:25:30.380 --> 00:25:31.559
It's still sort of important.
407
00:25:32.500 --> 00:25:34.900
You see that it's much harder to bring potential competition
408
00:25:34.940 --> 00:25:36.240
cases and vertical cases.
409
00:25:36.299 --> 00:25:38.240
You don't have that anchoring for the courts.
410
00:25:41.299 --> 00:25:42.819
The government, if they.
411
00:25:43.700 --> 00:25:47.359
convince the judge that the market definition is right and
412
00:25:47.420 --> 00:25:51.130
convince them that the shares are over 30%, they win
413
00:25:51.250 --> 00:25:53.650
the vast majority of those cases. It is harder to
414
00:25:53.730 --> 00:26:00.470
find the cases where the presumption is overcome. Now, is
415
00:26:00.509 --> 00:26:02.750
that because I really think that the judges are relying
416
00:26:02.789 --> 00:26:04.609
on the Philadelphia National Bank presumption?
417
00:26:04.990 --> 00:26:07.069
I'm not sure that I really think that.
418
00:26:07.710 --> 00:26:11.529
I think it's because When the government brings a case
419
00:26:11.569 --> 00:26:14.269
that's just slightly above 30 percent, if it's above 50
420
00:26:14.269 --> 00:26:16.450
or 60 percent, those are easy cases to bring. But
421
00:26:16.849 --> 00:26:20.390
if it's 30 or 40 percent, is it the presumption? No,
422
00:26:20.519 --> 00:26:23.779
it's all the other evidence that's going into it. When
423
00:26:23.819 --> 00:26:26.599
I first got to the FTC, Commissioner Wright sort of
424
00:26:26.640 --> 00:26:30.400
publicly said, you know, Debbie, you're the bureau director. you
425
00:26:30.440 --> 00:26:33.019
promised me that you won't cite Philadelphia National Bank. And
426
00:26:33.059 --> 00:26:36.500
I said, I will not promise that. It is the
427
00:26:36.559 --> 00:26:38.700
law of the land. The Supreme Court said it. But
428
00:26:38.809 --> 00:26:41.710
I will promise you this, that it'll be a paragraph
429
00:26:42.250 --> 00:26:44.470
and that the rest of the brief will be about
430
00:26:44.509 --> 00:26:45.809
the anti-competitive effects.
431
00:26:46.349 --> 00:26:46.950
Of the merger.
432
00:26:47.130 --> 00:26:51.730
It will be about the evidence of head-to-head closeness of competition,
433
00:26:52.029 --> 00:26:54.690
the evidence that there are entry barriers, the evidence that
434
00:26:54.730 --> 00:27:01.599
there are strong documents, and the evidence that customers are concerned.
435
00:27:01.680 --> 00:27:05.259
Concentration is really only a small part of this. In fact,
436
00:27:05.299 --> 00:27:07.660
the first time I was at the FTC, one of
437
00:27:07.779 --> 00:27:11.579
the senior managers said, look, To really build an antitrust
438
00:27:11.619 --> 00:27:14.119
merger case, you need three legs to the stools, the
439
00:27:14.180 --> 00:27:14.940
three H's.
440
00:27:15.420 --> 00:27:17.440
High Hervindals, high.
441
00:27:17.299 --> 00:27:19.220
Documents, hostile customers.
442
00:27:19.349 --> 00:27:22.569
And I think the latter two kind of help inform
443
00:27:22.589 --> 00:27:25.769
the market definition in the shares in a way that
444
00:27:26.089 --> 00:27:26.470
if you.
445
00:27:26.369 --> 00:27:28.690
Didn't have the Philadelphia National.
446
00:27:28.430 --> 00:27:30.549
Bank presumption, do I still think that the government could
447
00:27:30.569 --> 00:27:32.089
win the cases that it has won?
448
00:27:32.529 --> 00:27:32.869
Yes.
449
00:27:33.390 --> 00:27:36.190
It's just kind of a comfortable framework, just like I
450
00:27:36.230 --> 00:27:41.329
think market definition is a comfortable framework as a starting
451
00:27:41.390 --> 00:27:44.089
point to kind of get people's heads in the right places,
452
00:27:44.609 --> 00:27:47.809
even though you can often show that a market is
453
00:27:47.970 --> 00:27:51.730
problematic without relying on market definition as such. And in fact,
454
00:27:51.750 --> 00:27:54.690
the agency often does that. They decide there's a problem.
455
00:27:54.710 --> 00:27:56.390
Then it's like, OK, how should we talk about the
456
00:27:56.450 --> 00:27:58.319
market definition? Because we have to do that.
457
00:27:58.220 --> 00:28:03.079
For a court. Thanks, Debbie. That's very interesting. I wanted
458
00:28:03.119 --> 00:28:05.380
to turn to Rahul as.
459
00:28:05.299 --> 00:28:08.880
Another one of our private practitioners to comment both on
460
00:28:09.000 --> 00:28:10.720
what Debbie just said and also if you want to
461
00:28:10.759 --> 00:28:14.099
pull in your state AG experience as well. And I
462
00:28:14.319 --> 00:28:16.849
think the 2023 merger guidelines have been getting a little
463
00:28:16.869 --> 00:28:17.619
bit of a discussion.
464
00:28:17.640 --> 00:28:21.869
The various panelists, I suspect you have a view on
465
00:28:21.890 --> 00:28:24.890
that as well, given the tenure of your service.
466
00:28:25.309 --> 00:28:28.849
Yeah, well, my thanks for answering the question and thank
467
00:28:28.890 --> 00:28:32.450
you for... have me on this panel. A lot of
468
00:28:32.470 --> 00:28:34.990
what I'm going to say is that you're going to hear,
469
00:28:35.490 --> 00:28:37.230
we've already heard versions of it from the other speakers
470
00:28:37.279 --> 00:28:40.339
we know. I want to just start with this like
471
00:28:40.380 --> 00:28:44.579
level setting where the conversation that was heard earlier between
472
00:28:44.599 --> 00:28:50.269
Danny or Jeff Gainsbury I think that's a really important debate.
473
00:28:51.210 --> 00:28:55.349
It's a really important debate that captures really important policy
474
00:28:55.390 --> 00:28:58.170
questions of what are we trying to achieve with our
475
00:28:58.230 --> 00:28:59.230
merger control regimes?
476
00:28:59.630 --> 00:29:00.549
How do we achieve that?
477
00:29:01.950 --> 00:29:05.059
Are we fully focused on crisis and output? Do we
478
00:29:05.099 --> 00:29:09.299
care more about holistic and amorphous concerns about concentrated power,
479
00:29:09.339 --> 00:29:12.480
which could be political or social effects, or is there
480
00:29:12.519 --> 00:29:16.900
something else? And I think those are valid questions for
481
00:29:16.980 --> 00:29:21.539
policymakers to grapple with. And they're questions where people are
482
00:29:21.559 --> 00:29:23.579
going to fall on different points on the spectrum.
483
00:29:25.140 --> 00:29:28.829
But as a former.
484
00:29:27.769 --> 00:29:31.910
Federal and state enforcer, I do want to also ground
485
00:29:31.930 --> 00:29:37.069
this conversation from an enforcement vantage. And I think I
486
00:29:37.069 --> 00:29:39.599
want to start with you know, recognizing that there is
487
00:29:39.640 --> 00:29:43.720
a difference between practicing policy and enforcing the law. So
488
00:29:43.779 --> 00:29:45.460
when I was at the Bureau of Competitive Defense, when
489
00:29:45.480 --> 00:29:48.140
I was in the state AP's office, we were very
490
00:29:48.200 --> 00:29:52.440
intentional about not calling ourselves regulators. You know, we prefer
491
00:29:52.460 --> 00:29:55.910
to firm enforcers. We enforce the laws as it exists.
492
00:29:56.700 --> 00:29:59.319
We don't regulate markets according to what we prefer them
493
00:29:59.359 --> 00:30:02.609
to look like. That was a job for others. And
494
00:30:02.630 --> 00:30:05.490
so at the end of the day, Philadelphia National Bank,
495
00:30:05.609 --> 00:30:07.869
as old as it may be, as reliant as it
496
00:30:07.910 --> 00:30:12.170
may be on outdated economic thinking, it's still a good law.
497
00:30:12.960 --> 00:30:17.259
And as you foretold, It's still a controlling law. Neither
498
00:30:17.299 --> 00:30:20.539
the Supreme Court nor Congress has overturned it. So even
499
00:30:20.579 --> 00:30:25.630
if you think share-based or concentration-based metrics are an imprecise
500
00:30:26.109 --> 00:30:31.710
or even an outdated or incorrect economic proxy for predicting
501
00:30:31.730 --> 00:30:37.190
merger harm, that critique doesn't entirely capture the doctrine's legal relevance.
502
00:30:38.490 --> 00:30:41.109
So, until Congress.
503
00:30:40.900 --> 00:30:43.819
Or the Supreme Court will return Philadelphia NFL Bands, I
504
00:30:43.839 --> 00:30:48.420
think you're going to continue to see enforcers apply.
505
00:30:48.500 --> 00:30:51.740
That said, as a practical matter, this is, you.
506
00:30:51.700 --> 00:30:54.960
Know, I'm going to build on something Debbie had just mentioned,
507
00:30:55.000 --> 00:30:58.220
and I really doubt that you will see any federal
508
00:30:58.240 --> 00:31:04.930
agencies reputates solely on frothing affair or an A.J.-backed battle.
509
00:31:05.029 --> 00:31:07.829
They're not going to do it without some allegation of
510
00:31:08.269 --> 00:31:12.930
concrete armed attack. And that's not about doctrine or policy.
511
00:31:13.670 --> 00:31:17.170
That's about, at least for me, that's about litigation strategy. Litigation,
512
00:31:17.210 --> 00:31:20.119
at the end of the day, is a storytelling exercise.
513
00:31:20.900 --> 00:31:24.779
And a compelling story needs something tangible for the reader
514
00:31:24.799 --> 00:31:28.579
to connect with. And in this context, That's something like
515
00:31:28.599 --> 00:31:32.619
price effects. It's something like wage effects or reduced output. Judges,
516
00:31:32.660 --> 00:31:37.269
especially lay judges, they understand that. HHIs, they're kind of
517
00:31:37.289 --> 00:31:42.849
harder to internalize. So while the presumption does real legal work, right?
518
00:31:42.890 --> 00:31:46.049
It shifts the initial burden. I think it's not how
519
00:31:46.130 --> 00:31:50.029
the agencies factfully say it is for winning merger statements.
520
00:31:50.910 --> 00:31:53.799
I was looking, I didn't read Philadelphia National Bank this morning,
521
00:31:54.759 --> 00:31:57.660
but the description of today's panel did mention the Paramount
522
00:31:57.779 --> 00:32:01.759
Warner Brothers complaint. And I looked at that complaint and
523
00:32:01.819 --> 00:32:05.319
I think that's actually a pretty clean illustration of of
524
00:32:05.400 --> 00:32:09.740
this narrative focus, right? The states open with the structural presumption,
525
00:32:10.220 --> 00:32:14.750
but then they plead separate unilateral effects. They also plead
526
00:32:15.650 --> 00:32:19.750
coordinated effects. And so that's the storytelling architecture at work.
527
00:32:20.490 --> 00:32:23.579
It's giving the trier of fact room to not have
528
00:32:23.599 --> 00:32:27.400
to rely solely on HHI affairs. And now, as I
529
00:32:27.500 --> 00:32:30.019
also want to also build on something Danny had mentioned,
530
00:32:31.420 --> 00:32:35.650
relying on these structural presumptions, even if they may be outdated,
531
00:32:36.170 --> 00:32:39.430
I'm not entirely convinced that's a bad thing. You know,
532
00:32:39.670 --> 00:32:43.349
more than anything, and this is from a private practice perspective,
533
00:32:44.349 --> 00:32:48.630
businesses and markets need predictability for as much of it
534
00:32:48.970 --> 00:32:52.029
as these things need, like reasonably achieved. And so a
535
00:32:52.049 --> 00:32:56.450
bright line structural presumption gives merging parties a clear.
536
00:32:56.250 --> 00:32:57.490
Threshold to plan around.
537
00:32:57.589 --> 00:32:59.009
I mean, that's valuable.
538
00:32:59.690 --> 00:33:04.359
Even if it is not a perfect predictor of competitive harm.
539
00:33:04.730 --> 00:33:09.660
So at the end of the day, some benchmark to
540
00:33:09.700 --> 00:33:11.220
guide businesses is better than none.
541
00:33:13.559 --> 00:33:17.130
Yeah, can I just one minute disagree on that?
542
00:33:17.589 --> 00:33:20.470
Given how few cases are brought at 30% and given
543
00:33:20.509 --> 00:33:23.150
how few cases are brought with the numbers of the
544
00:33:23.190 --> 00:33:26.200
new guidelines, I actually think it was a step backward
545
00:33:26.539 --> 00:33:28.819
in suggesting that a whole bunch of mergers are problems
546
00:33:29.180 --> 00:33:32.869
that none of the administrations have brought cases under. I
547
00:33:33.410 --> 00:33:37.190
actually think that the Philadelphia National Bank presumption is what
548
00:33:37.210 --> 00:33:42.319
it is, but the guidelines need to give better guidance
549
00:33:42.400 --> 00:33:43.710
to businesses than they do now.
550
00:33:46.650 --> 00:33:49.089
So one of the questions in the chat actually kind
551
00:33:49.130 --> 00:33:51.670
of gets at this question about the 30% specifically, and
552
00:33:51.690 --> 00:33:54.849
I thought I would pose it maybe to Professor Hovindkamp
553
00:33:54.890 --> 00:33:57.769
first and then for others, partly because I think it's
554
00:33:57.789 --> 00:34:00.069
a good question, partly because I like to encourage questions
555
00:34:00.250 --> 00:34:04.009
in the chat. And so the question from the audience is,
556
00:34:04.109 --> 00:34:06.410
if 30% isn't the right number, and I think, Debbie,
557
00:34:06.450 --> 00:34:08.949
you were just expressing that 30% is way too low,
558
00:34:09.010 --> 00:34:10.760
and we heard that from others on the panel as well.
559
00:34:11.639 --> 00:34:15.320
Is there a number that is appropriate as either a
560
00:34:15.360 --> 00:34:21.960
screening mechanism or the right approach and that a guideline or,
561
00:34:22.219 --> 00:34:24.500
I suppose, a court should adopt?
562
00:34:26.780 --> 00:34:34.340
Well, if you take this recent research seriously, any number
563
00:34:34.380 --> 00:34:38.659
that expresses the market share of the post-merger firm is
564
00:34:38.739 --> 00:34:44.849
the wrong number. That's not what we should be looking at. Remember,
565
00:34:44.869 --> 00:34:51.329
the statute requires a change. It says a merger must
566
00:34:51.510 --> 00:34:57.619
lessen competition. And one of the points that Naki and
567
00:34:57.639 --> 00:35:01.699
Winston make is what we're trying to measure is what
568
00:35:01.860 --> 00:35:03.920
changes as a result of the merger.
569
00:35:04.420 --> 00:35:07.559
And for that, the important number is.
570
00:35:07.519 --> 00:35:11.380
The increase in concentration of both of these studies find
571
00:35:11.480 --> 00:35:17.179
no correlation between the absolute amount of core of concentration
572
00:35:18.719 --> 00:35:25.940
and price increases. The correlation is between the amount of
573
00:35:26.000 --> 00:35:29.860
the increase. I think that's the presumption. I'd be fine
574
00:35:30.280 --> 00:35:35.489
dropping the 30% rule from Philadelphia Bank, but focusing more
575
00:35:35.550 --> 00:35:38.769
closely on the amount of the increase. Now, there are
576
00:35:38.829 --> 00:35:43.400
other concerns, for example, the Sherman Act, which condemns mergers
577
00:35:43.460 --> 00:35:46.300
to monopoly. So you certainly might want to have a
578
00:35:46.360 --> 00:35:51.260
rule that says that a merger that creates a 60%
579
00:35:51.260 --> 00:35:56.420
firm might be unlawful on the basis of absolute concentration.
580
00:35:57.880 --> 00:36:03.510
But in the enforcement range, where most merger enforcement goes on,
581
00:36:03.590 --> 00:36:06.929
both of these studies, by the way, looked at mergers
582
00:36:06.969 --> 00:36:12.280
that survived scrutiny. They're in concentrated markets, but they're all
583
00:36:12.340 --> 00:36:17.630
mergers that actually occurred, so they weren't successfully challenged. And
584
00:36:17.650 --> 00:36:21.170
so looking back at those, both studies find we tend
585
00:36:21.190 --> 00:36:26.809
to under-enforce in that range, in that range where most
586
00:36:26.849 --> 00:36:32.000
mergers are challenged, and that the biggest price changes occur
587
00:36:33.409 --> 00:36:37.050
as the amount of concentration goes up. So I think
588
00:36:37.090 --> 00:36:40.949
we need to stop thinking about the 30% rule and
589
00:36:40.989 --> 00:36:42.230
think about the amount.
590
00:36:41.949 --> 00:36:46.800
Of the increase. Michael, if I may?
591
00:36:47.530 --> 00:36:48.139
Yes, of course.
592
00:36:49.099 --> 00:36:53.340
I'm glad to hear Herb saying that, to forget about
593
00:36:53.380 --> 00:36:55.920
the 30% rule and look at something that might be
594
00:36:55.960 --> 00:37:02.630
more productive. I just think at this early stage, I
595
00:37:02.650 --> 00:37:06.719
wouldn't put much stock in the papers that we have.
596
00:37:06.780 --> 00:37:11.579
The Mackey and Winston paper in particular, it's interesting, but
597
00:37:12.239 --> 00:37:17.420
it's a theoretical exercise in merger simulation models. It looks
598
00:37:17.480 --> 00:37:20.460
at possible mergers between different brands of beer in a
599
00:37:20.559 --> 00:37:25.559
highly regulated environment, one in which the vertical integration in
600
00:37:25.599 --> 00:37:29.190
most of the states is prohibited, and it's a peculiar
601
00:37:29.409 --> 00:37:34.670
market in itself for that reason. Still, fine, they've got
602
00:37:34.710 --> 00:37:37.969
something that's suggestive and I hope there'll be more research
603
00:37:38.630 --> 00:37:42.250
following it up. But just bear in mind about as
604
00:37:42.289 --> 00:37:46.489
long as we do have the 30% presumption, judges take
605
00:37:46.670 --> 00:37:50.989
presumptions like the rest of the law very seriously. And
606
00:37:51.030 --> 00:37:53.219
the agency can come in and of course it's gonna
607
00:37:53.260 --> 00:37:56.460
say more than 30%. It's gonna talk about its market
608
00:37:56.480 --> 00:37:59.460
definition at the very least and maybe something more.
609
00:38:00.159 --> 00:38:02.360
But nonetheless, the mindset.
610
00:38:02.159 --> 00:38:06.900
Is that, for its judge, that this is presumptively unlawful.
611
00:38:07.909 --> 00:38:10.510
And the defendant is going to have to find its
612
00:38:10.550 --> 00:38:13.650
way out of this to convince us that this is
613
00:38:14.230 --> 00:38:19.429
not actually harmful. And we're used to dealing with presumptions.
614
00:38:20.280 --> 00:38:23.800
Usually the presumptions are ones that are well tested, tried
615
00:38:23.860 --> 00:38:27.409
and true. The presumption of innocence in a criminal matter,
616
00:38:27.449 --> 00:38:31.269
the presumption of regularity in administrative affairs, things of that
617
00:38:31.329 --> 00:38:35.869
sort that shift the burden or define the burden of production.
618
00:38:37.090 --> 00:38:40.530
And so it's not a small matter. that this presumption
619
00:38:40.619 --> 00:38:45.510
still exists and is invoked regularly by the agencies. So
620
00:38:45.530 --> 00:38:53.590
I mentioned 116 times since it was decided. It's actually
621
00:38:53.610 --> 00:38:56.820
been cited in 167 of the government's briefs, but actually
622
00:38:56.860 --> 00:39:02.719
relied upon in 122 of them. And it's very effective.
623
00:39:03.360 --> 00:39:08.360
It does what Debbie talked about, creating a mindset is key,
624
00:39:08.619 --> 00:39:11.340
and this does so in a way that's just unfounded.
625
00:39:16.630 --> 00:39:17.449
Thanks, Judge Ginsburg.
626
00:39:18.110 --> 00:39:20.289
I wanted to shift gears a little bit and maybe
627
00:39:20.349 --> 00:39:22.889
pull on something, a little bit of what Judge Ginsburg said,
628
00:39:22.929 --> 00:39:26.000
and then I think also what Professor Sokol said about
629
00:39:26.619 --> 00:39:29.739
the Supreme Court, courts in general, and agencies. So it
630
00:39:29.780 --> 00:39:33.409
seems that we've been talking to some extent more explicitly
631
00:39:33.469 --> 00:39:36.610
on what the right approach should be. as opposed to
632
00:39:36.650 --> 00:39:40.050
perhaps a screening mechanism or a rule of decision, although
633
00:39:40.070 --> 00:39:42.989
Judge Ginsburg's comments just now brought out the idea of
634
00:39:43.030 --> 00:39:47.250
a presumption versus sort of a substantive rule of decision.
635
00:39:47.289 --> 00:39:49.449
I thought perhaps we'd start with Professor Sokol. I know
636
00:39:49.469 --> 00:39:51.750
you mentioned you think the Supreme Court might come out
637
00:39:51.789 --> 00:39:54.030
the other way. I guess there's another question as to
638
00:39:54.070 --> 00:39:56.119
whether this type of issue would ever get back up
639
00:39:56.159 --> 00:40:00.579
to the Supreme Court, given how merger litigation works. But
640
00:40:00.599 --> 00:40:03.519
I'm curious if you have a view regarding the issue
641
00:40:03.719 --> 00:40:07.480
of something like Philadelphia National Bank as a screening mechanism,
642
00:40:07.559 --> 00:40:10.769
as something that agencies or courts use, as opposed to
643
00:40:10.929 --> 00:40:14.150
the actual ultimate question.
644
00:40:15.070 --> 00:40:20.730
So let me handle both those. On how it would
645
00:40:20.769 --> 00:40:21.119
come out, I.
646
00:40:25.349 --> 00:40:29.840
I have 99% certainty it would not come out the
647
00:40:29.880 --> 00:40:32.980
same way today. And I think most commentators would agree
648
00:40:33.000 --> 00:40:33.739
with me on that.
649
00:40:36.900 --> 00:40:39.829
There are two issues that you raised. One is, one
650
00:40:39.849 --> 00:40:42.309
sort of gets to sort of the screening mechanism. I agree.
651
00:40:42.389 --> 00:40:45.610
That's essentially what it's there for. I also agree with Debbie,
652
00:40:45.929 --> 00:40:46.630
which is that.
653
00:40:48.480 --> 00:40:51.289
Some judges, they kind of know where they want to
654
00:40:51.349 --> 00:40:56.409
come out and they're going to find whichever mechanism necessary.
655
00:40:55.869 --> 00:40:58.250
To get to the outcome that they want.
656
00:40:58.969 --> 00:41:03.010
But overall, would it be useful if it was a
657
00:41:03.050 --> 00:41:06.440
more effective screening mechanism? Yes, but frankly, that's what the
658
00:41:06.519 --> 00:41:09.440
Merchant Guidelines should be there for, to be giving that
659
00:41:09.519 --> 00:41:10.519
kind of certainty.
660
00:41:10.539 --> 00:41:23.199
And I think the guidelines are.
661
00:41:25.349 --> 00:41:29.630
Difficult to provide that level of certainty. We could have
662
00:41:29.650 --> 00:41:33.170
a whole other session on what it did well and
663
00:41:33.250 --> 00:41:36.280
the many things that it did less well.
664
00:41:37.519 --> 00:41:38.800
I think, though, that.
665
00:41:40.369 --> 00:41:43.190
We wouldn't have had merger guidelines if we didn't have
666
00:41:43.230 --> 00:41:46.449
a case like a Philadelphia National Bank. But I think
667
00:41:46.489 --> 00:41:49.159
you raise another question, which is a really important one,
668
00:41:49.179 --> 00:41:51.599
which is why don't we see merger litigation that goes
669
00:41:51.659 --> 00:41:59.300
all the way to the Supreme Court? It's 50 years
670
00:41:59.360 --> 00:42:03.730
since we've seen substantive Supreme Court antitrust merger cases.
671
00:42:06.170 --> 00:42:08.650
It might be time, but the incentives.
672
00:42:08.230 --> 00:42:12.909
Are often not in favor of firms to push aggressively
673
00:42:12.969 --> 00:42:15.739
to get in front of the court to show that
674
00:42:15.780 --> 00:42:18.900
the court has, that it should be of interest to
675
00:42:18.920 --> 00:42:19.300
the court.
676
00:42:19.820 --> 00:42:23.699
But I think we are in desperate need of that.
677
00:42:24.800 --> 00:42:26.260
And I'll just leave it at that.
678
00:42:26.280 --> 00:42:33.150
That ain't going to happen, Barry. But it's because of
679
00:42:33.170 --> 00:42:37.440
a very specific reason. The Antitrust Expediting Act was essentially
680
00:42:37.519 --> 00:42:42.110
repealed as a part of the Tunney Act. So it's
681
00:42:42.150 --> 00:42:45.210
just not practical for these deals to hang around waiting
682
00:42:45.250 --> 00:42:49.269
for the Supreme Court. In the Montford of Colorado case,
683
00:42:49.309 --> 00:42:52.449
when I was assistant attorney general, I called the lawyer
684
00:42:52.489 --> 00:42:56.829
for the company, if I knew, and said, why are
685
00:42:56.869 --> 00:42:59.409
we in the Supreme Court on a $ 15 million deal?
686
00:43:00.110 --> 00:43:02.510
And the answer was, it was a fluke. Local counsel
687
00:43:02.530 --> 00:43:06.389
forgot to provide we get our money back. So, you know,
688
00:43:06.409 --> 00:43:08.130
that's the only kind of thing that's ever going to
689
00:43:08.150 --> 00:43:10.630
get it to the Supreme Court, and just can't count
690
00:43:10.670 --> 00:43:11.530
on that ever happening.
691
00:43:12.480 --> 00:43:15.880
Oh, they would come if the government lost and then
692
00:43:15.900 --> 00:43:19.239
won on appeal because then the parties do have an incentive.
693
00:43:19.280 --> 00:43:23.889
They've already closed their deal and they can continue. So
694
00:43:23.969 --> 00:43:27.250
that's the kind of case where you could see it
695
00:43:28.849 --> 00:43:31.050
going to the Supreme Court. I can think of a
696
00:43:31.070 --> 00:43:34.510
couple that would have gone to the Supreme Court if....
697
00:43:36.809 --> 00:43:38.909
You know, the government.
698
00:43:38.630 --> 00:43:41.489
Had decided to keep pursuing it after losing a.
699
00:43:41.610 --> 00:43:43.789
Circuit court opinion. That's the way it would get to
700
00:43:43.809 --> 00:43:44.719
the Supreme Court.
701
00:43:44.889 --> 00:43:47.570
If the parties won the first time and then the
702
00:43:47.639 --> 00:43:52.199
government won the neck at the appellate level and decided to,
703
00:43:53.579 --> 00:43:55.840
one of them decided to keep, and then the parties
704
00:43:55.860 --> 00:43:58.360
decided to fight that.
705
00:44:00.519 --> 00:44:02.340
But there just haven't been very many like that.
706
00:44:04.369 --> 00:44:06.920
Yes, I guess we'll see. It's going to be interesting
707
00:44:06.960 --> 00:44:07.219
to see.
708
00:44:07.260 --> 00:44:09.059
We have a lot of litigation going on right now,
709
00:44:10.059 --> 00:44:12.639
some of which is the folks you're involved with.
710
00:44:12.659 --> 00:44:13.780
So I guess we'll see if some of.
711
00:44:13.760 --> 00:44:18.449
Those defy the trend and are not another example like
712
00:44:18.489 --> 00:44:21.230
Judge Ginsburg just gave.
713
00:44:21.250 --> 00:44:22.250
There's one other.
714
00:44:24.530 --> 00:44:26.769
Concept that I think was pulled out.
715
00:44:26.789 --> 00:44:31.679
I'm forgetting actually who mentioned it, but there's the role
716
00:44:31.699 --> 00:44:34.179
of the Supreme Court, there's the role of the agencies. Obviously,
717
00:44:34.219 --> 00:44:38.150
there's another player in our trust law that occasionally comes up,
718
00:44:38.210 --> 00:44:40.610
and Judge Ginsburg just mentioned it about the Tunney Act.
719
00:44:41.309 --> 00:44:45.309
And that's Congress. And I'm curious if there's anyone on
720
00:44:45.349 --> 00:44:48.010
the panel who thinks that this type of issue is
721
00:44:48.449 --> 00:44:53.059
up and ripe for a congressional resolution, whether that would
722
00:44:53.079 --> 00:44:54.679
be a good intervention, a bad.
723
00:44:54.519 --> 00:44:57.719
Intervention on that particular issue.
724
00:44:57.739 --> 00:45:01.340
Maybe we'll start with Rahul and go from there.
725
00:45:03.030 --> 00:45:06.869
I mean, it would be at least an appropriate forum
726
00:45:07.130 --> 00:45:11.409
for Scott Simpson. I'll just be frank in my criticism
727
00:45:11.670 --> 00:45:15.179
that I think Congress is not the most functional at
728
00:45:15.199 --> 00:45:18.780
the moment. So I wouldn't bet a lot of money
729
00:45:19.300 --> 00:45:25.280
on there being a resolution or a good one out
730
00:45:25.300 --> 00:45:28.679
of the process today. But that's not to say that
731
00:45:29.679 --> 00:45:32.980
Congress conceptually doesn't have an important role to play here.
732
00:45:37.079 --> 00:45:40.360
Let's open that up if any others have thoughts.
733
00:45:39.590 --> 00:45:39.929
On the political.
734
00:45:39.969 --> 00:45:42.250
Yeah, I have a thought or two.
735
00:45:43.130 --> 00:45:45.289
I don't think this is a good issue for Congress
736
00:45:45.329 --> 00:45:48.769
to take on right now. It would be too political.
737
00:45:48.829 --> 00:45:52.880
There would be too many interest groups involved. What we've
738
00:45:52.940 --> 00:45:56.769
seen at the state level... has been more in the
739
00:45:56.829 --> 00:46:01.840
direction of using antitrust law to protect competitors.
740
00:46:01.199 --> 00:46:02.500
Rather than consumers.
741
00:46:02.519 --> 00:46:06.760
Uh, I just think there's so many sideways directions that
742
00:46:06.800 --> 00:46:09.780
could go that it would be, it would be risky.
743
00:46:10.219 --> 00:46:12.929
Uh, the real problem is of course, how.
744
00:46:12.789 --> 00:46:17.030
To interpret the may substantially lessen competition standard.
745
00:46:18.090 --> 00:46:21.460
Uh, and, uh, I think any.
746
00:46:21.519 --> 00:46:24.760
Guidance Congress could give on that would probably end up
747
00:46:24.800 --> 00:46:25.989
doing more harm than good.
748
00:46:30.409 --> 00:46:32.570
I've noticed in the attendee list that we have quite
749
00:46:32.610 --> 00:46:35.889
a few private practitioners in the list, as well as
750
00:46:35.929 --> 00:46:40.000
different members of the Bar Academy, et cetera. One question
751
00:46:40.019 --> 00:46:42.059
I wanted to pose, and maybe I'll start with Judge
752
00:46:42.099 --> 00:46:47.360
Ginsburg and then move over to Debbie Feinstein. We've talked
753
00:46:47.380 --> 00:46:51.099
a little bit about how these– How the presumption works
754
00:46:51.139 --> 00:46:52.170
in practice, whether it's in.
755
00:46:52.139 --> 00:46:55.650
The courts, whether it's in private practice, whether it's in
756
00:46:55.670 --> 00:46:58.690
the agencies, what is the best way.
757
00:46:59.190 --> 00:47:03.699
To advise and recognize that you're not doing the advising,
758
00:47:03.719 --> 00:47:06.300
but you're the person that maybe the case comes would
759
00:47:06.320 --> 00:47:10.059
have come before that. private parties given the current state
760
00:47:10.219 --> 00:47:15.139
of the law and scholarship and agency practice. When thinking
761
00:47:15.199 --> 00:47:19.510
about the presumption, I know you mentioned, Judge Ginsburg, there's,
762
00:47:20.150 --> 00:47:25.050
I thought you said, 166 citations in the agency pleadings.
763
00:47:26.469 --> 00:47:28.010
To the presumption.
764
00:47:28.630 --> 00:47:30.840
I'm curious how you think that that should affect how
765
00:47:30.860 --> 00:47:33.579
emerging parties should think about their deals.
766
00:47:37.840 --> 00:47:40.179
It's been a long time since I've been advising clients.
767
00:47:43.010 --> 00:47:47.849
I think any lawyer has got to be frankly skittish
768
00:47:47.909 --> 00:47:51.789
about any deal that crosses that 30% threshold for.
769
00:47:51.670 --> 00:47:52.510
The merged company.
770
00:47:52.849 --> 00:47:55.139
Now, maybe not based on what Debbie was saying. Maybe
771
00:47:56.000 --> 00:47:59.619
only a few of those get challenged. I'd like to
772
00:47:59.659 --> 00:48:01.159
know more about that.
773
00:48:00.980 --> 00:48:03.599
If I were advising. Yeah.
774
00:48:03.639 --> 00:48:05.679
Why don't we turn to Debbie to spell out a
775
00:48:05.699 --> 00:48:07.659
little bit more about what you had said before about
776
00:48:08.710 --> 00:48:13.309
your view on how the boardroom should think about the
777
00:48:13.309 --> 00:48:15.480
30% number or the option generally.
778
00:48:15.739 --> 00:48:17.679
I think there are so many other factors.
779
00:48:17.719 --> 00:48:20.300
Factors they should think about more it's not the first
780
00:48:20.380 --> 00:48:22.400
i mean i might ask about share sure because of
781
00:48:22.460 --> 00:48:25.639
course you would but my first question is how are
782
00:48:25.699 --> 00:48:27.730
customers going to react to this how do you run
783
00:48:27.769 --> 00:48:29.469
into each other in the marketplace.
784
00:48:29.070 --> 00:48:31.250
How do you compete do you compete for.
785
00:48:31.150 --> 00:48:34.630
The same customers there's a whole bunch of questions that
786
00:48:34.690 --> 00:48:38.590
i ask before i ask what are the market shares
787
00:48:38.869 --> 00:48:41.590
the ftc used to put out this document which i
788
00:48:41.610 --> 00:48:43.619
thought was so useful and i wish they would do
789
00:48:43.659 --> 00:48:49.000
it again um which was an economist went through all
790
00:48:49.059 --> 00:48:53.389
the memos that were sent out to the commission about mergers,
791
00:48:53.849 --> 00:48:58.650
and he categorized them. How many of them were two-to-ones, three-to-twos, four-to-threes,
792
00:48:58.710 --> 00:49:01.409
you know, in terms of substantial number of competitors. He
793
00:49:01.449 --> 00:49:04.050
would look at the shares, but that was kind of
794
00:49:04.900 --> 00:49:07.320
just one of many factors. Were there a lot of
795
00:49:07.380 --> 00:49:10.920
hot documents? Were there a lot of customer complaints?
796
00:49:10.960 --> 00:49:12.539
A whole bunch of other things.
797
00:49:12.960 --> 00:49:15.550
And what I found that I remember in terms of
798
00:49:15.610 --> 00:49:18.030
what I tell clients and I do today because I
799
00:49:18.090 --> 00:49:19.909
remember that is I talk to clients.
800
00:49:20.219 --> 00:49:23.119
About the number of meaningful competitors much more often than
801
00:49:23.139 --> 00:49:23.639
I talk.
802
00:49:23.480 --> 00:49:24.119
About the shares.
803
00:49:24.539 --> 00:49:26.500
If it's a two to one, I got one of
804
00:49:26.519 --> 00:49:29.679
those through once. It was a very unique circumstance. It
805
00:49:29.699 --> 00:49:33.059
was a potential competition case. Two to ones generally don't
806
00:49:33.079 --> 00:49:36.440
get through. Three to two generally don't get through. Four
807
00:49:36.480 --> 00:49:38.380
to three is the evidence show that it was 50-50
808
00:49:38.800 --> 00:49:39.860
and then you go on from there.
809
00:49:40.280 --> 00:49:42.260
That was much more helpful to me.
810
00:49:42.159 --> 00:49:44.619
Than the shares because shares alone don't tell you anything,
811
00:49:44.719 --> 00:49:46.500
as I think Professor Hovenkamp said.
812
00:49:46.519 --> 00:49:49.050
It's like, how much is it increasing, right?
813
00:49:49.050 --> 00:49:53.389
28 plus two is very different than 15 and 15
814
00:49:53.389 --> 00:49:56.170
when the other competitors are 60 and.
815
00:49:56.110 --> 00:49:57.389
Whatever the rest is.
816
00:49:58.909 --> 00:50:01.409
You know, there's all of those factors matter a lot
817
00:50:01.469 --> 00:50:03.489
more to me than what are the merging party shares.
818
00:50:07.500 --> 00:50:08.679
Thanks, Debbie.
819
00:50:09.199 --> 00:50:12.820
Professor Sokol, I saw that you were nodding vigorously during
820
00:50:12.860 --> 00:50:13.960
what Debbie was saying.
821
00:50:14.099 --> 00:50:15.940
I thought I'd give you an opportunity to weigh in
822
00:50:16.000 --> 00:50:16.219
on that.
823
00:50:16.239 --> 00:50:18.260
I know that you're teaching in the law school and
824
00:50:18.280 --> 00:50:19.969
you're also teaching in the business school.
825
00:50:20.050 --> 00:50:23.489
And so you're crafting young minds in an AI era.
826
00:50:24.150 --> 00:50:29.989
And I'm curious how you would propose this for thinking
827
00:50:30.030 --> 00:50:31.510
about for those contemplating transactions.
828
00:50:33.440 --> 00:50:36.199
So I think, first of all, Debbie's an amazing lawyer
829
00:50:37.300 --> 00:50:40.300
and I would always want her representing me.
830
00:50:41.179 --> 00:50:45.420
I think she's right.
831
00:50:45.960 --> 00:50:47.519
I think there are a lot of factors and you
832
00:50:47.539 --> 00:50:50.139
look across all the different risks and all the different
833
00:50:50.159 --> 00:50:56.280
things that matter. I think in practice, we see this
834
00:50:56.909 --> 00:51:02.420
already happening and it's been happening for decades, But I
835
00:51:02.480 --> 00:51:07.230
think that overall having certain cases helps anchor the debate
836
00:51:07.409 --> 00:51:12.090
on certain issues for that very, very difficult conversation where
837
00:51:12.110 --> 00:51:15.269
the parties and or the agencies say, I guess we're
838
00:51:15.289 --> 00:51:17.909
gonna sit in court. And the question is, are we
839
00:51:18.090 --> 00:51:20.900
anchoring on the right kind of.
840
00:51:22.840 --> 00:51:25.480
Presumptions? And here I think.
841
00:51:27.260 --> 00:51:32.329
We do need to fix some things, but overall, we
842
00:51:32.349 --> 00:51:34.989
do need to have that kind of anchor where we
843
00:51:35.010 --> 00:51:39.969
have a credible threat, but it needs to be a
844
00:51:40.210 --> 00:51:45.949
nuanced and well thought out set of discussions. It matters
845
00:51:46.010 --> 00:51:48.190
for credibility of the agencies, it matters in front of
846
00:51:48.210 --> 00:51:51.409
the courts, but frankly it also matters to the executives
847
00:51:51.469 --> 00:51:53.489
proposing these deals and negotiating them.
848
00:51:56.900 --> 00:51:57.239
Thank you.
849
00:51:57.260 --> 00:52:01.619
I think we have time for just Any last closing
850
00:52:01.679 --> 00:52:04.760
remarks or thoughts or reflections before we wrap up at
851
00:52:04.800 --> 00:52:09.090
the end of the hour? And so, of course, welcome
852
00:52:09.110 --> 00:52:11.309
to pass, but I thought I'd start with Professor Hobenkamp
853
00:52:11.329 --> 00:52:14.650
and then we'll go around virtually in my circle.
854
00:52:15.289 --> 00:52:18.690
Yeah, just one thing to follow up on something Debbie
855
00:52:18.710 --> 00:52:23.510
said a minute ago. One message of this recent scholarship
856
00:52:23.570 --> 00:52:26.070
is you can't throw all three to two mergers or
857
00:52:26.190 --> 00:52:28.190
all four to three mergers away.
858
00:52:29.030 --> 00:52:32.449
Into the same boat. If you were looking at the
859
00:52:32.510 --> 00:52:33.780
amount of the increase, it.
860
00:52:33.760 --> 00:52:36.380
Makes a whole lot of difference what the relative size
861
00:52:36.440 --> 00:52:39.099
of the merging firms are. There's a whole lot of
862
00:52:39.159 --> 00:52:45.889
difference between a merger of two 20% firms, which produces
863
00:52:45.949 --> 00:52:49.809
a huge increase in the HHI or a merger of
864
00:52:49.849 --> 00:52:54.610
a 39% firm and a 1% firm, which still give
865
00:52:54.650 --> 00:52:58.550
you the same market share, but only a very small increase.
866
00:52:58.630 --> 00:53:01.449
And those kinds of pairs of mergers need to be
867
00:53:01.469 --> 00:53:04.110
evaluated differently.
868
00:53:04.130 --> 00:53:12.570
I think that every three to two merger involves some
869
00:53:12.590 --> 00:53:18.889
kind of unique story. It's almost unimaginable. Otherwise, I had
870
00:53:19.070 --> 00:53:24.510
one when I was AAG. Three firms involved in very
871
00:53:24.550 --> 00:53:30.610
close rivals in desktop quotation systems. One of them, it
872
00:53:30.659 --> 00:53:33.699
was definitively shown, the first was acquiring the third, as
873
00:53:33.780 --> 00:53:38.179
I recall. In any event, the odd man out, not
874
00:53:38.219 --> 00:53:42.269
that it was being acquired, had not. invested in the
875
00:53:42.329 --> 00:53:45.150
technology that the other two had, it was going to
876
00:53:45.190 --> 00:53:47.360
be out of the business one way or another within
877
00:53:47.389 --> 00:53:50.800
a couple of years. And the acquiring firm basically wanted
878
00:53:51.760 --> 00:53:56.159
to be able to smoothly take a shot at their customers.
879
00:53:56.860 --> 00:53:58.679
But it was going to be two no matter what.
880
00:54:00.699 --> 00:54:03.780
That's an unusual story, to say the least.
881
00:54:04.500 --> 00:54:05.559
Yeah, yeah, yeah.
882
00:54:06.539 --> 00:54:10.360
Debbie, I think Professor Hovenkamp just said you're too cautious
883
00:54:10.659 --> 00:54:12.909
in some of your advice.
884
00:54:12.969 --> 00:54:15.289
So I'll let you respond on the three to two
885
00:54:15.329 --> 00:54:15.929
merger point.
886
00:54:16.269 --> 00:54:20.460
Look, I said that the government mostly does it.
887
00:54:21.050 --> 00:54:21.780
Mostly does it.
888
00:54:22.690 --> 00:54:25.239
Three to twos are mostly a problem and most three
889
00:54:25.260 --> 00:54:29.300
to twos don't. do get challenged, even if the share
890
00:54:29.360 --> 00:54:32.760
is below 30%. So I think whether or not they
891
00:54:32.860 --> 00:54:35.860
should be is a different question than whether I should
892
00:54:35.880 --> 00:54:39.820
be advising clients that if they're doing a three to
893
00:54:39.840 --> 00:54:42.849
two merger, they should be looking at a long, long battle.
894
00:54:42.869 --> 00:54:42.940
Yep.
895
00:54:45.010 --> 00:54:46.829
And Rahul, and then we'll close with Danny.
896
00:54:47.929 --> 00:54:51.300
Yeah, I'll just be really clear. I think there's a
897
00:54:51.360 --> 00:54:55.400
very important conversation about whether something makes sense. I think
898
00:54:55.440 --> 00:54:58.860
as a practical matter, I don't think they're going anywhere.
899
00:54:58.960 --> 00:55:02.489
It's a reality that we're going to have to deal
900
00:55:02.510 --> 00:55:06.030
with now. I don't think it's going to challenge something
901
00:55:06.050 --> 00:55:10.889
that's a 31% market share. But you're still going to
902
00:55:10.909 --> 00:55:13.110
use it as a rhetorical device, right? That this new
903
00:55:13.210 --> 00:55:17.110
earth is 50 times more, you know, whatever bigger than
904
00:55:17.550 --> 00:55:21.530
the UFO on HXI or something. So it's not going
905
00:55:21.550 --> 00:55:22.829
to work.
906
00:55:23.510 --> 00:55:23.630
Yeah.
907
00:55:23.650 --> 00:55:25.369
And last word out from the West Coast.
908
00:55:29.250 --> 00:55:40.349
Thank you. If I recall, Bork himself... set the limit at.
909
00:55:42.050 --> 00:55:46.329
Four to three as the important merger justification. Here, we're
910
00:55:46.389 --> 00:55:53.159
talking about three to two. I think that is a
911
00:55:53.179 --> 00:55:57.860
tough slog. But again, this is where competitive effects matter.
912
00:55:57.900 --> 00:55:58.699
This is where I have a.
913
00:55:58.699 --> 00:56:02.679
Lot of sympathy for Judge Ginsburg and his view of like,
914
00:56:03.099 --> 00:56:05.159
what do the economics show? The economics show that there
915
00:56:05.179 --> 00:56:06.570
are a lot of other things that matter too.
916
00:56:06.780 --> 00:56:15.980
And we should a first step, but God forbid it
917
00:56:16.000 --> 00:56:17.179
should be the final step.
918
00:56:17.360 --> 00:56:19.820
And I think that's my big takeaway.
919
00:56:20.550 --> 00:56:24.070
Merger control should not begin and end with.
920
00:56:25.889 --> 00:56:29.809
How do we define the market and do we hit
921
00:56:29.809 --> 00:56:34.000
30% or something like that. That's inconsistent with merger practice,
922
00:56:34.039 --> 00:56:38.800
with the economics, and frankly, Even the ICN thinks about
923
00:56:38.820 --> 00:56:42.699
these issues simply as a first step in terms of
924
00:56:42.820 --> 00:56:44.179
global best practices.
925
00:56:44.280 --> 00:56:46.880
And I think we just need to keep that in mind.
926
00:56:48.239 --> 00:56:50.679
Thank you. Thank you for all the panelists.
927
00:56:51.159 --> 00:56:53.860
This has been a great discussion that I was very
928
00:56:53.900 --> 00:56:57.280
honored to attempt to facilitate. I'm going to turn it
929
00:56:57.300 --> 00:57:00.539
back over to Emma for concluding the remark.
930
00:57:01.579 --> 00:57:04.300
Thanks, Michael. On behalf of the Federalist Society, I want
931
00:57:04.340 --> 00:57:06.989
to thank our experts for the benefit of their valuable
932
00:57:07.030 --> 00:57:09.510
time and expertise today. I also want to thank you,
933
00:57:09.570 --> 00:57:13.769
our audience, for joining and participating. We also welcome listener feedback,
934
00:57:13.829 --> 00:57:17.909
if you have any, by email at fedsocforums at fedsoc.org.
935
00:57:18.429 --> 00:57:20.889
As always, please keep an eye on our website and
936
00:57:20.949 --> 00:57:24.090
on your emails for any announcements about upcoming virtual events.
937
00:57:24.679 --> 00:57:26.800
Thank you all again for joining us today, and we
938
00:57:26.840 --> 00:57:27.219
are adjourned.
939
00:57:28.269 --> 00:57:32.030
Thanks for listening to FedSoc Forums, a Federalist Society podcast.
940
00:57:32.440 --> 00:57:34.960
Be sure to subscribe wherever you get your podcasts so
941
00:57:34.980 --> 00:57:39.190
you won't miss an episode. To learn more, visit our website, FedSoc.org.