Merlin Partners LP and AAMAF, LP v. AutoInfo, Inc. ( 2015 )

EFiled: Apr 30 2015 03:25PM EDT
Transaction ID 57163687
Case No. 8509-VCN
IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE
MERLIN PARTNERS LP, and :
AAMAF, LP, :
:
Petitioners, :
:
v. : C.A. No. 8509-VCN
:
AUTOINFO, INC., a Delaware :
corporation, :
Respondent. :
MEMORANDUM OPINION
Date Submitted: January 9, 2015
Date Decided: April 30, 2015
Ronald A. Brown, Jr., Esquire, Marcus E. Montejo, Esquire, Kevin H. Davenport,
Esquire, and Eric J. Juray, Esquire of Prickett, Jones & Elliott, P.A., Wilmington,
Delaware, Attorneys for Petitioners.
A. Thompson Bayliss, Esquire and David A. Seal, Esquire of Abrams & Bayliss
LLP, Wilmington, Delaware, Attorneys for Respondent.
NOBLE, Vice Chancellor.
Petitioners Merlin Partners LP and AAMAF, LP are former common
stockholders of Respondent AutoInfo, Inc. (“AutoInfo” or the “Company”).
Pursuant to 8 Del. C. § 262, they demanded appraisal of their shares in connection
with a merger (the “Merger”) whereby AutoInfo’s common stockholders were
cashed out at a price of $1.05 per share. This memorandum opinion sets forth the
Court’s post-trial findings of fact and conclusions of law.
I. BACKGROUND
A. AutoInfo’s Business
At the time of the Merger, AutoInfo was a public non-asset based
transportation services company operating through two wholly-owned
subsidiaries.1 It did not own any equipment and provided brokerage and contract
carrier services through a network of independent sales agents in the United States
and Canada. AutoInfo and its agents split fees generated by freight transportation
transactions.2 The agents developed and maintained all important client
relationships.3
The Company also provided support services to its agents. Its assistance
was primarily financial, such as making long-term loans and short-term advances.
1
This memorandum opinion does not distinguish between AutoInfo and its
subsidiaries; they are collectively referred to as AutoInfo.
2
Trial Tr. 145 (Puglisi).
3
Trial Tr. 34 (Patterson).
1
AutoInfo also supplied non-financial services, such as training, marketing
assistance, market segment data, and business analysis tools.4
The Company’s 100% agent-based model distinguished it from many others
in the transportation logistics industry that rely on a “company store” model.
While AutoInfo’s brokers were independent contractors, “[b]rokers [in a company
store model] are direct employees of the company.”5
B. AutoInfo’s Board and Management
AutoInfo’s management (the “Management”) consisted of Harry Wachtel
(“Wachtel”), the Chairman and Chief Executive Officer (“CEO”); Michael
Williams (“Williams”), the President, Chief Operating Officer, and General
Counsel; William I. Wunderlich (“Wunderlich”), an Executive Vice President and
the Chief Financial Officer (“CFO”); Mark Weiss (“Weiss”), an Executive Vice
President; and David Less, the Chief Information Officer and Vice President.
Throughout the sales process, and at the time of the Merger, AutoInfo’s
board (the “Board”) consisted of five directors. Two, Wachtel and Weiss, were
inside directors. The others, Peter Einselen, Thomas C. Robertson, and Mark K.
Patterson (“Patterson”), were outside directors. Wachtel served as the Board’s
chairman.6
4
JX 335 (“AutoInfo 2012 Form 10-K”) at 2.
5
JX 179 (“L.E.K. Consulting Due Diligence Presentation”) at 32.
6
AutoInfo 2012 Form 10-K at 28.
2
C. The Merger
1. AutoInfo Considers Strategic Alternatives
During a regularly scheduled meeting in the first quarter of 2011, the Board
discussed AutoInfo’s financial results, budget, business, and financial prospects. It
was concerned that the market undervalued AutoInfo relative to comparable agent-
based, non-asset based transportation services companies. Part of the problem was
that the Company was small, thinly traded on the Nasdaq Over-the-Counter
Bulletin Board, and did not receive much analyst coverage. The Board decided
that exploring strategic options, including a potential sale, was in the best interests
of AutoInfo’s stockholders.7
The Board was not the only AutoInfo constituent disappointed with the
Company’s stock price. Around this time, Patterson (a Board member) was
contacted by Kinderhook, LP (“Kinderhook”), a stockholder with which he had a
relationship.8 Kinderhook believed that AutoInfo’s stock price failed to reflect its
financial performance. Although it did not push for a sale of the Company, it
encouraged the Board to develop a strategy to increase the stagnant stock price,
which was then trading in the $0.50-0.60 per share range.9
7
JX 334 (“Apr. 1, 2013, AutoInfo Schedule 14A”) at 23.
8
Trial Tr. 7 (Patterson). Kinderhook controlled 6,278,312 AutoInfo shares,
representing approximately 18.3% of the Company’s outstanding common shares.
JX 336 (“Apr. 1, 2013, AutoInfo Form DEFM14A”) at 72.
9
Trial Tr. 12, 23-24 (Patterson).
3
2. AutoInfo Retains Stephens
In summer 2011, Patterson contacted Stephens Inc. (“Stephens”), an
investment bank with experience in the transportation industry, to explore
AutoInfo’s strategic options. Stephens prepared and presented on July 29, 2011, a
Strategic Initiatives Overview, outlining avenues for enhancing stockholder
value.10 While AutoInfo had “built a solid legacy within the transportation and
logistics industry,” it “consistently traded at valuation multiples well below its peer
group due to the Company’s relatively small scale and corresponding lack of
interest from the investment community.”11 Stephens believed that if the Company
could grow its market capitalization from $20 million to approximately $400-500
million, then it would gain greater Wall Street attention and access capital at a
lower cost.12 The investment bank concluded that AutoInfo might need to alter its
strategy to achieve the necessary growth.13
Stephens thus proposed strategic alternatives, including organic projects,
shareholder distributions, and acquisitions.14 It identified pros and cons for each
option. For example, it suggested that “[e]xecution risk,” related to Management’s
ability to execute, would be a concern should the Company decide to pursue an
10
JX 19 (“Stephens’s Strategic Initiatives Overview”).
11
Id. at 5.
12
Trial Tr. 276-77 (Miller); Stephens’s Strategic Initiatives Overview 12.
13
Stephens’s Strategic Initiatives Overview 5.
14
Id. at 14.
4
organic project.15 Stephens also preliminarily valued the Company within a range
of $0.59 to $1.76 per share.16 The average of its valuations was $0.98 per share,
above the Company’s then-current $0.60 price.17
In August 2011, after considering its various options, the Board began
reaching out to potential purchasers.18 Patterson contacted parties that were active
in mergers and acquisitions in the transportation industry. While there was some
interest, AutoInfo could not reach a satisfactory agreement.19
Several months later, in November 2011, activist hedge funds Baker Street
Capital L.P. and Khrom Capital Management, through affiliated entities (“Baker
Street”), acquired a 13% equity interest in AutoInfo.20 Baker Street began
expressing its desire that AutoInfo be sold. According to Patterson, those demands
did not impact the Board’s sales process, which was already underway.21
In early 2012, after interviewing several investment banks, AutoInfo
formally retained Stephens to run a sales process.22 The parties agreed to an
incentive-based fee structure whereby Stephens would be paid 2% on the first $54
15
Id. at 15; Trial Tr. 16 (Patterson).
16
Stephens’s Strategic Initiatives Overview 19.
17
Id.
18
Trial Tr. 17 (Patterson).
19
Trial Tr. 19 (Patterson).
20
JX 23 (Baker Street November 10, 2011, Schedule 13D); JX 86 (Baker Street
Apr. 20, 2012, Schedule 13D, Amendment No. 1).
21
Trial Tr. 20 (Patterson).
22
Trial Tr. 25 (Patterson).
5
million of a transaction price and 5% on any additional value.23 Stephens had
extensive industry experience; Michael Miller (“Miller”), who worked on
AutoInfo’s engagement, had focused on the transportation logistics space since
2002.24
3. Management’s Financial Projections
To implement the sales process, Stephens asked Management to prepare a
bottoms-up five-year financial forecast (the “Management Projections”).25
Stephens specified that because they would be used to market the Company, the
projections should be optimistic.26 Management had never prepared multi-year
projections before and its first attempt fell largely on Wunderlich’s (its CFO)
shoulders.27 Internally, Management doubted its ability to forecast the Company’s
future performance accurately and perceived its attempt as “a bit of a chuckle and a
joke.”28 It questioned how to go about a process it had never before attempted.29
Recognizing that the Management Projections would be used to shop the
Company, Wunderlich focused on painting an “aggressively optimistic” picture.30
23
Trial Tr. 280 (Miller).
24
Trial Tr. 274 (Miller).
25
Trial Tr. 281 (Miller).
26
Id.
27
Trial Tr. 481-82 (Wachtel).
28
Williams Dep. 170.
29
Trial Tr. 354 (Williams).
30
Wunderlich Dep. 49. See also Caple Dep. 38 (“[The Management Projections
were] about the most optimistic you could make them.”); Trial Tr. 237 (Puglisi)
6
Williams, AutoInfo’s President, helped develop the forecast by projecting agent
revenue.31 He started with each agent’s historical revenue and “took the most
optimistic view of [the] agents’ performance in the marketplace . . . .”32 He
categorized agents by size and assumed that larger agents would grow at a lower
percentage than smaller agents.”33 Williams testified that there “was no science”
behind those assumptions.34 He also looked at agent-by-agent historical results and
predicted, based on knowledge of the individual agents, how much the agent’s
business could grow during 2012-2013.35 Those growth assumptions were
extrapolated to later years.36 The Management Projections also included estimates
of how successfully the Company would recruit new agents.37
(“They were optimistic. I didn’t see anybody who said they weren’t optimistic.”);
Trial Tr. 359 (Williams) (“Overly optimistic, really to the exclusion of external and
internal risk factors that otherwise are part of the business.”); Trial Tr. 399
(Williams) (“[W]e prepared those projections with the most optimistic view of the
future that we could possible conceive.”).
31
Williams Dep. 168-70. Weiss and AutoInfo’s director of corporate marketing
and communications assisted this effort. Trial Tr. 395 (Williams).
32
Trial Tr. 396 (Williams).
33
Williams Dep. 175.
34
Id.
35
Id. at 169.
36
Id. at 169-70.
37
Id. at 168.
7
4. Comvest Emerges as the Highest Bidder
In the spring of 2012, Stephens contacted 164 potential strategic and
financial acquirers, focusing on those most interested in the transportation space.38
Approximately seventy bidders signed non-disclosure agreements (“NDAs”) and
received a Confidential Information Memorandum (“CIM”).39 Those interested
were provided several weeks for due diligence before a deadline to submit an
indication of interest (“IOI”).40 By the end of May, ten bidders had presented IOIs,
with bids ranging from $0.90-$1.36 per share.41 Nine moved on to a second round
of the sales process, at which point they attended Management presentations and
received access to an electronic data room.42
On June 28, 2012, the Board formed a special committee (the “Special
Committee”) to evaluate the competing offers. The Special Committee consisted
38
Trial Tr. 33 (Patterson); Trial Tr. 282-83 (Miller). The Board opted against
publicly announcing a sales process because it did not want to disrupt its agent
base. The possibility of losing agents is particularly troublesome for a 100%
agent-based company because the agents maintain all client relationships. Trial Tr.
33-34 (Patterson). If a public announcement caused agents to leave the company,
then AutoInfo would not likely have maintained its revenue and earnings. Trial
Tr. 34 (Patterson).
39
Trial Tr. 285 (Miller).
40
Id.
41
JX 295 (“Stephens’s Special Committee Presentation”) at 9.
42
Id. The one party that did not advance to the next round had provided the lowest
IOI. Trial Tr. 287 (Miller).
8
of the three outside directors, with Patterson serving as chair.43 It proceeded, with
the assistance of a legal advisor and a financial advisor, to review the bids.44
By July, three would-be acquirers had submitted written letters of intent
(“LOI”) and two others had presented verbal valuation ranges.45 After receiving
legal advice regarding its fiduciary duties, the Special Committee weighed the
proposals as against each other and the alternative option of foregoing a sale at that
time.46 It decided to continue with the sales process and instructed Stephens to
negotiate with the bidders over price.47
Later that month, Stephens updated the Special Committee with final terms
for the written bids. HIG Capital (“HIG”) had made the highest offer at $1.30 per
share.48 The Special Committee determined that the highest offer was also the best
and recommended that the Board pursue a transaction with HIG. The Board
accepted this determination and on August 14, 2012, executed an LOI at the $1.30
43
JX 114 (June 28, 2012, Board minutes).
44
Patterson Dep. 102-03.
45
Stephens’s Special Committee Presentation 9. Comvest Partners was one of the
bidders which expressed verbal interest with the caveat that it would need
additional time for due diligence because of conflicts with other transactions.
JX 117 (Stephens’s July 2, 2012, Process Update) at 4.
46
Apr. 1, 2013, AutoInfo Schedule 14A at 26.
47
Id. at 26-27.
48
Stephens’s Special Committee Presentation 9.
9
per share price, which provided for a forty-five day exclusivity period to negotiate
and perform further due diligence.49
HIG conducted due diligence for the next thirty days but by mid-September,
it decided not to proceed with the purchase.50 HIG’s lead partner on the deal had
left the firm, apparently due to various disagreements with his colleagues,
including whether HIG should decrease its offer for AutoInfo.51 After that
partner’s departure, HIG opted against pursuing AutoInfo.52 The parties
terminated their LOI, and AutoInfo decided to continue with the sales process.
Stephens contacted previously interested parties, as well as others it recommended
to AutoInfo.53
By October 2012, two interested parties had submitted written LOIs and two
others had indicated interest verbally. The highest offer came from Comvest
Partners (“Comvest”) and valued the Company at $1.26 per share.54 The others
were substantially lower, ranging from $1.00-$1.07 per share.55 After determining
that Comvest’s offer was the best, the Special Committee recommended that the
Board pursue that transaction. The Board unanimously agreed and on
49
Apr. 1, 2013, AutoInfo Schedule 14A at 27.
50
Id.
51
Trial Tr. 290 (Miller).
52
Id.
53
Id.
54
Stephens’s Special Committee Presentation 9.
55
Id.
10
November 12, 2012, AutoInfo executed an LOI with Comvest at $1.26 per share
with a thirty day exclusivity period.56 Comvest then hired accounting, legal,
industry, and other advisors to conduct due diligence.57
5. Comvest’s Due Diligence Process
Comvest hired L.E.K. Consulting (“LEK”), a strategy consultant, to assess
AutoInfo’s competitive positioning in the trucking freight brokerage market.58
LEK evaluated growth trends and dynamics in the brokerage market generally, as
well as concerns associated with AutoInfo’s agent-based business.59 Comvest
considered LEK’s findings as very positive.60
LEK’s analysis came relatively early in the due diligence process, and as
that process evolved, Comvest learned of potential issues associated with
AutoInfo’s business.61 For example, AutoInfo’s infrastructure for recruiting new
agents, which represented the lifeblood of the Company, was lacking.62 Comvest
determined that it would need to address that deficiency, and others, before it could
56
Id.
57
Id.
58
L.E.K. Consulting Due Diligence Presentation 3.
59
Trial Tr. 445 (Caple).
60
Trial Tr. 446 (Caple).
61
Id.
62
Id.
11
effectively recruit agents and grow AutoInfo’s business.63 Its biggest concerns,
however, arose during its accounting due diligence.
Comvest retained McGladrey LLP (“McGladrey”) to perform financial due
diligence; its work included conducting a quality of earnings analysis to test the
accuracy of the Company’s stated historical earnings and its ability to achieve
projections.64 McGladrey began its review in November 2012, with Wunderlich,
AutoInfo’s CFO, serving as its primary Company contact. McGladrey was
immediately taken aback by the poor quality of AutoInfo’s financial records,
which were unusually bad for a publicly traded company.65 The state of the
financials caused the due diligence process to be more difficult than McGladrey
had anticipated.66
McGladrey was surprised that AutoInfo used QuickBooks, accounting
software popular among small businesses, but rarely employed by public
companies.67 Also troubling to McGladrey was the fact that a Florida-based public
company would engage a one-office, Connecticut-based accounting firm as its
outside auditor.68 More importantly, McGladrey believed that some of AutoInfo’s
63
Trial Tr. 447 (Caple).
64
Trial Tr. 405, 408 (Spizman).
65
Trial Tr. 412 (Spizman); JX 159 (emails among McGladrey personnel).
66
Trial Tr. 414 (Spizman); JX 159. McGladrey also considered Wunderlich to be
“in over his head” as a public company CFO. Trial Tr. 424 (Spizman).
67
Trial Tr. 414-15 (Spizman).
68
Trial Tr. 415 (Spizman).
12
accounting practices violated generally accepted accounting principles.69
McGladrey raised these concerns with an increasingly troubled Comvest.70
In December 2012, McGladrey reported its findings to Comvest (the
“McGladrey Report”).71 AutoInfo’s Management had estimated the Company’s
2012 adjusted EBITDA as $10 million.72 McGladrey concluded that $7.7 million
was an appropriate estimate, representing a 23% reduction.73 Comvest considered
the McGladrey Report a “huge problem” with the potential to “blow[] up” the
deal.74 Not only was AutoInfo’s EBITDA apparently much lower than initially
assumed, but there was “a whole series of weaknesses in the company’s financial
reporting practices . . . .”75
AutoInfo responded to the McGladrey Report through a memorandum
prepared by Wunderlich.76 McGladrey considered the rebuttal unconvincing.77 At
the beginning of January, Wunderlich, Wachtel, and a representative from
Stephens met with a Comvest representative to discuss the McGladrey Report and
69
Spizman Dep. 65-66.
70
Trial Tr. 417 (Spizman).
71
JX 223.
72
Trial Tr. 418-19 (Spizman).
73
Trial Tr. 419 (Spizman).
74
Trial Tr. 453 (Caple).
75
Caple Dep. 116.
76
JX 208; Trial Tr. 419-20 (Spizman).
77
Trial Tr. 420 (Spizman).
13
AutoInfo’s response.78 While Comvest listened to AutoInfo’s arguments, it
remained convinced that the McGladrey Report raised valid issues and McGladrey
did not change its conclusions.
After that meeting, Comvest lowered its offer to $0.96 per share and
AutoInfo countered at $1.15.79 During ensuing negotiations, Comvest learned that
AutoInfo had guaranteed some loans, the existence of which had been undisclosed
and unreported. Some of the borrower’s creditors had filed an involuntary
bankruptcy petition and AutoInfo was facing the possibility of having to satisfy the
guarantees.80 Comvest was concerned not only by AutoInfo’s increased liabilities,
but more importantly, it was troubled by the fact that the guarantees had not been
properly identified in the first place.81 Its confidence in the quality of AutoInfo’s
financial information and controls further deteriorated.82
On January 18, 2013, the Special Committee and Comvest agreed to a new
price of $1.06 per share.83 Comvest had successfully negotiated for Wachtel
(AutoInfo’s CEO) to roll over $500,000 and for Weiss (another executive) to roll
78
JX 211 (email from Wachtel to Patterson regarding Comvest meeting).
79
Trial Tr. 294 (Miller).
80
JX 231 (memo to Special Committee).
81
Trial Tr. 460 (Caple).
82
Caple Dep. 176-77.
83
JX 236 (emails among Comvest employees).
14
over 25% of his deal proceeds.84 The deal process then resumed, until discovery of
another accounting deficiency. AutoInfo had improperly booked a transaction,
worth approximately $1,000,000 in EBITDA, in the third quarter of 2012 before
the deal had closed.85 Comvest was shocked at this revelation and was worried
that AutoInfo would need to restate its financials. Characterizing his reaction,
John Caple, Comvest’s lead partner on the AutoInfo deal, testified, “As much as I
had seen financial weaknesses in the business, the fact that the company could
book a million dollar transaction that hadn’t actually happened, I’ve just never seen
that before in any business I’ve worked with, public or private.”86 AutoInfo
determined, after an approximately two week review, that its financials would not
need to be restated.87 Nonetheless, Comvest was “disturb[ed that the error] could
have happened at all, particularly given the size and the impact of the
transaction.”88 Comvest’s already low confidence in AutoInfo’s Management and
internal controls eroded further and it revised its offer to $1.00 per share.89
84
Id. Comvest demanded the rollover agreements as a condition to executing at
$1.06 so that Management would retain an economic stake in AutoInfo’s business
moving forward. Trial Tr. 458-59 (Caple).
85
Trial Tr. 460 (Caple).
86
Id.
87
Trial Tr. 461 (Caple). “The auditors determined that because the transaction
could be closed now that it was simply . . . sort of a paperwork error.” Id.
88
Id.
89
Id.
15
On February 28, 2013, after additional negotiations, the parties ultimately
reached an agreement at $1.05 per share, with Wachtel entering into an
indemnification agreement for potential breaches of AutoInfo’s representations and
warranties, whereby $500,000 of his proceeds would be held in escrow.90 The
Board approved the Merger pursuant to the Special Committee’s unanimous
recommendation. Stephens had provided a fairness opinion and presentation to the
Special Committee. AutoInfo announced the Merger on March 1, 2013.91
On April 25, 2013, AutoInfo’s stockholders approved the deal and the
transaction closed later that day. No topping bids had emerged between the deal’s
announcement and closing.92
II. THE PARTIES’ COMPETING VALUATIONS
Both parties retained well-qualified experts to opine on the fair value of
Petitioners’ stock as of the date of the Merger. Petitioners’ expert, Donald Puglisi
(“Puglisi”), suggests that AutoInfo’s fair value was $2.60 per share. He places
equal weight on three valuation calculations: a discounted cash flow (“DCF”)
90
Trial Tr. 462 (Caple); Apr. 1, 2013, AutoInfo Form DEFM14A at 5. Wachtel,
Williams, and Weiss entered a rollover agreement whereby they acquired an
indirect ownership interest in AutoInfo upon the closing of the Merger. Wachtel
and Williams also entered into new employment agreements with AutoInfo. Id.
91
JX 302.
92
This was despite at least one stockholder’s attempts to solicit topping bids. See,
e.g., JX 309; JX 314; JX 318.
16
analysis, and two comparable companies analyses, one using a historical based
multiple and the other a forward looking multiple.93
AutoInfo’s expert, Mark Zmijewski (“Zmijewski”), submits that AutoInfo’s
fair value on the date of the Merger was $0.967 per share. Unlike Puglisi,
Zmijewski does not believe that a DCF or comparable companies analysis can be
reliably performed with available data. Instead, he analyzed the Merger price and
the market evidence regarding the strength of AutoInfo’s sales process. He
concluded that the Merger price, minus cost savings arising from the Merger, is the
best available evidence of the Company’s fair value on the Merger date.94
III. ANALYSIS
A. The Appraisal Statute
Under 8 Del. C. § 262, stockholders who elect against participating in
certain merger transactions may petition the Court to determine the fair value of
their stock.95 Assuming all procedural requirements are satisfied, the Court
determine[s] the fair value of the shares exclusive of any element of
value arising from the accomplishment or expectation of the merger or
consolidation, together with interest, if any, to be paid upon the
93
JX 380 (“Puglisi Opening Report”).
94
JX 381 (“Zmijewski Opening Report”). Zmijewski did conduct a DCF analysis,
for illustrative purposes, for his rebuttal expert report. See JX 415 (“Zmijewski
Rebuttal Report”) at 22. That did not affect his fair value conclusion.
95
8 Del. C. § 262.
17
amount determined to be the fair value. In determining such fair
value, the Court . . . take[s] into account all relevant factors.96
“Fair value” represents “the value to a stockholder of the firm as a going
concern, as opposed to the firm’s value in the context of an acquisition or other
transaction.”97 To discharge its statutory responsibility, the Court independently
evaluates the evidence concerning fair value and does not presumptively defer to
any particular valuation metric.98 The Court may consider “any techniques or
methods which are generally considered acceptable in the financial community and
otherwise admissible in court . . . .”99 Depending on the case, a DCF analysis, a
comparable transactions analysis, a comparable companies analysis, or the merger
price itself may inform the Court’s determination.100 “[A]n arms-length merger
price resulting from an effective market check” is a strong indicator of actual
value.101
In a Section 262 appraisal proceeding, “both sides have the burden of
proving their respective valuation positions by a preponderance of the evidence.”102
96
8 Del. C. § 262(h). There is no dispute that Petitioners have met all procedural
requirements.
97
Golden Telecom, Inc. v. Global GT LP, 11 A.3d 214, 217 (Del. 2010).
98
Id. at 217-18.
99
Weinberger v. UOP, Inc., 457 A.2d 701, 713 (Del. 1983).
100
Huff Fund Inv. P’ship v. CKx, Inc., 2013 WL 5878807, at *9 (Del. Ch. Nov. 1,
2013), aff’d, -- A.3d --, 2015 WL 631586 (Del. Feb. 12, 2015) (“Huff”).
101
Global GT LP v. Golden Telecom, Inc., 993 A.2d 497, 507 (Del. Ch. 2010),
aff’d, 11 A.3d 214 (Del. 2010).
102
M.G. Bancorporation, Inc. v. Le Beau, 737 A.2d 513, 520 (Del. 1999).
18
The Court may select one of the parties’ valuation models, make adjustments to a
proffered model, or fashion its own framework.103
B. Puglisi’s DCF Analysis
Puglisi bases his valuation of AutoInfo in part on a DCF analysis. “DCF, in
theory, is not a difficult calculation to make—five-year cash flow projections
combined with a terminal value are discounted to their present value to produce an
overall enterprise value.”104 However, when reliable inputs are unavailable, “any
values generated by a DCF analysis are meaningless.”105 Puglisi used the
Management Projections in his DCF calculation. The first question is: are those
projections reliable?106
The Court will often give weight to management projections made in the
regular course of business because “management ordinarily has the best first-hand
knowledge of a company’s operations.”107 Nonetheless, “management projections
[may be disregarded] where the company’s use of such projections was
unprecedented, where the projections were created in anticipation of litigation, or
where the projections were created for the purpose of obtaining benefits outside the
103
Cede & Co. v. Technicolor, Inc., 684 A.2d 289, 299 (Del. 1996).
104
Huff, 2013 WL 5878807, at *9.
105
Id.106
That the projections were not ultimately realized does not foreclose the potential
conclusion that they were reliable as of their preparation date.
107
Doft & Co. v. Travelocity.com Inc., 2004 WL 1152338, at *5 (Del. Ch. May 20,
2004).
19
company’s ordinary course of business.”108 If management had never prepared
projections beyond the current fiscal year, the Court may be skeptical of its first
attempt.109
Here, Petitioners have failed to establish that the Management Projections
can be relied upon.110 Management prepared them at Stephens’s request and with
the guidance that they “need[ed] to be optimistic” to maximize the effort to market
the Company.111 Management had never prepared anything resembling the
Management Projections before and “hadn’t analyzed the business historically in a
way that would allow [it] to predict the future.”112 Stephens had advised, “You’re
trying to sell the business. You need to paint the most optimistic and bright current
and future condition of the company that you can. All positive. Let’s get the most
interest by painting the most positive picture of this business.”113
108
Huff, 2013 WL 5878807, at *9.
109
Merion Capital, L.P. v. 3M Cogent, Inc., 2013 WL 3793896, at *11 (Del. Ch.
July 8, 2013) (citing Gearreald v. Just Care, Inc., 2012 WL 1569818, at *4 (Del.
Ch. Apr. 30, 2012)).
110
AutoInfo’s expert agrees that “the Management Projections are not a reliable
forecast of the Company’s expected future performance and, thus, would not yield
a reliable indication of the Fair Value of AutoInfo common stock.” Zmijewski
Opening Report ¶ 53.
111
Trial Tr. 281-82 (Miller).
112
Trial Tr. 354 (Williams).
113
Trial Tr. 355 (Williams).
20
As discussed in Section I.C.3 above, the Management Projections were
indisputably optimistic.114 Puglisi, Petitioners’ own expert, testified that he would
have implied a discount factor to back out the optimism if the record had provided
a basis for calculating one.115 Even if Management had not been motivated to paint
a bright picture, its projections would have been unreliable. Again, Management
itself had no confidence in its ability to forecast.116 If Management could not have
been trusted to produce credible projections in the ordinary course of business, the
projections it created during the sales process deserve little deference. Because
Petitioners have failed to establish the credibility of a key component in their
expert’s DCF analysis, the Court gives that analysis no weight.117
C. Puglisi’s Comparable Companies Analyses
Puglisi performed two comparable companies analyses, one using a 2012
EBITDA figure derived from AutoInfo’s 2012 10-K, and the other using an
estimated 2013 EBITDA created by modifying the Management Projections. To
perform a comparable companies analysis, one must first identify a set of actively
traded public companies sharing similar business characteristics with the subject
114
See supra note 30.
115
Trial Tr. 237 (Puglisi).
116
See supra text accompanying note 28 (describing the Management Projections
as “a bit of a chuckle and a joke”).
117
This conclusion is corroborated by the dramatic difference between Puglisi’s
DCF value and the Merger price. As discussed below, the Merger price, unlike
Puglisi’s DCF output, is indicative of fair value.
21
company. Using available information, one then derives a valuation multiple that,
when multiplied by a relevant financial performance metric, such as EBITDA,
provides an estimate of the value of a company as a whole.
The Court may credit a comparable companies analysis in an appraisal
proceeding; however, “[t]he utility of the comparable company approach depends
on the similarity between the company the court is valuing and the companies used
for comparison.”118 Petitioners bear the burden of proving that Puglisi’s
“comparables are truly comparable.”119 Because they fail to meet their burden, the
Court gives no weight to Puglisi’s comparable companies analyses.120
1. AutoInfo is Significantly Smaller than Puglisi’s Supposed Comparables
The Court may reject comparable companies analyses based on purported
comparables that differ significantly in size from the company being appraised.121
118
In re Radiology Assocs., Inc. Litig., 611 A.2d 485, 490 (Del. Ch. 1991).
119
In re AT & T Mobility Wireless Operations Hldgs. Appraisal Litig., 2013 WL
3865099, at *2 (Del. Ch. June 24, 2013) (quoting ONTI, Inc. v. Integra Bank, 751
A.2d 904, 916 (Del. Ch. 1999)).
120
Of course, if the Court had accepted that the comparables are truly comparable,
it would have needed to test the reliability of the EBITDA figures that Puglisi used
as inputs.
121
See, e.g., Merion Capital, 2013 WL 3793896, at *6 (“[I]t would be
inappropriate to compare a company with an enterprise value of $14.7 million . . .
to a company . . . with an enterprise value more than 25 times higher.”); Reis v.
Hazelett Strip-Casting Corp., 28 A.3d 442, 477 (Del. Ch. 2011) (rejecting the
comparable companies approach because the comparables were “much bigger than
[the subject company] . . . [and] enjoy[ed] better access to capital . . .”); In re PNB
Hldg. Co. S’holders Litig., 2006 WL 2403999, at *25 n.125 (Del. Ch. Aug. 18,
2006) (finding a comparable companies analysis flawed where the “comparable
22
It is undisputed that Puglisi’s comparables are all significantly larger than
AutoInfo. As of the Merger date, their market capitalizations ranged from more
than twice, to more than 300 times, AutoInfo’s size.122 All but two of Puglisi’s
comparables had a market capitalization more than ten times AutoInfo’s. While
recognizing this fact, Petitioners argue that size, while relevant in other contexts, is
not a determining factor here.
Puglisi testified that he did not observe a meaningful relationship between a
company’s size and its multiple among his comparables. He could not recall “ever
discriminating inclusion in comparable companies based on company size . . .
[because] size itself should not have an impact on the ultimate valuation.”123
Although there may be little theoretical basis for discriminating comparables based
on size, doing so has empirical support and is common both in practice and in this
Court.124 Zmijewski suggests that it would be inappropriate to select comparables
publicly-traded companies all were significantly larger than [the subject company],
with one having assets of $587 million as compared to [the subject company’s]
assets of $126 million . . .”); Gray v. Cytokine Pharmasciences, Inc., 2002 WL
853549, at *9 n.19 (Del. Ch. Apr. 25, 2002) (finding a comparable companies
analysis unreliable where the comparables “taken together had a market
capitalization with a median 24 times higher than [the appraised company] . . . [and
t]he median revenue of the comparable companies was 12 times larger than [the
appraised company]”).
122
Puglisi Opening Report Ex. C.
123
Trial Tr. 155-56 (Puglisi).
124
See supra note 121. See also ROBERT W. HOLTHAUSEN & MARK E. ZMIJEWSKI,
CORPORATE VALUATION THEORY, EVIDENCE & PRACTICE 525 (Cambridge
Business Publishers, LLC 2014). Puglisi did employ a size premium in his DCF
23
without regard to relative market capitalization without otherwise controlling for
risk and other differences.125
All else equal, smaller firms are riskier and thus face higher costs of equity
capital. This higher cost of capital leads to lower market multiples.126 Miller,
Stephens’s representative, suggests
Typically in this sector, small cap companies tend to be valued
at lower multiples. That’s generally been the case in the . . . dozen
years that I’ve spent covering this sector. The market tends to ascribe
premium multiples to companies that are larger . . . [and] are
considered more stable businesses. And therefore, investors are
willing to . . . afford those companies . . . a higher trading multiple.127
Before delivering its fairness opinion, Stephens performed a comparable
companies analysis. Based on its experience in the transportation services
industry, Stephens, unlike Puglisi, did not rely on the median multiple of its
comparables. It selected a lower multiple range, based on differences between
AutoInfo and the comparables, including size, business model, and the quality of
management.128 Stephens grouped its comparable companies by size, which
showed a relationship between size and multiples.129
analysis, thus recognizing the empirically observed size effect whereby the capital
asset pricing model understates the returns to small firms. See Trial Tr. 198-99
(Puglisi).
125
Zmijewski Rebuttal Report ¶ 30.
126
Id. at ¶ 28. See also Merion Capital, 2013 WL 3793896, at *6.
127
Miller Dep. 148.
128
Miller Dep. 154-55.
129
Stephens’s Special Committee Presentation 18.
24
While Petitioners criticize Stephens’s size grouping as arbitrary and self-
serving, in its initial July 29, 2011, Strategic Initiatives Overview presentation to
AutoInfo, Stephens highlighted the fact that AutoInfo had “consistently traded at
valuation multiples well below its peer group due to the Company’s relatively
small scale . . . .”130 Petitioners have failed to show that the size difference
between AutoInfo and Puglisi’s supposedly comparable companies is
immaterial.131
2. AutoInfo’s 100% Agent-Based Model
Puglisi did not consider the differences between freight brokerage businesses
that use the company store model and those that employ an agent-based model as
important for valuation purposes.132 As described in Section I.A above, in a
company store model, “[b]rokers are direct employees of the company,” while in
an agent-based model, the brokers are independent contractors. 133 According to
130
Stephens’s Strategic Initiatives Overview 5.
131
Petitioners note the Court’s usual skepticism of “an expert [who] throws out his
sample and simply chooses his own multiple in a directional variation from the
median and mean that serves his client’s cause . . . .” In re Orchard Enters., Inc.,
2012 WL 2923305, at *11 (Del. Ch. July 18, 2012). While Petitioners contend that
Puglisi’s use of a median multiple is thus preferable to accepting Stephens’s lower
numbers, AutoInfo has not suggested that the Court rely on any comparable
companies analysis. Also, Stephens’s choice of multiple was not a post hoc
determination made during litigation, but a reasoned selection based on its industry
experience. Regardless, the Court need not consider the soundness of Stephens’s
choice to view Puglisi’s methodology as unreliable.
132
Puglisi Dep. 125.
133
L.E.K. Consulting Due Diligence Presentation 32.
25
Miller, who has years of experience in the transportation sector, “agent-based
models . . . are generally less desirable. They’re perceived as riskier. The
company does not have control over the customer relationship. The agent does.
And so the agent-based models are generally . . . less desirable and generally they
tend to trade at lower multiples than the company store models.”134
That the market perceives the agent-based model as inferior was
corroborated by the reaction that one AutoInfo stockholder received while
soliciting topping bids for the Company. That stockholder learned that “the agent-
based model with no company-owned locations, especially in important shipping
hubs, was a bigger deal to potential acquirers than . . . [initially] realized.”135
AutoInfo’s 100% agent-based model was a “problem” for potential buyers.136
At trial, Puglisi, who lacks Miller’s experience in the freight brokerage
sector, could not identify which of his comparable companies used which type of
business model, but suspected that the majority were agent-based.137 Miller
testified specifically regarding the business models of Stephens’s comparables and
134
Trial Tr. 304 (Miller). Miller testified regarding the many differences between
AutoInfo and the supposedly comparable companies. See Trial Tr. 302-15
(Miller).
135
JX 357 (email exchange regarding AutoInfo’s valuation).
136
JX 346 (email to uninterested solicited buyer).
137
Trial Tr. 238-39 (Puglisi).
26
explained that they mostly use company store models.138 In its fairness opinion,
Stephens had taken advantage of its industry experience and its knowledge of
AutoInfo’s business to select a below-median multiple for its comparable
companies analysis.139 Petitioners have not established that the differences
between AutoInfo’s business model and those of Puglisi’s comparable companies
are unimportant.
3. Summary of Puglisi’s Comparable Companies Analyses
Because the weight of the evidence suggests that size and business model
affect the multiples at which companies trade in the freight brokerage industry,
Puglisi’s comparable companies analyses are not reliable indicators of value. The
Court’s confidence in this conclusion is bolstered by the facts that (i) all of the bids
received by AutoInfo during the sales process implied market multiples well below
Puglisi’s, and (ii) AutoInfo ultimately sold, through a thorough sales process, at a
price less than half of Puglisi’s comparable companies valuations.140 The Court
was unable independently to derive in any reasoned manner a valuation multiple
from the purported comparables. Accordingly, the Court gives no weight to any
comparable companies analysis.
138
Trial Tr. 302-14 (Miller). Some companies used a mixed model. AutoInfo
used a 100% agent-based model.
139
Trial Tr. 314-15 (Miller).
140
See RX-9; RX-10 (demonstrative exhibits charting market multiples implied by
bids for AutoInfo).
27
D. Merger Price
Zmijewski, AutoInfo’s expert, relies on the Merger price as a reliable
indication of AutoInfo’s fair value at the time of the Merger. “[W]here no
comparable companies, comparable transactions, or reliable cash flow projections
exist, . . . the merger price [may be] the most reliable indicator of value.”141
Nonetheless, the Court will give little weight to a merger price unless the record
supports its reliability.
The dependability of a transaction price is only as strong as the process by
which it was negotiated.142 For example, a transaction that implicates self-
interested parties or an inadequate market check may generate a price divergent
from fair value. Conversely, where a company “was marketed to potential
buyers . . . [through a process that was] thorough, effective, and free from any
spectre of self-interest or disloyalty,” the outcome of that process is significant.143
Petitioners argue that the Merger price deserves no weight because (i) the
Merger price is not a business valuation methodology, (ii) the Court cannot rely on
the price if no business valuation methodology, e.g., a DCF analysis, was
performed to corroborate the price, and (iii) even if the Merger price could be
considered, AutoInfo’s sales process was deficient.
141
Huff, 2013 WL 5878807, at *13.
142
Id.143
Id.28
Petitioners’ first two contentions are easily dismissed. As discussed, this
Court can, and has, relied on a merger price when appraising a company. When it
is the best indicator of value, the Court may assign 100% weight to the negotiated
price.144 Although the Court may not presumptively defer to price, no particular
valuation methodology must provide corroboration. Rather, the Court may, in its
discretion, look to any “evidence tending to show that [the merger price] represents
the going concern value of the company rather than just the value of the company
to one specific buyer.”145 Here, evidence regarding AutoInfo’s sales process
substantiates the reliability of the Merger price.
The manner by which AutoInfo was sold is described in Section I.C. above.
This case does not involve self-interest or disloyalty; nothing like a controlling
stockholder’s freezing out the minority is at issue. The Merger was negotiated at
arm’s length, without compulsion, and with adequate information. It was the result
of competition among many potential acquirers. However, Petitioners argue that
the sales process was flawed and cannot be expected to have produced a price
representative of value. Based on the evidence, the Court concludes that
Petitioners’ objections, discussed next, are either unwarranted or overblown.
144
See, e.g., Union Ill. 1995 Inv. Ltd. P’ship v. Union Fin. Gp., Ltd., 847 A.2d 340,
357 (Del. Ch. 2004).
145
M.P.M. Enters., Inc. v. Gilbert, 731 A.2d 790, 797 (Del. 1999).
29
1. Lack of Analyst Coverage
AutoInfo was thinly traded and lacked financial analyst coverage.
Petitioners contend that the market underpriced the Company because it was
ignorant of its potential. While “[t]he court cannot defer to market price as a
measure of fair value if the stock has not been traded actively in a liquid
market,”146 the Merger price does not reflect the value that a potentially
uninformed market attributed to AutoInfo. The Merger price represented a 22%
premium to AutoInfo’s average stock price during the six months before
February 28, 2013, the last trading day before public announcement of the
Merger.147 At no time in the two years before the Merger’s announcement had the
market price for the Company’s stock reached $1.00.148 Further, the Merger price
exceeded the highest price that AutoInfo stock had reached during the previous
five years.149
To shop the Company, AutoInfo retained an experienced investment bank
with knowledge of the transportation industry. Stephens’s fee had an incentive-
based component, which allowed the bank to earn a higher percentage fee the
146
Applebaum v. Avaya, Inc., 812 A.2d 880, 890 (Del. 2002). In fact, as discussed,
in an appraisal, the Court may never defer to market price without independently
testing its reliability.
147
Apr. 1, 2013, AutoInfo Schedule 14A at 31.
148
Id.149
Stephens’s Special Committee Presentation 11.
30
larger the deal.150 Stephens reached out to and provided information on AutoInfo
to many potential bidders. Part of the reason for hiring the bank would have been
to educate the market and assure the Company that it was not leaving value on the
table.151 The Board formed a Special Committee to pursue the sales process.
Ultimately, AutoInfo was sold at a premium to market. Despite attempts by a
stockholder to solicit interest, no topping bid emerged during the time frame
between announcement and closing of the Merger.152 While the market may have
been uninformed about AutoInfo before the sales process, it subsequently gained
ample information.
2. Alleged Pressure from Large Stockholders
Petitioners contend that large stockholders pressured the Board to sell
quickly. Approximately 31.4% of AutoInfo’s voting power was held by Baker
Street and Kinderhook.153 According to Petitioners, those hedge funds sent a clear
message that if a liquidity event were not achieved, then they would get active and
Management would potentially face a control contest.
150
Trial Tr. 280 (Miller).
151
In explaining Stephens’s request that the Management Projections be optimistic,
Miller stated “You certainly don’t want to be conservative and leave potential
shareholder value on the table.” Trial Tr. 282 (Miller).
152
One investment advisor who had initially been skeptical of the merger
concluded, after learning of the issues associated with an agent-based model, that
“the deal was done at a fair, or very close to fair, price.” JX 357 (email to the
soliciting stockholder).
153
Kinderhook held an 18.4% stake and Baker Street held 13%.
31
Baker Street purchased its stake in the Company in November 2011. By that
time, AutoInfo had already begun to consider strategic alternatives, including a
potential sale. The Board had reached out informally to potential purchasers
months before Baker Street became a stockholder. Stephens’s July 29, 2011,
presentation to AutoInfo had indicated “that now is an opportune time to explore
initiatives to maximize shareholder value, including . . . [a c]hange of control
transaction.”154 By the time Baker Street arrived on the scene, AutoInfo was
already contemplating the selection of a bank to lead the formal sales process.155
Unlike Baker Street, Kinderhook was not adamant that AutoInfo be sold.
Rather, like the Board, Kinderhook desired change to address AutoInfo’s low stock
price.156 Patterson, the Special Committee’s chair, testified that neither Baker
Street nor Kinderhook impacted the sales process. Before retaining Stephens, the
Board had received early indications of interest and “absolutely” could have sold
quickly if the terms had been right.157 Instead, the Board retained Stephens and
embarked on a sales process lasting over a year. Near the end of that process,
Patterson told the rest of the Special Committee “I plan to tell [Comvest] to pay
$1.06 or walk away.”158 If necessary, the Special Committee was prepared to
154
Stephens’s Strategic Initiatives Overview 5.
155
Trial Tr. 20 (Patterson).
156
Trial Tr. 23-24 (Patterson).
157
Trial Tr. 32 (Patterson).
158
JX 277 (email from Patterson to other Special Committee members).
32
“regroup, push some changes through and clean up” for a future sale. 159 Based on
the evidence, neither Baker Street nor Kinderhook appear to have materially
impacted the sales process.
3. Negotiations with Comvest
Petitioners next argue that Comvest completely overwhelmed AutoInfo’s
Management and Board during negotiations. More specifically, they contend that
Comvest commissioned the McGladrey Report as a tool to drive down the Merger
price. According to Petitioners, AutoInfo was incapable of adequately responding
to that report.
Hiring an accounting firm to conduct due diligence is standard practice for
Comvest.160 While due diligence sometimes flags issues, in other cases, the
process is positive and the accounting firm concludes that the target company is
actually a better deal than Comvest initially believed.161 McGladrey was not the
only outside firm hired to conduct due diligence. For example, Comvest engaged a
strategy consultant, whose review of AutoInfo’s business was very positive.162
159
Id. Petitioners argue that Baker Street had demanded a deal by June 2012 and
had suggested that any sale at or above $1.00 per share would suffice. The Board
did not approve the Merger until 2013 and the Special Committee was “not
comfortable” with a $1.00 price. See id.160
Trial Tr. 451-52 (Caple).
161
Trial Tr. 452-53 (Caple).
162
Trial Tr. 445-46 (Caple).
33
It is mostly undisputed that AutoInfo’s CFO was below-average, the
Company used relatively unsophisticated accounting software, and its accounting
records contained errors discovered throughout negotiations. There is room to
debate whether all of McGladrey’s adjustments to AutoInfo’s financials were
necessary. However, the record does not support the notion that McGladrey’s
auditors would have sacrificed their professional independence to benefit Comvest
on this one particular transaction. AutoInfo did attempt to rebut the McGladrey
Report, but many of McGladrey’s findings “were valid issues.”163 Because
AutoInfo had sub-par accounting and financial controls, McGladrey was
understandably alert to potential problems, and Comvest was understandably
concerned by the issues raised. Comvest viewed the agreement it eventually
reached with AutoInfo as inferior to the deal it had initially anticipated.164 The
record does not support the allegation that McGladrey was a hired gun employed to
overwhelm AutoInfo.165
163
Trial Tr. 334 (Miller); see also Trial Tr. 105 (Patterson).
164
Trial Tr. 458 (Caple).
165
Those contacted by an AutoInfo stockholder soliciting topping bids for the
company shared at least some of McGladrey’s concern. See, e.g., JX 320 (email
from accountant questioning “why . . . a Shelton, CT based firm (not even a
regional firm) [would] be auditing a Miami based company . . .”); JX 325 (email
from disinterested party stating: “Just as a personal aside I also wonder about the
accounting. They convert notes to goodwill ($10M) in exchange for cash flow but
then they don’t amortize the goodwill against that cash flow at all. I doubt that
cash flow will continue infinitely.”).
34
4. Stephens’s Process
Petitioners suggest that (i) Stephens’s market canvas was unfocused,
(ii) Stephens improperly suggested a valuation of AutoInfo to some bidders,
(iii) Stephens did not provide a formal valuation of the Company until the Merger
was negotiated, and (iv) the Board did not adequately oversee the sales process.
The sales process is described supra Section I.C. The weight of the
evidence discredits Petitioners’ stated concerns. The Court concludes that the sales
process was generally strong and can be expected to have led to a Merger price
indicative of fair value. Accordingly, it deserves weight in the Court’s valuation.
E. The Court’s Determination
Any real-world sales process may be criticized for not adhering completely
to a perfect, theoretical model. Nonetheless, AutoInfo’s process was
comprehensive and nothing in the record suggests that the outcome would have
been a merger price drastically below fair value, as Petitioners’ expert suggests.
Placing heavy weight on the Merger price “is justified in light of the absence of
any other reliable valuation analysis.”166 Not only are other credible valuations
unavailable, but the record also contains evidence corroborating the Merger price’s
reliability. Even Petitioners’ expert agrees that AutoInfo was “shopped quite a bit”
166
Huff, 2013 WL 5878807, at *13.
35
and that the sales process was arm’s length.167 The Merger was the result of “an
adequate process.”168 The Merger price is thus a strong indicator of value.169
Before placing full weight on the Merger price, the Court performed its own
DCF analysis. Having rejected the Management Projections, the Court relied on
financial projections that Comvest had prepared for internal use in evaluating the
AutoInfo deal.170 In a February 25, 2013, Investment Committee Memo, Comvest
projected five-year financials for AutoInfo based on both a base case (the “Base
Case Projections”) and a downside case scenario. Comvest’s projections were
prepared during due diligence to provide more detail than the Management
Projections. They represented Comvest’s then-current belief regarding AutoInfo’s
likely future performance.171 After Comvest’s investment committee requested “a
number of alternative scenarios below the down side case,” a revised downside
case and a “shock case” were also produced.172
When preparing his expert report, Zmijewski considered using the Base
Case Projections in a DCF valuation. While he concluded that those projections
would not yield a reliable indication of fair value, he did use them to conduct a
167
Trial Tr. 221-22 (Puglisi).
168
Trial Tr. 222 (Puglisi).
169
Delaware law does not require that a sales process conform to any theoretical
standard. Huff, 2013 WL 5878807, at *14.
170
See JX 282 (email to Caple attaching Comvest’s presentation to its investment
committee).
171
Trial Tr. 449 (Caple).
172
Trial Tr. 450-51 (Caple).
36
DCF analysis included in his rebuttal report. AutoInfo has argued that Comvest’s
projections are a better forecast of the Company’s future performance as of the
date of the Merger than are the Management Projections.
In his rebuttal expert report, Puglisi analyzed the Comvest Base Case
Projections. He considered them reasonably reliable, observing that
after months of due diligence and hundreds of thousands of dollars
spent, up until days prior to the stockholder vote on the transaction,
Comvest continued to focus its internal investment committee
presentations on its Base case projections, including in its closing
memo, noting the Company’s strong 2013 first quarter results, and
highlighting that the Company had outperformed revenue and gross
margins stated in its Base case projections.173
Because the Base Case Projections are the most reliable forecast in the
record, the Court employed them in its DCF analysis. The Court generally adopted
the DCF framework used by Zmijewski in his rebuttal expert report.174 However,
as explained in Section 3.F below, the record does not support Zmijewski’s
decision to remove $1,449,000 per year (before tax) in purported merger cost
savings. The Court added back that value to arrive at a corrected estimate of
AutoInfo’s forecasted free cash flows. The Court otherwise credited the
173
JX 382 (Puglisi Rebuttal Report) at 10.
174
Despite the gulf between the parties’ fair value estimates, there is little dispute
over the appropriate DCF model. Rather, the parties disagree on whether there are
reliable inputs to run a DCF and the appropriate equity size premium, which
impacts AutoInfo’s cost of equity and thus its weighted average cost of capital.
37
uncontroversial assumptions underlying Zmijewski’s model, as well as his use of
17.57% as AutoInfo’s weighted average cost of capital (“WACC”).
The parties disagree on AutoInfo’s WACC, which is used in a DCF analysis
to discount cash flow projections and a terminal value to estimate the Company’s
enterprise value as of the Merger. Zmijewski used a WACC of 17.57%, while
Puglisi used 11.30%. The difference stems entirely from debate regarding the
appropriate equity size premium to be added to AutoInfo’s cost of equity.175 The
most common method for estimating a company’s cost of equity, and the method
employed by both experts, is application of the capital asset pricing model (the
“CAPM”). Because empirical evidence suggests that the CAPM understates small
companies’ costs of equity, valuation professionals often add a size premium,
based on historically observed data, to a CAPM-derived cost of equity.176
Zmijewski and Puglisi each added a size premium to AutoInfo’s CAPM-based cost
of equity; Zmijewski used 11.65%, and Puglisi selected 3.81%.
Following standard practice, both experts derived the size premium using
data from Ibbotson Associates (“Ibbotson”). The 2013 edition of Ibbotson breaks
down publicly traded stocks into deciles based on market capitalization.177 It
further breaks down the 10th decile, which includes the smallest companies, into
175
See RX-1 (demonstrative exhibit comparing the experts’ WACC calculations).
176
SHANNON P. PRATT & ROGER J. GRABOWSKI, COST OF CAPITAL: APPLICATIONS
AND EXAMPLES 232-61 (John Wiley & Sons, Inc. 4th ed. 2010).
177
JX 201.
38
four subdeciles. Subdecile 10z subsumes the smallest companies in Ibbotson’s
data set.
Puglisi chose the size premium for Ibbotson’s micro-cap category, which
includes the 9th and 10th deciles, i.e., companies with market capitalizations
ranging from $1.139 million to $514.209 million. Zmijewski looked to the 10z
subdecile, which consists of companies with market capitalizations from $1.139
million to $96.164 million. At the time of the Merger, AutoInfo had a market
capitalization of approximately $30 million. AutoInfo thus fell comfortably within
subdecile 10z based on its market capitalization. For several reasons, the Court
relied on the 10z size premium.
First, Puglisi testified that he “would have used [a size premium] close to the
10z category, if not 10z itself,” had he not believed it necessary to strip out a
marketability factor.178 Puglisi’s adjustment to the size premium runs counter to
Delaware law.179 In Gearreald v. Just Care, Inc., this Court “decline[d] to reduce
the Company’s size premium to less than what is implied by its actual size.”180 In
that case, as here, the parties agreed as to which Ibbotson subdecile applied based
on size alone, yet petitioners’ expert used a lesser size premium to “eliminate[e]
178
Puglisi Dep. 156.
179
See Gearreald v. Just Care, Inc., 2012 WL 1569818, at *10-12 (Del. Ch.
Apr. 30, 2012).
180
Id. at *12.
39
the ‘well-documented liquidity effect’ contained within the size premium.” 181 The
Court rejected the adjustment “because the liquidity effect at issue relate[d] to the
Company’s ability to obtain capital at a certain cost, . . . [and was therefore] related
to the Company’s intrinsic value as a going concern and should be included when
calculating its cost of capital.”182 Petitioners attempt to distinguish between a
marketability discount and an illiquidity discount, which may represent distinct
concepts in a separate context. However, AutoInfo’s cost of capital directly affects
transactions between the Company and providers of capital, and is thus part of its
value as a going concern. Because in these circumstances there is an insufficient
factual basis for doing so, the Court declines to depart from the size premium
implied by AutoInfo’s actual size.183
The Court also considered the fact that Stephens, when valuing AutoInfo,
used a size premium and WACC even higher than what Zmijewski recommends.
Stephens believed that AutoInfo would need to significantly increase its market
capitalization to benefit from a lower WACC.184 Perhaps most importantly,
relying on Puglisi’s WACC produces an estimate of fair value completely divorced
181
Id. at *10.
182
Id. at *11.
183
Id. at *12. While Ibbotson no longer publishes 10z size premium data, Duff &
Phelps, LLC has “pick[ed] up the mantle.” Trial Tr. 590 (Zmijewski). Duff &
Phelps is a widely used and well-respected source of size premium data. See Pratt
& Grabowski, supra note 176, at 110.
184
Stephens’s Strategic Initiatives Overview 12.
40
from the negotiated Merger price (and the other bids offered for the Company).
The discrepancy between Puglisi’s estimates and the market’s valuation of
AutoInfo cannot be explained by anything in the record.185
Using a WACC of 17.57% and the Base Case Projections, the Court
performed a DCF analysis that resulted in a fair value determination of
approximately $0.93 per share on the date of the Merger.186 Under Delaware law,
it would be appropriate to provide weight to the value as implied by the Court’s
DCF analysis.187 Nonetheless, because the Merger price appears to be the best
estimate of value, the Court will put full weight on that price.188
F. Must the Merger Price Be Adjusted for Cost Savings?
While the Merger price was the baseline for Zmijewski’s fair value opinion,
he adjusted that amount downward to account for the portion of the price that he
deemed attributable to the consummation or prospect of the Merger.189 In this, as
in any appraisal action, the Court must value Petitioners’ shares “exclusive of any
element of value arising from the accomplishment or expectation of the
185
Cf. Union Ill. 1995 Inv. Ltd. P’ship, 847 A.2d at 359 n.43 (citing to a highly-
regarded corporate finance text for the proposition “that if the DCF analysis you
perform of a stock does not match the market price, you have probably used poor
forecasts”).
186
The Base Case Projections were provided to the Court in native format at
JX 390. The Court used Zmijewski’s basic model, as set forth in his rebuttal
report.
187
See Union Ill. 1995 Inv. Ltd. P’ship, 847 A.2d at 364.
188
Id.189
Zmijewski’s fair value estimate was thus below the Merger price.
41
merger . . . .”190 AutoInfo argues that two categories of cost savings, which
increased the price that Comvest was willing to pay for it, must be backed out of
the Merger price to arrive at AutoInfo’s fair value as a going-concern as of the
Merger date. Those categories are (i) public company costs that Comvest could
eliminate once AutoInfo ceased trading as a public company, and (ii) executive
compensation costs that Comvest planned to eliminate. AutoInfo bears the burden
of showing that adjustments should be made to the Merger price.191
Zmijewski suggests backing out these cost savings because AutoInfo’s
stockholders likely captured 100% of the value created by those savings and, thus,
the value is embedded in the Merger price.192 He cites academic literature that
concludes that target firms capture virtually all of the value created by corporate
combinations through the price paid by the acquirer.193 Because the $1.05 price
would be expected to reflect anticipated cost savings, Zmijewski adjusted the
Merger price downwards to account for Merger-related effects on the stock’s
value.
190
8 Del. C. 262(h).
191
See Huff Fund Inv. P’ship v. CKx, Inc., 2014 WL 2042797, at *2 (Del. Ch.
May 19, 2014), aff’d, -- A.3d --, 2015 WL 631586 (Del. Feb. 12, 2015) (“Huff
Fund”).
192
Zmijewski Opening Report ¶ 98.
193
Id. at ¶ 97.
42
This Court only excludes from an appraisal award value that is merger-
specific.194 An appraisal award does not include “the amount of any value that the
selling company’s shareholders would receive because a buyer intends to operate
the subject company, not as a stand-alone going concern, but as a part of a larger
enterprise, from which synergistic gains can be extracted.”195
Zmijewski based his calculation of cost savings on adjustments that
Comvest made to AutoInfo’s earnings when preparing the Base Case Projections.
Comvest apparently anticipated savings related to public company costs and
executive compensation. It assumed that the savings would not grow over time
and would persist into perpetuity.196
In Huff Fund, the respondent company urged the Court to subtract $0.29
from the merger price to arrive at fair value.197 Its rationale was that prior to the
merger, the acquirer had identified $4.6 million in annual cost savings that it hoped
to realize by converting the target from a publicly held corporation to a privately
held firm.198 The evidence for those anticipated cost savings was an investment
memorandum that the acquirer had prepared. The Court did not need to “reach[]
the theoretical question of under what circumstances cost-savings may constitute
194
Huff Fund, 2014 WL 2042797, at *3.
195
Union Ill. 1995 Inv. Ltd. P’ship, 847 A.2d at 356.
196
Zmijewski Opening Report ¶ 100.
197
Huff Fund, 2014 WL 2042797, at *3.
198
Id.43
synergies excludable from going-concern value under Section 262(h)” because the
record did not establish that the acquirer had based its bid on cost savings that the
target could not have itself realized had it continued as a going concern.199
Accepting Zmijewski’s adjustments would appear to require the Court to
reduce for cost savings the fair value established in an appraisal proceeding
through reliance on the transaction price. Allowing a near automatic reduction in
price would reverse the burden that is on the party arguing that adjustments are
warranted. Zmijewski derived his cost savings figures from three lines of data
included in Comvest’s development of its Base Case Projections.200 The Court
does not know how Comvest arrived at its numbers or even what it included as
“public company costs.” Unlike the Merger price, which was corroborated by a
thorough and public sales process, the reliability of the purported cost savings has
not been tested.201 AutoInfo has thus failed to establish that any downward
adjustment to the Merger price is warranted.202
199
Id.200
Zmijewski Opening Report ¶ 100. AutoInfo cites to one other one-page
document that purports to show Comvest’s plan to save on executive
compensation. See JX 348. No context for that document was provided and
Zmijewski did not rely on it in calculating cost savings.
201
Because AutoInfo has failed to provide adequate evidence to support its
adjustments to the Merger price, the Court need not reach the issue of whether
similar cost savings would be excluded from fair value in another context.
202
Further, AutoInfo has not established that the executive compensation cost
savings, which represent the bulk of Zmijewski’s adjustments, could only have
been realized through accomplishment of a merger. The Special Committee
44
IV. CONCLUSION
Where, as here, the market prices a company as the result of a competitive
and fair auction, “the use of alternative valuation techniques like a DCF analysis is
necessarily a second-best method to derive value.”203 The result of a DCF analysis
depends critically on its inputs. For example, small changes to the assumed cost of
capital can dramatically impact the result.
AutoInfo’s expert, a tenured professor at the University of Chicago Booth
School of Business, concluded that there is no reliable data to input into a DCF or
comparable companies model. He determined that the process by which AutoInfo
was marketed and sold would be expected to have led to a price indicative of the
fair value of the Company’s stock. The Court has independently reached these
same conclusions.
For the reasons set forth above, the fair value of one share of AutoInfo at the
time of the Merger was $1.05. Petitioners are entitled to interest at the legal rate.
Counsel are requested to confer and to submit an implementing form of order.
expected that if the Comvest deal fell through, the Board would push through
Management-related changes in the hope of increasing share price. See, e.g.,
JX 277 (Patterson email to other Special Committee members).
203
Union Ill. 1995 Inv. Ltd. P’ship, 847 A.2d at 359.
45