William Richard Kruse v. Synapse Wireless, Inc.

IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE
WILLIAM RICHARD KRUSE, )
individually and as TRUSTEE for THE )
VIVIAN CALVERT KRUSE LIVING )
TRUST and THE WILLIAM )
RICHARD KRUSE LIVING TRUST, )
)
Petitioners, )
)
v. ) C.A. No. 12392-VCS
)
SYNAPSE WIRELESS, INC., )
)
Respondent. )
MEMORANDUM OPINION
Date Submitted: April 22, 2020
Date Decided: July 14, 2020
Evan O. Williford, Esquire of The Williford Firm LLC, Wilmington, Delaware,
Attorney for Petitioner.
Richard P. Rollo, Esquire of Richards, Layton & Finger, P.A., Wilmington,
Delaware and Michael D. Mulvaney, Esquire, J. Ethan McDaniel, Esquire and
James C. Lester, Esquire of Maynard, Cooper & Gale P.C., Birmingham, Alabama,
Attorneys for Respondent.
SLIGHTS, Vice Chancellor
To follow is the Court’s fair value appraisal of Respondent, Synapse Wireless,
Inc. (“Synapse” or the “Company”), per 8 Del. C. § 262. Synapse is an internet of
things (or “IoT”) company that sells hardware and software to industrial clients.1
The IoT industry has experienced tremendous growth in recent years, particularly
within the industrial machinery space. Sensing that Synapse was well positioned to
take advantage of this market opportunity, McWane Inc. (“McWane”), a large,
traditional manufacturer, gained control of Synapse through a merger in 2012
(the “2012 Merger”).
In contrast to McWane’s high hopes, Synapse did not grow as expected
following the 2012 Merger. It repeatedly missed its management’s financial targets,
usually by wide margins. As Synapse hemorrhaged cash, McWane propped up its
subsidiary through loans and equity purchases at the price set by the 2012 Merger.
In 2016, McWane decided to buyout the remaining minority shareholders and make
Synapse a wholly owned subsidiary (the “2016 Merger”). In connection with that
transaction, McWane offered the remaining Synapse stockholders $0.42899 per
1
The Internet of Things “refers to a network comprised of physical objects capable of
gathering and sharing electronic information. The Internet of Things includes a wide
variety of smart devices, from industrial machines that transmit data about the production
process to sensors that track information about the human body.” Will Keaton,
The Internet of Things (IoT), INVESTOPEDIA (Feb. 9, 2020)
https://www.investopedia.com/terms/i/internet-things.asp (last visited July 6, 2020).
1
share, and all but one Synapse stockholder accepted. That one stockholder,
Petitioner, William Richard Kruse (“Kruse”), seeks appraisal of his Synapse stock
under 8 Del. C. § 262.
While Delaware courts typically look for market-based evidence of fair value
in appraisal proceedings, there is no contemporaneous market evidence available
here with respect to the 2016 Merger. It is undisputed that there was no market-
check or competitive sales process for Synapse leading up to that transaction.
Instead, a controlling shareholder of a private company bought out the minority
stockholders at a price set by a previously bargained-for settlement agreement. With
neither party arguing this Court should defer to the deal price, Kruse and Synapse
both have relied on expert witnesses to value Kruse’s Synapse shares.
The experts used similar valuation techniques: each presented valuations
based on discounted cash flow (“DCF”) models, comparable transactions and
McWane’s prior purchases of Synapse’s stock (the only market-based evidence
presented). And, in doing so, the experts materially agreed on several important
inputs in their valuation models. Nevertheless, as has become standard fare for
appraisal litigation,2 the experts reached monumentally different valuations. Kruse’s
2
See, e.g., Golden Telecom, Inc. v. Global GT LP, 11 A.3d 214, 218 (Del. 2010) (“[I]t is
difficult for . . . Vice Chancellors to assess wildly divergent expert opinions regarding
value.”); In re Appraisal of Jarden Corp., 2019 WL 3244085, at *1 (Del. Ch. July 19,
2019) (observing that well credentialed experts were “miles apart”); Gonsalves v. Straight
Arrow Publ’rs, Inc., 1996 WL 696936, at *1 (Del. Ch. Nov. 27, 1996) (“Gonsalves I”),
2
expert opined that each Synapse share was worth $4.1876 at the time of the 2016
Merger. Synapse’s expert declined to give a single valuation, but his summary
values ranged from $0.06 to $0.11 per share.
When dueling experts proffer wildly divergent valuations, the resulting trial
dynamic presents difficult and, frankly, frustrating challenges for the judicial
appraiser. This case presents another, more fundamental challenge; after carefully
reviewing the evidence, it is difficult to discern any wholly reliable indicators of
Synapse’s fair value. There is no reliable market evidence, the comparable
transactions analyses both experts utilized—a dicey valuation method in the best of
circumstances—have significant flaws and the management projections relied upon
by both experts in their DCF valuations are difficult to reconcile with Synapse’s
operative reality.
In the typical litigation context, the lack of fully reliable evidence might lead
the factfinder to conclude that neither party carried their burden of proof and neither
party, therefore, is entitled to a verdict. But “no” is not an answer in the unique
world of statutory appraisal litigation. If the parties fall short in their respective
burdens, the court must still reach an answer—a fair value appraisal must still be
rev’d, 701 A.2d 357 (Del. 1997) (“Gonsalves II”) (stating it is “rather a typical appraisal
trial” when experts advance “absurdly differing values”).
3
provided.3 In this case, one expert credibly made the best of less than perfect data
to reach a proportionately reliable conclusion, while the other did not. In such
circumstances, this court is free to adopt, in part or in whole, the more credible
valuation. 4 This is especially so when the court is satisfied that it can do no better
on its own.
After carefully reviewing the evidence, I find that Synapse has marshalled
sufficient evidence to carry its burden of proving a reliable appraisal of Synapse’s
fair value as of the 2016 Merger. Accordingly, with two minor adjustments, I adopt
one of the discounted cash flow valuations proffered by Synapse and appraise the
fair value of Synapse’s equity as of the 2016 Merger at $20,347,822, or $0.228 per
share. Kruse’s 582,216 shares, therefore, are appraised at a fair value of
$133,015.09.
3
M.G. Bancorporation, Inc. v. LeBeau, 737 A.2d 513, 526 (Del. 1999).
4
See id.; Cede & Co. v. Technicolor, Inc., 884 A.2d 26, 35–36 (Del. 2005) (“Cede III”)
(“It is often the case in statutory appraisal proceedings that a valuation dispute becomes a
battle of experts. This is evidenced by the fact that the Court of Chancery is frequently
presented with conflicting expert testimony. The Court of Chancery, as the finder of fact
in an appraisal case, enjoys the unique opportunity to examine the record and assess the
demeanor and credibility of witnesses. Thus, the Court of Chancery is the sole judge of
the credibility of live witness testimony. This Court will accept the Court of Chancery's
factual determinations if they turn on a question of credibility and the acceptance or
rejection of particular pieces of testimony. A factual finding made by the Court of
Chancery based on a weighing of expert opinion may be overturned only if arbitrary or
lacking evidential support.”) (quotations omitted).
4
I. BACKGROUND
The following facts were proven by a preponderance of the evidence after a three-
day trial.5
A. Parties and Relevant Non-Parties
Respondent, Synapse, is a Delaware corporation that manufactures products
in the “Internet of Things” and “Smart Manufacturing” spaces.6 Synapse has several
business lines. In its “Core” or “Legacy Core” business, the Company provides
customers a small hardware module with an operating system, a radio and the ability
to accept instructions in the Python programming language, allowing customers to
build and connect their own applications for the technology.7 The “Lighting”
business sells special purpose controls for LED lighting, and configuring networks
for those controls.8 Synapse’s “Smart Manufacturing” business provides products
as services to manufacturing companies, which allows Synapse to obtain recurring
revenue by offering continuing configuration and computing services.9
5
I cite to the trial record as “Tr.__ (name)”, the Joint Pre-Trial Stipulation and Order as
“PTO ¶ __”, the joint trial exhibits as “JX__” and Depositions as “(Name) Dep.__.”
6
JX 305 (“Myers Dep.”) 29:22–30:10.
7
Myers Dep. 25:9–27:25.
8
Myers Dep. 28:1–29:13.
9
Myers Dep. 30:3–31:13.
5
Non-party, McWane, is a major national manufacturer of products including
“cast iron, drainpipe fittings, couplings, ductile pipe, fire hydrants, underground
valves, fittings like the bends and Ts that connect the water underground, [and]
propane tanks, like [] gas grill tanks.”10 McWane is a private company and remains
under the control of its founders, the McWane family. 11
Petitioner, William Richard Kruse, is a former aerospace engineer who has
invested in numerous startup companies, both as an individual and through the
Huntsville Angel Network, an investor group.12 Kruse first invested in Synapse in
2009 after the company pitched the Huntsville Angel Network, and he invested
additional money in 2009 and 2010 through his IRA, his trust and his wife’s trust.13
Kruse, individually, and as trustee for the Vivian Calvert Kruse Living Trust and the
William Richard Kruse Living Trust, held 582,216 shares of Synapse before the
2016 Merger.14
10
JX 301 (“Page Dep.”) 18:15–19:7.
11
Tr. 22:21–22:2 (Page); Tr. 61:15–22 (Page).
12
Tr. 128:1–131:4 (Kruse).
13
Tr. 133:4–134:3 (Kruse).
14
PTO ¶ 7.
6
B. The 2012 Merger
On May 23, 2012, Synapse entered into an Agreement and Plan of
Reorganization with McWane. 15 The Agreement and Plan of Reorganization
contained two key parts: the 2012 Merger between Synapse and McWane, which
gave McWane a controlling stake in Synapse, and a Stockholders Agreement
(the “2012 Stockholders Agreement”) between Synapse, McWane, the designated
stockholder representative and consenting Synapse stockholders. 16
In the 2012 Merger, McWane acquired Synapse stock at a per-share price of
$4.997, implying an enterprise value of $109 million.17 The 2012 Merger
Agreement contains certain Company representations, warranties and covenants,
and an agreement to indemnify the purchasers against losses arising out of or
connected to breaches of those representations, warranties and covenants.18
The parties established an $8 million escrow account as partial security for the
15
JX 52 (the “2012 Merger Agreement”).
16
PTO ¶¶ 3–4.
17
Id. Synapse effectuated a 3:1 stock split on September 30, 2014, implying a pre-merger
price per share of $1.667. PTO ¶ 6.
18
2012 Merger Agreement §§ 3.1–3.28, 8.2(a).
7
sellers’ indemnification obligations.19 Moro Lanier, III was named as the
Stockholder Representative. 20
The 2012 Stockholders Agreement contains a provision giving the minority
stockholders the right, beginning in 2016, to demand that McWane purchase their
outstanding Synapse stock based on Synapse’s then-current valuation—subject to a
valuation floor of $76.3 million.21 That agreement also gave McWane the right,
beginning in 2018, to require the remaining minority stockholders to sell their
Synapse stock to McWane. 22 Important here, the agreement further provided
McWane with the right, expiring in 2016, to purchase any newly issued Synapse
shares at a price per-share based on the 2012 Merger.23 As discussed below,
McWane would go on to exercise this option on numerous occasions.
19
2012 Merger Agreement § 8.4, 6.
20
2012 Merger Agreement § 8.6. Kruse has criticized the selection of Mr. Lanier as
stockholder representative, suggesting that his relationship with Ruffner Page, McWane’s
President, made Mr. Lanier an inappropriate choice to represent Synapse stockholders.
See, e.g., Pet’r’s Post-Trial Opening Br. (“Post-Trial OB”) 8–9, 58. After reviewing the
evidence of the Page/Lanier relationship, I find the argument unpersuasive.
21
PTO ¶ 10.
22
Id.
23
PTO ¶ 11.
8
C. Synapse Continuously Misses its Revenue Targets
In the years following the 2012 Merger, Synapse repeatedly missed its
revenue targets, at times by wide margins. For instance, Synapse’s 2013 revenue
target was $11.78 million;24 its actual 2013 revenue was $2.154 million. 25 Synapse’s
2014 revenue target was $7.1 million; 26 its actual 2014 revenue was $3.025 million.27
Synapse’s 2015 revenue target was $10.14 million; 28 its actual 2015 revenue was
$2.33 million. 29 Synapse’s 2016 revenue target was $11.91 million; 30 its actual 2016
revenue was $3.96 million. 31 And so it went; suffice it to say, Synapse’s
management were not proficient forecasters. Indeed, in 2015, Synapse missed its
revenue projections for every month, even when accounting for monthly downward
revisions. 32 The chart below illustrates the forecast misses:
24
JX 76 at 2.
25
Id.
26
JX 98 at 2.
27
Id.
28
JX 197 at 14.
29
Id.
30
JX 277 at 13.
31
Id.
32
Tr. 179:4–180:11 (Reinhardt); JX 197 at 22.
9
33
Synapse’s disappointing post-2012 Merger performance stands in stark
contrast to the financial projections used to calculate the 2012 Merger price.
As depicted in the chart below, before that merger, Synapse’s projected revenues
were $43.6 million for 2013, $106.18 million for 2014, $196.43 million for 2015,
and $324.11 million for 2016. 34 Cumulatively, Synapse’s actual revenues missed
these projections by hundreds of millions of dollars.35
33
Respondent’s Demonstrative 2.
34
JX 25 at 7.
35
Compare id. with JX 76 at 2; JX 98 at 2; JX 197 at 14; JX 277 at 13.
10
36
Synapse developed six distinct business lines after the 2012 Merger, but by 2015, it
had already shut down three of them. 37
Kruse waves away this dismal history and makes much of Synapse’s 2016
business plan. He argues it represented a turning point for the Company that “put it
on an accelerated path to success,” using this argument to justify a higher valuation.38
According to Kruse, the shuttering of business units meant, “Synapse was retooled
to be a more profitable vertical business.”39 This supposed “retooling,” however,
36
Respondent’s Demonstrative 1.
37
JX 170 at 5; JX 303 (“Reinhardt Dep.”) 217:4–13.
38
Petitioner’s Pre-Trial Opening Br. 9.
39
Id.
11
did nothing to improve Synapse’s fortunes. The Company’s revenue was only
$5.7 million in 2017 and $7.5 million in 2018, orders of magnitude lower than the
lofty revenue projections used to justify the 2012 Merger price. 40
D. McWane Provides Synapse with Additional Funds
With Synapse repeatedly missing its revenue targets and being miles away
from profitability, the Company relied on McWane to finance its operations.41
To fund its subsidiary, McWane purchased newly-issued Synapse stock eight times
after the 2012 Merger, eventually acquiring over ten million newly issued Synapse
shares at the contractually mandated pre-split price of $4.997 per share.42
The majority of these shares were purchased in one block in January 2014, when
McWane paid $31 million for 6,203,660 shares of Synapse.43 In addition to new
stock issuances, McWane also acquired nearly a million existing shares from
Synapse stockholders at the contractually pre-determined price.44
40
Tr. 196:19–21 (Reinhardt); Tr. 123:3–9 (Page). While these post-merger numbers
cannot impact my appraisal of Synapse’s fair value as of the 2016 Merger, they are useful
to keep in mind when assessing the fanciful nature of Synapse’s management’s financial
targets.
41
Tr. 186:10–192:15 (Reinhardt).
42
JX 256.
43
Tr. 186:21–23 (Reinhardt).
44
JX 256. This excludes purchases associated with the 2016 Merger and Lanier Settlement
Agreement, both discussed later.
12
Shortly after the January 2014 equity purchase, Synapse loaned the entire
$31 million back to McWane. 45 This transaction structure allowed McWane to fund
Synapse’s operations while keeping Synapse on a short leash—each month Synapse
requested McWane repay a portion of the $31 million according to the Company’s
current financial needs, and each month McWane repaid the requested amount.46
Importantly, by purchasing the shares up front, McWane increased its ownership
interest in Synapse beyond 80%. This permitted McWane to consolidate Synapse’s
and McWane’s federal tax returns, allowing it to employ Synapse’s prior tax losses
fully for its own benefit.47
The parties dispute what happened next. The $31 million was paid back and
spent by November or December of 2014, after which McWane began to loan money
(the “Disputed Money”) to Synapse directly. 48 The first loan occurred on
December 3, 2014, for $870,000.49 By January 11, 2016, the balance on the loan
45
Tr. 187:13–188:1 (Reinhardt).
46
Tr. 187:16–188:4 (Reinhardt).
47
Tr. 371:22–372:22 (Petty). These tax benefits would eventually exceed the $31 million
McWane paid for these shares. Tr. 373:3–11.
48
Tr. 188:9–21 (Reinhardt).
49
JX 349 (“Sweet Opening Report”) 15.
13
totaled $29,343,000.50 The loan carried a rock-bottom interest rate of 1.16%, and
while Synapse appears to have made some interest payments, it never generated
enough free cash to pay back principal on the loan. 51
The parties dispute whether this money was truly a loan from McWane to
Synapse, or was, instead, a capital contribution.52 Synapse and McWane describe
the money as a loan, support that characterization with loan documentation and
identify “interest payable” and “intercompany notes payable” attributable to the loan
on Synapse’s balance sheet.53 Kruse responds that Synapse never made any
principal payments, McWane intended to forgive the loan and, therefore, the
Disputed Money should be treated as contributed capital.54
After reviewing the evidence, I am satisfied the Disputed Money is debt.
Synapse’s executives credibly testified to that fact, and the Company was able to
support this claim with contemporaneous, original loan documentation.55 Although
Synapse had not repaid any principle as of the 2016 Merger, Synapse did appear to
50
Sweet Opening Report 15–16; see JX 197 at 11 (noting a $27,747,500 balance on
December 31, 2015).
51
Tr. 810:5–16 (Noe); Tr. 389:4–15 (Petty). Sweet Opening Report 15–16.
52
See Sweet Opening Report 16; JX 358 (“Noe Rebuttal Report”) ¶ 14.
53
Noe Rebuttal Report ¶¶ 14, 14 n.1; JX 197 at 11; JX 217; JX 252.
54
Sweet Opening Report 15–16; JX 207.
55
Tr. 182:13–22 (Reinhardt); Tr. 183:11–16 (Reinhardt); JX 197 at 11; JX 217; JX 252.
14
have made interest payments and McWane has never forgiven this debt.56 I address
the implications of this finding below.
E. The Lanier Action
On December 13, 2013, McWane filed a claim certificate related to the 2012
Merger in which it alleged it had suffered losses due to Sellers’ breaches and
misrepresentations exceeding the $8 million escrow account.57 The parties
commenced negotiations to resolve the claim, but were unable to do so.58
On March 31, 2014, McWane filed a complaint in this court (the “Lanier Action”)
to recover its purported losses.59
The Lanier Action focused on alleged “channel stuffing,” a practice in which
a company intentionally sells more inventory to retailers and distributors than it
expects those intermediaries will ultimately sell to consumers. 60 This practice allows
56
Tr. 188:22–190:9 (Reinhardt).
57
PTO ¶ 12.
58
JX 421 (the “Lanier Action”) 6.
59
Id.
60
See generally, Tracy Byrnes, Too Many Thin Mints: Spotting the Practice of ‘Channel
Stuffing’, WALL ST. J. (Feb. 7, 2002)
https://www.wsj.com/articles/SB1013117089572671360; Tr. 263:3–8 (Foster).
15
a company to recognize revenue earlier than it otherwise would. 61 Channel stuffing
often requires the company to extend discounts or other incentives to the
intermediaries, lowering the ultimate price the company receives for its products.62
Additionally, intermediaries typically have the right to return unsold inventory,
potentially rendering the sales boost illusory. 63 Channel stuffing is only possible
when a company uses a “sales-in” method of revenue recognition (which allows the
seller to recognize revenue when the product goes to a distributor), rather than a
“sales-through” method (which requires the seller to wait to recognize revenue until
the distributor sells to the ultimate consumer). 64
The complaint in the Lanier Action alleges, “Synapse improperly recognized
revenue from sales to distributors prior to the time of the Merger, resulting in inflated
and misstated revenue numbers.” 65 The parties spent considerable time at trial
presenting evidence related to purported channel stuffing before the 2012 Merger
and the propriety of Synapse’s revenue recognition and accounting methods.
61
There may be legitimate reasons for a company to recognize revenue in an earlier sales
period. The practice is illegitimate, however, when used to create a misleading impression
of current sales. See id.
62
Id.
63
Tr. 396:24–398:2 (Petty).
64
Tr. 392:23–395:6 (Petty).
65
Lanier Action 12.
16
The parties settled the Lanier Action in December 2015, prior to the court’s
verdict. According to the settlement documents (collectively, the “Lanier Settlement
Agreement”), McWane received the following: (1) $4.65 million from the escrow
account; (2) a reduction in the price of McWane’s call option to a post-split price of
$0.42899 per share; and (3) an acceleration of McWane’s call option to make it
immediately exercisable.66 After the Lanier Settlement Agreement was executed,
McWane exercised its accelerated call option, giving it beneficial ownership of
99.346% of the outstanding Synapse stock. 67 Kruse was the only stockholder who
refused to sell his shares pursuant to this agreement.68
F. The 2016 Merger
On February 2, 2016, Synapse and McWane effectuated a squeeze-out
merger, and offered Kruse the Lanier Settlement price of $0.42899 per share for his
stock. 69 Instead of accepting the 2016 Merger price, Kruse brought this action for
appraisal.
66
JX 199 (the “Lanier Settlement Agreement”).
67
PTO ¶ 16.
68
PTO ¶ 18.
69
PTO ¶ 19.
17
G. The Experts
Both Kruse and Synapse retained experts who opined on the fair value of
Synapse as of the 2016 Merger. Kruse retained Athen Sweet (“Sweet”), whose
reports I will refer to as the “Sweet Opening Report” and the “Sweet Rebuttal
Report.”70 Synapse retained Christopher Noe (“Noe”), whose reports I will refer to
as the “Noe Opening Report” and the “Noe Rebuttal Report.”71
Sweet applied three valuation techniques to reach his fair value conclusion.
First, he performed a “Prior Company Transactions” analysis that yielded an
enterprise value for Synapse of $386,622,000.72 Next, he performed a DCF analysis,
resulting in an enterprise value of $331,973,000.73 Finally, he performed a
70
Athen Sweet is the founder and CEO of Innovatus IQ, LLC, a merger and acquisition
advisory firm. Sweet Opening Report Appx. A. He previously worked as a Certified
Public Accountant at Faulk & Winkler, LLC, an accounting firm. Id. Mr. Sweet received
his B.S. in Accounting and M.B.A. from Southeastern Louisiana University, where he
currently serves as Professional Chairman of the school’s Accounting Advisory
Committee. Id.
71
Christopher Noe, Ph.D is a Senior Lecturer at the Massachusetts Institute of
Technology’s Sloan School of Management, where he teaches classes on accounting,
financial statement analysis and valuation. JX 319 (“Noe Opening Report”) ¶ 1; Noe
Opening Report Ex. 1. He previously worked as a consultant at Charles River Associates,
an economics, finance and business consulting firm, and as an Assistant Professor at
Harvard Business School. Id. Dr. Noe received his Ph.D in Accounting and his M.S. in
Applied Economics from the William E. Simon Graduate School of Business
Administration at the University of Rochester and his B.A. in Economics from Emory
University. Id.
72
Sweet Opening Report 17; Tr. 582:18–583:16 (Sweet).
73
Sweet Opening Report 17; Tr. 595:8–596:2 (Sweet).
18
“Guideline Transactions (Private)” analysis (or “Comparable Transactions”
analysis), which reached an enterprise value of $190,356,000.74 Ultimately, and for
reasons not entirely clear, Sweet relied 75% on his Prior Company Transactions
analysis, 25% on his Discounted Cash Flow analysis, and 0% on his Comparable
Transactions analysis, to land on a final valuation of $372,960,000, and a per-share
fair value of $4.1876. 75
Noe utilized the same three valuation techniques as Sweet, but reached very
different conclusions. He first performed DCF valuations utilizing two different sets
of assumptions, leading to enterprise values in one model of $48.9 million or $0.24
per share,76 and in the other model a value of negative $3.4 million or $0 per share.77
Noe then employed a “Comparable Transactions” analysis, which resulted in a value
of $0.01 per share.78 Last, Noe performed a Prior Company Transactions analysis
that produced a value of $0.19 per share.79 Ultimately, Noe declined to provide a
single valuation for Synapse stock. Instead, he offered two possibilities. If the 2012
74
Sweet Opening Report 17; Tr. 632:6–20 (Sweet).
75
Sweet Opening Report 17.
76
Noe Opening Report ¶ 13; Tr. 763:18–764:1 (Noe).
77
Noe Opening Report ¶ 16; Tr. 770:19–777:7 (Noe).
78
Noe Opening Report ¶ 21; Tr. 779:5–783:9 (Noe).
79
Noe Opening Report ¶ 24.
19
Merger price was unreliable, he applied equal weight to his DCF and Comparable
Transactions analyses to reach a fair value share price of $0.06. 80 If the 2012 Merger
was a reliable indication of value, he weighed all three of his techniques equally and
valued Synapse at $0.11 per share. 81
H. Procedural History
Kruse filed this action on May 31, 2016.82 This Court held a three-day trial
from February 11, 2019 to February 13, 2019, and post-trial oral argument was held
on September 19, 2019. After reassignment of this case to me, 83 I certified my
familiarity with the case and trial evidence (including a video recording of the trial
testimony) under Court of Chancery Rule 63 on April 16, 2020, and deemed the
matter submitted for decision on April 20, 2020. 84
II. ANALYSIS
Kruse seeks appraisal of his Synapse shares under 8 Del. C. § 262. That
statute entitles shareholders who satisfy certain form and manner requirements to
“an appraisal by the Court of Chancery of the fair value of the stockholder’s shares
80
Noe Opening Report ¶ 25; Tr. 738:7–15 (Noe).
81
Id.
82
JX 246.
83
D.I. 170.
84
D.I. 176.
20
of stock . . . .” 85 The parties do not dispute that Kruse continuously held shares or
that he has properly perfected his appraisal rights under Section 262. Instead, as
usual, the dispute centers on the proper determination of Synapse’s fair value.
A. The Statutory Appraisal Remedy
“An action seeking appraisal is intended to provide shareholders who dissent
from a merger, on the basis of the inadequacy of the offering price, with a judicial
determination of the fair value of their shares.”86 “‘The underlying assumption in
an appraisal valuation is that the dissenting shareholders would be willing to
maintain their investment position had the merger not occurred.’ Accordingly, the
corporation must be valued as a going concern based upon the ‘operative reality’ of
the company as of the time of the merger.” 87 “[T]he purpose of an appraisal is . . .
to make sure that [shareholders] receive fair compensation for their shares in the
sense that it reflects what they deserve to receive based on what would fairly be
given to them in an arm’s-length transaction.” 88 “Since every company is different,
85
8 Del. C. § 262.
86
Cavalier Oil Corp. v. Harnett, 564 A.2d 1137, 1142 (Del. 1989).
87
M.G. Bancorporation, 737 A.2d at 525 (quoting Cede & Co. v. Technicolor, Inc.,
684 A.2d 289, 298 (Del. 1996) (“Cede II”)).
88
DFC Global Corp. v. Muirfield Value P’rs, L.P., 172 A.3d 346, 370–71 (Del. 2017).
21
and every merger is different, the appraisal endeavor is by design, a flexible
process.”89
The Appraisal Statute is a “[b]roth of many cooks and opaque of intent.”90
In this spirit, the statute eschews specific guidance in favor of a more cryptic
mandate that this court consider “all relevant factors” as it determines the fair value
of a dissenting stockholder’s shares. 91 As I undertake my independent appraisal of
Synapse, I may consider “proof of value by any techniques or methods which are
generally considered acceptable in the financial community and otherwise
admissible in court. . . .” 92 Because “corporate finance is not law,” expert witnesses
often play the lead role in sponsoring competing appraisals to the court.93
89
In re Appraisal of Jarden Corp., 2019 WL 3244085, at *23 (Del. Ch. July 19, 2019)
(quotations omitted), aff’d sub nom. 2020 WL 3885166 (Del. July 9, 2020).
90
In re AOL Inc., 2018 WL 1037450, at *1 (Del. Ch. Feb. 23, 2018).
91
8 Del. C. § 262(h). “Relevant factors” do not include value deriving from the transaction
itself. “The court should exclude ‘any synergies or other value expected from the merger
giving rise to the appraisal proceeding.’ ‘[O]nce the total standalone value is determined,
the court awards each petitioning stockholder his pro rata portion of this total . . . plus
interest.’” Verition P’rs Master Fund Ltd. v. Aruba Networks, Inc., 2018 WL 922139,
at *23 (Del. Ch. Feb. 15, 2018) (rev’d on different grounds, 2019 WL 1614026(Del. Apr. 16, 2019)) (quoting Global GT LP v. Golden Telecom, Inc., 993 A.2d 497, 507
(Del. Ch. 2010) (aff’d, 11 A.3d 214 (Del. 2010)).
92
Weinberger v. UOP, Inc., 457 A.2d 701, 713 (Del. 1983). See also Laidler v. Hesco
Bastion Envtl., Inc., 2014 WL 1877536, at *6 (Del. Ch. May 12, 2014) (discussing the
parties’ competing burdens of proof and the court’s obligation independently to appraise
the target company).
93
In re Jarden, 2019 WL 3244085, at *1.
22
When it comes to evaluating expert testimony, appraisal cases are no different
from any other adversarial proceeding. The court’s fact-finding role allows it to
consider reliable and credible expert testimony and to reject unreliable or incredible
expert testimony. 94 In this regard, the court “has the discretion to select one of the
parties’ valuation models as its general framework or to fashion its own.”95 And,
“although not required to do so, it is entirely proper for the Court of Chancery to
adopt any one expert’s model, methodology, and mathematical calculations, in toto,
if that valuation is supported by credible evidence and withstands a critical judicial
analysis on the record.” 96
As mentioned, Kruse and Synapse’s expert witnesses both relied on three
valuation techniques. The first technique links the value of Synapse to the price
McWane paid for Synapse shares in prior transactions. I refer to this approach as
the Prior Company Transactions model. The second is a Comparable Transactions
analysis, which looks to other transactions within a similar timeframe, industry and
company size, and uses financial metrics from those transactions to estimate
Synapse’s value. The third technique is a DCF, which builds a valuation from the
94
M.G. Bancorporation, 737 A.2d at 526.
95
Id. at 525–26.
96
Id. at 526.
23
ground up using projections of future cash flows, and then discounting those
projections to present value. I address each approach in turn.
B. Market Evidence of Fair Value
Both experts performed valuations based on McWane’s previous purchases of
Synapse stock, though they differ on how much weight to give these valuations in
the final determination of Synapse’s fair value. As these purchases, and in particular
the 2012 Merger, are the transactions most likely to provide the Court with market-
based evidence of Synapse’s value, they must be examined first. 97
Recent decisions of our Supreme Court have emphasized the important role
market evidence plays in valuing a company. 98 Vice Chancellor Glasscock
concisely summarized this focus noting, “[w]here [] transaction price represents an
unhindered, informed, and competitive market valuation,” and the “terms of the
transaction are not structurally prohibitive or unduly limiting to [] market
participation,” then “the trial judge must give particular and serious consideration to
transaction price as evidence of fair value.” 99 While the cases emphasizing the
97
See Dell, Inc. v. Magnetar Global Event Driven Master Fund Ltd., 177 A.3d 1, 35
(Del. 2017) (holding that the price of transactions occurring in an efficient market are to
be given “heavy weight” in appraisal cases).
98
See id.; DFC, 172 A.3d at 366.
99
In re AOL, 2018 WL 1037450, at *1 (citations omitted); see also DFC, 172 A.3d at 367
(holding “the most reliable evidence of value is that produced by a competitive market, so
long as interested buyers are given a fair opportunity to price and bid on the something in
question”); Aruba, 2019 WL 1614026, at *5 (“DFC and Dell recognized that when a public
24
importance of market evidence focus on the transaction that caused a dissenting
shareholder to sue for appraisal, purchases of company stock predating the event
triggering appraisal could potentially provide market evidence of fair value if those
transactions were “Dell Compliant.” 100
1. McWane’s Stock Purchases and the 2016 Merger
Kruse argues that McWane’s purchases of Synapse stock preceding the 2016
Merger provide important insight into Synapse’s fair value because McWane
voluntarily bought Synapse stock with full information about the Company’s
struggling business prospects. 101 Synapse disagrees, arguing McWane made these
purchases at a contractually mandated price to provide Synapse with desperately
needed capital. 102 Stock purchases in such circumstances, Synapse argues, provide
little to no reliable evidence of Synapse’s value.
Neither McWane’s series of stock purchases after the 2012 Merger nor the
2016 Merger occurred in a competitive market. These purchases involved no robust
market check, and each occurred in the absence of unhindered, informed and
company with a deep trading market is sold at a substantial premium to the
preannouncement price, after a process in which interested buyers all had a fair and viable
opportunity to bid, the deal price is a strong indicator of fair value. . . .”).
100
In re AOL, 2018 WL 1037450, at *1.
101
Post-Trial OB 31.
102
Tr. 374:5–22 (Petty).
25
competitive market conditions. 103 Information was not widely disseminated to other
potential buyers, and there is no evidence other willing purchasers of Synapse stock
existed, let alone bid for the stock. 104 In each of these transactions, McWane was
the dominant controlling stockholder making purchases at a price set not by market
conditions or independent appraisers, but by a years-old contract. 105 And, the largest
of these capital investments—the $31 million January 2014 purchase—was
motivated, in large part, by McWane’s desire to keep Synapse afloat and to increase
its ownership of Synapse over the 80% threshold so that it could reap the attendant
tax benefits.106 As such, I do not find these purchases to be reliable or relevant
indicators of Synapse’s fair value.
2. The 2012 Merger
While the 2012 Merger was an arm’s-length transaction between a willing
buyer and seller, I likewise find it is not probative of Synapse’s value as of the 2016
Merger. 107 The 2012 Merger was nearly four-years-old at the time of the 2016
103
Tr. 366:20–369:19 (Petty).
104
See Tr. 368:11–16 (Petty); Tr. 369:14–19 (Petty); Tr. 387:1–7 (Petty).
105
JX 61 at 1.
106
Tr. 371:22–373:16 (Petty); JX 443 at 2. The tax savings from this transaction eventually
exceeded the $31 million paid for the shares. Tr. 373:3–11 (Petty).
107
See DFC, 172 A.3d at 370–71 (noting the purpose of appraisal is to give a shareholder
the price they would fairly receive in an arms-length transaction).
26
Merger. It was, in a word, stale as of the 2016 Merger. Moreover, Synapse faced
dramatically different prospects in 2016 than it did in 2012. 108 In the years since the
2012 Merger, the Company had demonstrated a serial inability to meet even
conservative financial targets. 109 Of particular consequence, its 2015 revenue was
nearly 84 times lower than the projection used to calculate the 2012 Merger price—
a miss of nearly $200 million. 110
The parties argued at great length about whether the 2012 Merger involved
fraud. 111 That issue was front and center in the Lanier Action, but that case is not
before me. Instead, I am tasked with deciding whether the 2012 Merger is evidence
of Synapse’s value as of the 2016 Merger. After reviewing the evidence and
arguments of counsel, I am satisfied that the 2012 Merger was either the product of
Synapse’s officers’ misleading inflation of the company’s value, or the product of
McWane’s failure to perform adequate due diligence regarding Synapse’s revenue
recognition model. In either instance, McWane did not understand the company it
108
See Respondent’s Demonstrative 1.
109
See, e.g., supra Section I.C; Tr. 183:17–186:9 (Reinhardt).
110
JX 25 at 7; JX 197 at 14.
111
See, e.g., Post-Trial OB 31, 41; Respondent’s Post-Trial Answering Br. (“Post-Trial
AB”) 47–59.
27
was buying in 2012, limiting the probative value of the 2012 Merger price in my
appraisal of the Company’s fair value as of the 2016 Merger.112
**********
The experts’ determinations of fair value using Prior Company Transaction
methods rest on an assumption that the prior transactions bear some relevance to
Synapse’s fair value in 2016. Because I am not persuaded that this connection is
factually supported, I reject the experts’ Prior Company Transaction analyses as
unreliable and, therefore, irrelevant.
3. Pre-litigation Valuations
The parties dispute the relevance of several valuations carried out by Stout
Risius Ross (“SRR”), an investment bank, before the appraisal litigation began.113
While SRR valued Synapse numerous times over a period of a year, 114 its final June
2015 valuation concluded that Synapse was worth only $0.02 per share. 115 Kruse
112
See Dell, 177 A.3d at 25 (“A market is more likely efficient, or semi-strong efficient . . .
if information about the company is widely available and easily disseminated to the market.
In such circumstances, a company’s stock price reflects the judgments of many
stockholders about the company’s future prospects, based on public filings, industry
information, and research conducted by equity analysts. In these circumstances, a mass of
investors quickly digests all publicly available information about a company, and in trading
the company’s stock, recalibrates its price to reflect the market’s adjusted, consensus
valuation of the company.”) (quotation omitted).
113
See, e.g., JX 107; JX 122; JX 123; JX 132; JX 229; JX 259.
114
Id.
115
JX 229.
28
argues that SRR’s work should not be trusted because it was pressured by a potential
client, McWane, to produce a valuation favorable to McWane. 116 Synapse responds
that SRR was never beholden to McWane, and that Synapse’s management’s
reliance upon the SRR valuations in 2015 to deem stock options worthless is a strong
indication they believed those valuations to be accurate.117
While Kruse asks me to disregard these valuations because SRR’s valuation
teams were biased, I do not need to reach that issue. No SRR employee testified at
trial, and therefore the sponsors of those valuations were not subject to the “crucible
of cross-examination,” a key tool in the assessment of evidence during the
deliberations of any factfinder, including the appraisal factfinder.118 Accordingly, I
do not rely on the SRR valuations in appraising the fair value of Synapse.119
C. Comparable Transactions Analyses
Both Sweet and Noe utilize a Comparable Transaction analyses in valuing
Synapse.120 “It [is] well within the [Court’s] discretion to view [a] comparable
116
Post-Trial OB 10–12.
117
Post-Trial AB 10.
118
In re Jarden, 2019 WL 3244085, at *1 (quoting Gilbert v. M.P.M. Enters., Inc., 1998
WL 229439, at *3 (Del. Ch. Apr. 24, 1998)).
119
This does not mean, however, that these valuations are of no aid to me in determining
Synapse’s fair value. SRR’s valuations can function as something of a “sanity check”
comparison point if my final valuation is miles from theirs.
120
Noe Opening Report ¶¶ 19–21; Sweet Opening Report 12–14.
29
companies analysis as providing relevant insights into [the company’s] value based
on inferences from how the market value[s] companies in the same industry, facing
most of the same risks.” 121 It is important to keep in mind, however, that the financial
ratios utilized in a comparables analyses “can vary widely even within the same
industry.” 122 One prominent valuation treatise, therefore, advises that comparables
analyses are a “cautious” enterprise and not as useful in determining a company’s
value as a DCF analysis.123
Sweet ultimately assigned no weight to his Comparable Transactions analysis,
noting “[t]he difficulty associated with identifying truly comparable private
transactions is great and should not be understated.”124 This fundamental
shortcoming did not stop him from completing the analysis, however. To find
comparable transactions, he searched the PitchBook database for transactions that
meet the following criteria: (1) the company operates in the IoT industry; (2) the
121
DFC, 172 A.3d at 387.
122
JX 425 (Richard A. Brealey, Stewart C. Myers and Franklin Allen, Principles of
Corporate Finance 81 (McGraw-Hill Irwin, 12th ed. 2017) [hereinafter Brealey Myers &
Allen]).
123
See id. (Noting that to understand why comparables analyses can fluctuate widely
“we need to look more carefully at what determines a stock’s market value. We start by
connecting stock prices to the cash flows that stockholders receive from the company in
the form of cash dividends. This will lead us to a discounted cash flow (DCF) model of
stock prices.”).
124
Sweet Opening Report 13; Tr. 634:15–635:10 (Sweet).
30
transaction occurred no more than 6 years before the 2016 Merger; (3) the company
generates revenue; (4) the company is not profitable; (5) the company operates in
the United States; (6) the company has revenue no more than twice Synapse’s
revenue; and (7) there exists sufficient revenue data.125 These search parameters
yielded seven transactions, of which five did not involve transfers of control, leading
him to apply a control premium to those five. 126 He then weighted the revenue
multiples for these transactions based on how similar in size the companies were to
Synapse—with weights ranging from 1.69 to 85.53—and calculated a weighted
revenue multiple of 66.26. 127 Applying this weighted revenue multiple to Synapse’s
revenue, Sweet valued the Company at $190,356,000.128
Noe approaches the problem in a similar manner but, unsurprisingly, utilizes
a different set of comparable transactions. Noe looked to the Capital IQ database
for “transactions between January 1, 2012 and December 31, 2016 involving U.S.
targets where the business description of the target or buyer contains keywords
‘internet of things,’ ‘IoT,’ ‘machine to machine,’ or ‘M2M,’” and selected only those
125
Sweet Opening Report 12–13; Tr. 632:21–633:14 (Sweet).
126
Sweet Opening Report 13.
127
Id.
128
Id.; Tr. 633:15–17 (Sweet).
31
transactions with sufficient data to construct a revenue multiple.129 He then excluded
transactions involving companies that he considered to be only tangentially related
to the IoT industry. 130 This left him with 15 transactions, which valued the firms at
a median of 2.4x revenue and a mean of 2.53x revenue. 131 Applying a 2.47x revenue
multiplier—the midpoint between the median and mean—to Synapse’s historical
revenue, and then adding Synapse’s cash on hand, Noe reached a firm value of
$6.6 million, implying negative equity value. 132 Nonetheless, he concluded that the
equity had value as an option, and valued the equity at $0.01 per share using the
Black-Scholes option pricing formula. 133
This dramatic difference between the experts neatly captures the perils of the
comparable transactions method. The two experts largely agree on Synapse’s
revenue baseline and that a revenue multiplier should be used to calculate enterprise
value, and neither uses facially unreasonable methods in selecting that revenue
129
Noe Opening Report ¶ 19; Tr. 779:8–780:16 (Noe).
130
Id.; Tr. 780:7–16 (Noe).
131
Id.; Tr. 781:10–782:5 (Noe).
132
Noe Opening Report ¶ 20; Tr. 782:6–783:9 (Noe). Noe operated under the assumption
the Disputed Money is debt. Id.
133
Noe Opening Report ¶¶ 20–21; Tr. 782:21–783:9 (Noe).
32
multiplier. 134 Nonetheless, Sweet selected a revenue multiplier nearly 27 times
larger than Noe’s, and the experts reach valuations nearly $200 million apart. 135
As is typical, both experts had much to say about the other’s analysis. Sweet
points out that some of the companies in Noe’s model had vastly more revenue than
Synapse—including companies with annual revenues of $370 million, $340 million
and $258 million.136 He argues these inclusions skew Noe’s results because,
“as companies grow in size, the revenue multiple is less and less reliable.” 137 Sweet
also notes that Noe incorporated profitable companies in his comparable set, and that
profitable companies are often valued based on earnings multiples rather than
revenue multiples.138
Noe argues that Sweet’s seven comparable transactions set included two
inappropriate entries: one included as a result of a data error,139 and one transaction
134
See Sweet Opening Report Ex. 5 (applying a revenue multiple to a baseline of
$2,460,210); Noe Opening Report Ex. 5 (applying a revenue multiple to a revenue of
$2,335,466).
135
Id.
136
Tr. 683:9–18 (Sweet); Noe Opening Report at Ex. 4.
137
Tr. 683:9–10 (Sweet).
138
Tr. 684:16–23 (Sweet).
139
Sweet’s model included a company, Newport Media, listed as having $0.043 million in
revenue. JX 343. Newport Media actually had revenue of $43 million in the year before
its acquisition. Noe Rebuttal Report ¶ 20; Tr. 790:10–791:8 (Noe). Correcting this error
33
where the target’s value derived from its possession of user data, not its potential in
the IoT space. 140 Additionally, he finds Sweet’s use of the 2012 Merger as a
comparable transaction to be inappropriate. 141 After adjusting Sweet’s findings to
remove these transactions, and correct what he saw as other, less consequential
errors, 142 Noe’s Rebuttal Report argues that Sweet’s model should find a value of
$0.01 per-share.143
I reject both the Sweet and Noe Comparable Transactions analyses. The
parties bear the burden of establishing that their respective Comparable Transactions
analyses are reliable and probative of the fair value of Synapse. 144 Neither did so.
Each expert was able to make well-considered, convincing objections to the other’s
model that were not effectively rebutted.145 Accordingly, I find these models are not
probative of Synapse’s fair value.
results in that transaction having a revenue multiple of 3.95x, not the 3,953x used by Sweet.
Id.
140
Noe Rebuttal Report ¶¶ 19–23; Tr. 791:9–792:4 (Noe).
141
Noe Rebuttal Report ¶ 23.
142
Noe Rebuttal Report ¶ 22–23 (removing value of Synapse’s NOLs from the model and
changing the cash on hand figure).
143
Noe Rebuttal Report ¶¶ 20–23; Tr. 793:11–15 (Noe).
144
In re Jarden, 2019 WL 3244085, at *33.
145
See Noe Rebuttal Report ¶¶ 19–23; Sweet Rebuttal Report 4–6. Sweet, to his credit,
gave his Comparable Transactions Analysis no weight.
34
D. Discounted Cash Flow
I turn next to the parties’ competing DCF models. Our Supreme Court has
noted that a DCF is “widely considered the best tool for valuing companies when
there is no credible market information and no market check. . . .” 146 The Supreme
Court has also recognized the reality, however, that “DCF valuations involve many
inputs—all subject to disagreement by well-compensated and highly credentialed
experts—and even slight differences in these inputs can produce large valuation
gaps.”147 With these limitations in mind, I don my waders and begin the trek into
the DCF marsh.
A DCF requires three primary inputs: (1) a projection of future cash flows;
(2) a terminal value; and (3) a discount rate.148 The company’s available cash is then
added to this calculation and its debt subtracted to arrive at a final enterprise value.149
In reviewing these inputs, I will discuss where the parties’ experts disagreed and,
146
Dell, 177 A.3d at 38; see also In re Petsmart, Inc., 2017 WL 2303599, at *23 (Del. Ch.
May 26, 2017) (Noting that, “DCF is considered by many to be the ‘gold standard’ of
valuation tools . . . .”).
147
Dell, 177 A.3d at 38. Stated more colorfully by an expert in a prior case, “garbage in,
garbage out.” In re Petsmart, 2017 WL 2303599, at *22.
148
JX 329 (Robert W. Holthausen & Mark E. Zmijewski, Corporate Valuation: Theory,
Evidence, and Practice 182 (first ed. 2013) [hereinafter Holthausen & Zmijewski]);
Noe Opening Report ¶¶ 7–9.
149
Id.35
when appropriate, determine if either expert has credibly offered a reliable input that
this Court should adopt in its own appraisal. 150
1. Free Cash Flows
The first ingredient in a DCF model is a company’s future profits. 151 One
common measure of profits, utilized by both experts in this case, is free cash flows.
“Free cash flows can be thought of as the cash generated by a company’s net
operating assets before any payments are made to debtholders in the form of interest
and/or principal to shareholders in the form of dividends.”152 They include several
inputs, including: Earnings Before Interest, Taxes, Depreciation, and Amortization
(“EBITDA”); capital expenditures (“CAPEX”)—the money a company spends on
maintaining and upgrading its capital stock—and changes in working capital and
150
M.G. Bancorporation, 737 A.2d at 526 (noting, “it is entirely proper for the [court] to
adopt any one expert’s model, methodology, and mathematical calculations, in toto, if that
valuation is supported by credible evidence and withstands a critical judicial analysis on
the record.”).
151
Brealey Myers & Allen 84.
152
Noe Opening Report ¶ 9. While free cash flows are not accepted under the Generally
Accepted Accounting Principles (“GAAP”), they are routinely used in DCF models
accepted by this Court. See generally Merion Capital, L.P. v 3M Cogent, Inc., 2013
WL 3793896, at *12–15 (Del. Ch. July 8, 2013) (calculating free cash flows); Blueblade
Capital Opportunities LLC v. Norcraft Cos., Inc., 2018 WL 3602940, at *36–39 (Del. Ch.
July 27, 2018) (same).
36
taxes.153 Here, both experts assume Synapse would have no future tax liabilities.154
Thus, taxes require no treatment separate from free cash flows.
For the most part, Noe and Sweet agree on projections of free cash flows for
2016 through 2020—although Noe’s are marginally higher—and both base their
numbers on Synapse’s management’s projections.155 I acknowledge I have some
reservations about relying on Synapse’s management’s projections given the
Company’s serial inability to meet its financial targets.156 But, both experts rely on
management projections in their analyses, and no alternate projections were offered
for my consideration. 157 With little dispute between the parties’ proffered
153
Noe Opening Report Ex. 3A; Brealey Myers & Allen 84.
154
Although Sweet and Noe originally treated the net operating losses differently, with
Sweet counting the present value of the net operating losses separately from profits, Sweet
revised his analysis in that regard. See Tr. 629:9–631:24 (Sweet). Both experts now agree
not to account for taxes, although they get there through different methods. I have adopted
Noe’s more straightforward method of not counting taxes at all, rather than Sweet’s method
of adding the value of the previous losses back by valuing net operating losses.
155
Tr. 807:10–13 (Noe). Sweet used the Synapse 2016 Plan and supplemented it with an
Excel projection file, both prepared by Synapse management before this litigation.
Tr. 603:14–24 (Sweet); Sweet Opening Report 14. Noe used a different set of projections,
also prepared by Synapse management. Noe Opening Report ¶ 8. While Kruse attacks
Noe’s use of projections he claims are “incorrect,” looking to the projections used by both
experts confirms they do not contain any material differences. Tr. 640:6–9 (Sweet);
Compare JX 178 at 5 (used by Noe) with JX 188 at 2; JX 193 (relied on by Sweet).
156
See Respondent’s Demonstratives 1–2.
157
Sweet Opening Report 14; Noe Opening Report ¶ 15. If there were other financial
projections in the record that more closely reflected the operative reality of the Company,
I may well have relied on them.
37
projections, at least for this period, I adopt Noe’s free cash flow projections for
2016–2020 for the simple reason that I find his overall DCF analysis more credible.
In accepting Noe’s free cash flow estimates, I also necessarily adopt the inputs he
used when calculating the free cash flows for this period.
Noe and Sweet differ considerably on projections of free cash flows from
2021 to 2024, although they rely on substantially similar revenue estimates. 158 These
differences are the product of very different estimates of the Company’s profit
margins.159 It is necessary, therefore, first to analyze the experts’ profit margin
projections in order better to understand their differing free cash flow projections for
this period.160
Profit margin is a measure of the difference between the cost of producing a
good and the price at which that good is ultimately sold, expressed as a percentage.161
In his Opening Report, Noe uses EBITDA profit margins from forecasts developed
by Synapse management through 2020, and then assumes the profit margin will stay
constant into the future at 12.3%—the 2020 management projection. 162 Noe then
158
Compare Noe Opening Report at 17 with JX 345 (“Sweet DCF”) at 1.
159
Tr. 748:21–749:4 (Noe).
160
See Post-Trial OB 44–47; Post-Trial AB 28–32.
161
See Tr. 602:20–603:13 (Sweet).
162
Tr. 749:7–16 (Noe).
38
compares this assumed EBITDA margin with the margins of other companies in the
IoT industry. 163 His search revealed that these companies have a median EBITDA
margin of 5.3%, with profitable companies having a median of 14.2%, supporting
his 12.3% assumption. 164
Sweet likewise uses management projections for profit margins through
2020. 165 For projecting profit margins for 2020–24, Sweet relies on a data set from
Aswath Damodaran, a highly-regarded finance professor at NYU, which shows an
industry weighted average profit margin of 24%. 166 To state the experts’
disagreement succinctly, Noe assumes profit margins will be static after 2020, while
Sweet assumes they will continue to increase dramatically into the terminal period.
Sweet justifies this assumption by arguing that cost of sales will remain at 49% of
total revenue, the same as in management’s 2020 projections, and operating
expenses will increase 5% annually from 2021 to 2024. 167 Meanwhile, he projects
depreciation and amortization will stay at the 2020 level.168
163
Tr. 750:9–751:15 (Noe).
164
Id.165
Sweet DCF at 1; Tr. 596:7–599:11 (Sweet).
166
Tr. 616:9–617:10 (Sweet); 674:1–12; 751:6–753:5 (Noe).
167
Sweet Opening Report 15.
168
Id.39
After carefully considering the competing approaches, I find that Noe has
utilized the more credible projections of profit margins. Noe’s method of estimating
a 12.3% margin is facially reasonable, and he backed this assumption with a
comparison to the median profit margin of other companies in the IoT industry.169
Although this estimate is almost certainly too optimistic given Synapse’s dismal
historic performance, Noe’s projection at least partially accounts for these past
failures by projecting Synapse’s profit margin to be slightly lower than other
profitable companies in the industry. 170
Sweet, by contrast, refers to a proprietary dataset generated for the tech
industry as a whole, including firms that are not remotely comparable to Synapse.171
Sweet initially argued these companies are a solid comparison for Synapse because
their net incomes were similar to projections of Synapse’s net income for the period
2020–24. 172 On cross-examination, however, it was revealed that Sweet had made
a fundamental error. He had mistakenly assumed the income numbers in the
169
Tr. 749:21-751:5 (Noe); Noe Rebuttal Report ¶ 18.
170
Noe Opening Report ¶ 18; Tr. 750:22–751:5 (Noe).
171
Tr. 751:11–753:14 (Noe). While I am entirely confident that the dataset is reliable,
I have no confidence that it has been reliably applied to this case. See Tr. 752:14–19 (Noe)
(“The Damodaran database doesn't allow you to calculate different metrics, means,
medians, because the individual company data is not provided because of licensing
agreements. So he only provides these industry aggregates.”).
172
Tr. 621:17–622:22 (Sweet).
40
Damodaran database were measured in “millions,” when, in fact, they were
measured in “billions,” making the companies he was comparing to Synapse much
larger than he believed. 173
With Sweet’s error revealed, Kruse attempted to pivot and argue that because
larger companies do not necessarily have the highest profit margins, the error was
harmless.174 Noe demonstrated, however, that in Damodaran’s database the largest
companies in Synapse’s industry did, in fact, have the highest profit margins.175
Because that database uses a “weighted average,” not a median, these large
companies with high profit margins skew the numbers higher. 176
Kruse’s disagreement with Noe’s calculated 12.3% profit margin, by contrast,
amounts to little more than “it is too low.”177 In particular, Kruse criticizes Noe for
173
Tr. 713:23–718:6 (Sweet).
174
Tr. 799:15–24 (Noe). (“Q: Do the largest companies always have the highest EBITDA
margins? A: No, that need not always be the case. For this particular industry at this
particular time, it was the case. But it need not be the case. Q: In fact, it is possible that
many large companies may have EBITDA margins lower than the average? A: That is
possible. Not in this case, but it is possible.”).
175
Id.; Tr. 753:17–22 (Noe) (“Q: And so what’s the significance of those large players
when we’re talking about Synapse? A: Well, those are the largest companies in the
industry, and, therefore, they’re going to have the largest effect on this industry aggregate
metric. And it turns out, for this industry at this time period, that those large companies
were also some of the most profitable companies.”).
176
Tr. 752:8–10 (Noe).
177
Petitioner’s Post-Trial Reply Brief 20 n.15 (“Synapse concedes Noe’s EBITDA
margin is lower than the median margin for profitable companies in those industries.”).
41
assuming Synapse’s profit margin would flatten out after 2020 despite growing the
previous years. 178 In doing so, Kruse does not identify any academic literature that
suggests Noe’s assumptions are incorrect or disfavored. 179 Indeed, Kruse’s breezy
assertion that Noe’s 12.3% profit margin is too pessimistic is hard to square with the
undisputed fact that Synapse’s management’s financial projections, as a matter of
course, wildly overestimate the Company’s future earnings.180
As Sweet’s method includes an error that appears artificially to inflate his
proffered profit margin, and Kruse is unable to mount any persuasive criticism of
Noe’s methodology, I find Noe’s more conservative methods to be credible and most
reliable. Thus, I adopt a 12.3% profit margin going forward.
As noted, the experts used materially similar revenue estimates in their
projections of Synapse’s free cash flows for 2021–24.181 The differences in the
178
Tr. 643:21–644:5 (Sweet); JX 371 (“Sweet Rebuttal Report”) 2; Post-Trial OB 45
(“Noe’s sharp year-to-year swing contrasts unfavorably to Sweet’s reasonable and gradual
decline . . . .”).
179
Sweet relies heavily on Professor Everett Rogers’ book Diffusion of Innovations to
support his growth trajectory of free cash flows (and therefore profit margins) and to argue
against Noe’s leveling off. Sweet Rebuttal Report 4. As was made clear at trial, however,
Diffusion of Innovations discusses revenue growth, not free cash flow growth. Tr. 747:16–
748:20 (Noe). It is, therefore, irrelevant here as both Sweet and Noe’s revenue projections
do not contain material differences. Compare JX 178 at 5 (used by Noe) with JX 188 at 2;
JX 193 (relied on by Sweet).
180
See Respondent’s Demonstratives 1–2.
181
Compare JX 178 at 5 (used by Noe) with JX 188 at 2; JX 193 (relied on by Sweet).
42
experts’ free cash flow projections for this period emanate from their competing
profit margin estimates. 182 In line with his profit margin estimates, Sweet generates
his free cash flow projections for 2020–24 by reference to Synapse’s projected 2019
growth rate. 183 Management’s projected 2019 revenue growth rate was 48%, and
Sweet projects this will drop to a projected 2020 growth rate of 38%. 184 Continuing
this trend, Sweet projects a 28% growth rate in 2021 and an 18% growth rate in
2022, followed by a 13% growth rate in 2023, an 8% growth rate in 2024, and a 3%
terminal growth rate.185 He then uses his profit margin calculation—subtracting out
Synapse’s capital expenditures and additions to working capital—to estimate
Synapse’s free cash flows for this period.186 By contrast, consistent with his profit
margin estimates, Noe projects that Synapse’s free cash flows would rise 20% per
182
For a company with no taxes, FCF = EBITDA-CAPEX-Working Capital Investment.
Noe Opening Report Ex. 3A. Noe did not include estimates for CAPEX and Working
Capital Investments in his DCF. Id. Sweet, however, did include these numbers. As his
estimates for 2021–24 hew closely to the estimates Noe used for the 2016–2020 period, it
is reasonable to infer that their wildly different FCF estimates result from differences in
calculated profit margins. See Sweet DCF; Noe Opening Report Ex. 3A. Noe confirmed
at trial that the experts’ different FCF estimates resulted from their different estimations of
Synapse’s profit margins. Tr. 748:21–749:4 (Noe).
183
Sweet Opening Report 14; Tr. 616:11–617:1 (Sweet).
184
Sweet Opening Report 14.
185
Sweet Opening Report 14–15.
186
Sweet DCF.
43
year through 2025 based on a report about future growth prospects in the industry,
and then used a 3.1% terminal growth rate.187
Sweet’s free cash flow projections for the disputed period are as follows:
Year 2021 2022 2023 2024
FCF 188 18,455,394 29,521,774 38,372,499 44,223,998
And, Noe’s projections for the disputed period are as follows:
Year 2021 2022 2023 2024
FCF 189 8,956,673 10,748,008 12,897,610 15,477,131
Because I have adopted Noe’s projected profit margin, I also adopt his projected free
cash flows for 2021–24.190
2. Discount rates
Discount rates are the second crucial ingredient of a DCF analysis.191
Discount rates reflect the common-sense notion that a dollar tomorrow is not as
187
Noe Opening Report ¶ 10 (citing an April 2016 report by IHS Markit about projected
IoT growth rates from 2021 to 2025 as an industry); Noe Opening Report Ex. 3A.
Tr. 643:5–15 (Sweet). As discussed below, Noe only projected free cash flows through
2024 before making his terminal value calculation.
188
Sweet DCF (“AMS-7 DCF” tab).
189
Noe Opening Report Ex. 3A.
190
I again express hesitancy in accepting any calculations based on management’s
projections. Given the lack of credible alternate indications of value in this case, however,
I find using a DCF, even based on these questionable projections, to be the most (and, here,
only) reliable indicator of Synapse’s value.
191
Brealey Myers & Allen 83–84.
44
valuable as a dollar today. 192 Accordingly, a discount rate is applied to each future
cash flow to determine that cash flow’s present value.193 The applied discount rate
reflects a combination of factors and can be calculated several different ways. Both
experts here relied on the well-accepted weighted average cost of capital (“WACC”)
methodology in calculating Synapse’s discount rate.194 WACC is the “cost of capital
(discount rate) determined by the weighted average, at market value, of the cost of
all financing sources in the business enterprise’s capital structure.” 195
Noe used two different models to calculate Synapse’s WACC. His first model
calculates a WACC of 12 percent; the second calculates a WACC of 40 percent.196
For the first model, Noe looked to cost of capital estimates from Duff & Phelps 2015
Valuation Handbook—a common source for valuation experts—in industries that he
found to be related to the IoT. 197 He identified four comparable industries, which,
combined with Duff & Phelps’ eight estimation methods, gave him 32 estimates of
192
JX 357 at 3, 7.
193
Noe Opening Report ¶ 7.
194
Noe Opening Report ¶ 12: Sweet Opening Report 5.
195
JX 357 at 9.
196
Noe Opening Report ¶¶ 13, 16; Tr. 757:1–2 (Noe); Tr. 770:20–771:2 (Noe). Both
experts adopted a half-year convention for their discounting, a convention I adopt as well.
Sweet DCF n.6; Noe Opening Report Ex. 3A n.8.
197
Noe Opening Report ¶ 12; Tr. 742:13–20 (Noe).
45
Synapse’s WACC. 198 He averaged the 32 to arrive at a 12% estimate of Synapse’s
WACC. 199 In his second model, Noe added a premium based on a startup company’s
risk of complete failure. 200 He used a 40% discount rate between 2016 and 2020 in
this model, which he described as “the low end of the discount rate range applied to
first-stage ventures by venture capitalists, to incorporate the probability of
failure.” 201
Sweet used the Capital Asset Pricing Model (“CAPM”) to calculate Synapse’s
equity cost of capital. 202 Because Sweet assumes that Synapse’s capital structure
includes no debt, its equity cost of capital in his model is equal to its WACC.203
CAPM is “a model in which the cost of capital for any stock or portfolio of stocks
equals a risk-free [interest] rate plus a risk premium that is proportionate to the
systematic risk of the stock or portfolio.”204 Estimating a company’s cost of equity
capital using the CAPM requires estimates of three main inputs: the risk-free interest
198
Noe Opening Report ¶ 12.
199
Id. (citing 2015 Valuation Handbook – Industry Cost of Capital (Duff & Phelps, 2015)).
200
Noe Opening Report ¶ 16: Tr. 770:17–772:1 (Noe).
201
Noe Opening Report ¶ 16. Noe then used a 12% discount rate after this first four years.
Id.202
Sweet Opening Report 15; Tr. 652:19–23 (Sweet).
203
Sweet DCF at 1; Tr. 652:19–653:16 (Sweet).
204
JX 357 at 2.
46
rate, the company’s beta, and the market risk premium. 205 Sweet found a risk free
rate based on the 20-year US Treasury Bond yield of 2.68%; an equity risk premium
of 6.03%; a full information beta multiplier of 1.01; and a size premium of 5.4%.206
Thus, he finds an equity cost of capital, WACC, and discount rate of (2.68% +
6.03%) * 1.01 + 5.4% = 14.2%.207
After carefully considering the evidence, I give no weight to Noe’s
“probability of failure” WACC model. While Noe was able to point to academic
studies backing use of a company specific risk adjustment, Synapse was not able
convincingly to argue that its actual risk of failure justifies this adjustment.208
McWane—a mature, profitable company—had demonstrated a persistent
willingness to provide seemingly unlimited capital financing to Synapse.209
Accepting Noe’s risk adjustment would require this Court to assume, in essence, that
McWane would abandon this practice and let its subsidiary fail. I decline to do so.
205
JX 399 (Shannon P. Pratt, Cost of Capital 76 (John Wiley & Sons, Inc., 2nd ed. 2002)
[hereinafter Pratt]). More sophisticated models, like the one used by Sweet, also include
“size effect and specific risk” inputs. Id.; JX 346 at 1.
206
JX 346 at 1.
207
Id.208
Noe Opening Report ¶ 16; Tr. 770:17–773:19 (Noe).
209
See JX 256; JX 346.
47
I also decline to adopt Sweet’s WACC calculation. As I have found that the
Disputed Money is debt, I cannot accept Sweet’s WACC model that assumes
Synapse’s capital structure is 100% equity, and accordingly does not include a cost
of debt. 210
As noted, Noe relied on industry estimates to peg Synapse’s WACC at 12%.
Kruse argues this was in error because, assuming, as Noe does, the Disputed Money
is debt, the industry estimates Noe relied on do not properly value the unique benefit
McWane provided Synapse by loaning it money at a rock bottom interest rate.211
Synapse responds by invoking Modigliani and Miller’s theory that altering a firm’s
capital structure by substituting cheaper debt for more expensive equity does not
reduce its WACC because investors will react by treating that firm’s equity as
increasingly risky. 212
The difficulty with Kruse’s position is that his expert did not provide a
calculation of what Synapse’s WACC would be if the Company’s capital structure
included debt; Sweet simply criticized Noe’s choice without providing an
210
JX 346. As Sweet’s calculated WACC is higher than Noe’s, this choice is favorable to
Kruse, resulting in a higher valuation.
211
Tr. 623:11–19 (Sweet); Tr. 646:23–647:14 (Sweet); Post-Trial OB 52. Presumably, the
typical IoT firm does not receive such heavily subsidized loans.
212
Tr. 765:3–766:22 (Noe); Post-Trial AB 36. See generally JX 397.
48
alternative. 213 Consequently, even assuming Noe’s choice not to account for
Synapse’s low cost of debt was improper (an assumption I do not make), I have no
“corrected” WACC calculation to consider as an alternative.
While I could attempt this calculation sua sponte, such an exercise would be
of questionable reliability. Given that Synapse has no warrants, no preferred stock
and an assumed 0% tax rate, rWACC = rE * (VE/VF) + rD * (VD/VF). 214 While rD
is known (1.16%), and rE could be calculated using the information the experts have
provided, solving VE/VF and VD/VF—the capital structure ratio of equity and
debt—would be far more difficult. 215 Solving for these ratios would require
estimating the market value of both the debt and equity. 216 And there are serious
problems with attempting to undertake these estimates. First, as Noe admits, the
interest rate paid by Synapse on its debt is dramatically below the rate it would be
able to secure in a market loan.217 Given that the typical early-stage, unprofitable
213
See Sweet DCF; Sweet Rebuttal Report 3; Tr. 647:10–14 (Sweet).
214
Pratt at 46. Where rWACC is the weighted average cost of capital, rE is the cost of equity
capital, (VE/VF) is the capital structure ratio of equity (meaning the percentage of the
company’s capital structure that is equity), rD is the cost of debt, and (VD/VF) is the capital
structure ratio of debt (meaning the percentage of the company’s capital structure that is
debt).
215
Id. This is, in no small part, because neither expert provided this calculation in their
reports or in their trial testimony.
216
Id. at 48–49.
217
Tr. 743:15–744:10 (Noe).
49
company like Synapse would be required to pay a far higher interest rate than 1.16%
on any market loan, calculating the debt’s market value is especially tricky—it might
well be close to zero. 218 Second, estimating the market value of Synapse’s equity as
a step in valuing Kruse’s shares would involve an obviously recursive process. And,
even assuming both these exercises could be undertaken, the iterative method
utilized in estimating the value of a closely held company’s equity involves a healthy
dose of guesswork. 219
Accordingly, I accept Noe’s 12% calculation based on industry estimates.
Utilizing industry estimates of WACC when, for the reasons discussed above, a
company specific number is difficult to calculate is a well-accepted method in the
valuation profession.220 While it is possible this method does not properly value the
extremely cheap debt that Synapse has access to, given the problems with the other
proffered estimates, I am convinced Noe’s methods result in the most reliable
calculation of Synapse’s WACC.
218
Tr. 744:5–10 (Noe); Tr. 646:8–647:14 (Sweet).
219
Pratt at 48–52.
220
Tr. 742:13–743:7 (Noe); Noe Opening Report ¶ 12.
50
3. Terminal value
In a DCF analysis, future cash flows are only projected out for a certain period
of time—five years in this instance.221 Rather than continuing to forecast annual
cash flows after this period, valuation experts typically calculate a firm’s terminal
value as an estimation of its cash flows into perpetuity. 222 Two common methods
for computing a firm’s terminal value are the perpetual growth model (commonly
referred to as the Gordon Growth model) and an exit multiples method (which
assumes a sale of the company). 223 This terminal value is then discounted to a
present value. 224 As a company’s terminal value can account for a large percentage
of its value in a DCF (it is nearly 100% of Synapse’s value in Sweet’s DCF and
100% of the value in Noe’s), understanding the experts’ disagreements on this point
is essential.225
221
JX 330 (Paul Asquith & Lawrence A. Weiss, Lessons in Corporate Finance 348
(John Wiley & Sons, Inc.) [hereinafter Asquith & Weiss]); Noe Opening Report Ex. 3A;
Sweet DCF.
222
Asquith & Weiss 348.
223
Id.; Tr. 605:20–606:13 (Sweet).
224
Brealey Myers & Allen 83.
225
Sweet DCF at 1; Noe Rebuttal Report ¶ 13. Without the terminal value in Noe’s
calculation, Synapse has a negative equity value of over $30 million.
51
Both Noe and Sweet estimate free cash flows through 2024, then project those
numbers into perpetuity to calculate a terminal value. 226 Noe calculates his terminal
value using the Gordon Growth model.227 This is a standard and accepted method
of measuring terminal value; it assumes that the company’s free cash flows will grow
at a constant rate in perpetuity. 228 Noe’s model uses a perpetual growth rate of 3.1%,
in line with the long-term growth rate of the United States economy. 229 Again, this
is a standard and accepted choice of a growth rate for a terminal value.230 This
method results in a terminal value of $206,680,424, with a present value of
$79,639,499. 231
Sweet calculates a terminal value using a different, but also well-accepted,
method—the exit multiple model. 232 An exit multiple calculates a company’s
terminal value as some multiple of future earnings, in this case as a multiple of
226
Noe Opening Report ¶ 10; Noe Opening Report Ex. 3A; Sweet Opening Report 14;
Sweet DCF.
227
Noe Opening Report ¶ 10; Tr. 741:6–9 (Noe).
228
Tr. 754:17–755:2 (Noe); Asquith & Weiss 348.
229
Noe Opening Report 10; Tr. 741:6–9 (Noe).
230
Asquith & Weiss 350.
231
Noe Opening Report Ex. 3A.
232
Sweet DCF; Asquith & Weiss 348.
52
Synapse’s projected EBITDA. 233 Sweet derived this multiple by examining the sales
of companies similar to Synapse and calculating the sale price as a multiple of the
target company’s EBITDA. 234 This method resulted in an exit multiple of 21.9x.235
Applying this multiple to his projection of Synapse’s 2024 EBITDA results in a
terminal value of $1,074,777,895, with a present value of $305,491,909. 236
While neither of the terminal valuation methods utilized by the experts is, in
a vacuum, superior to the other, Noe credibly points out that Sweet’s terminal value
calculation suffers from a serious flaw. 237 Although Sweet calculated his terminal
valuation based on an EBITDA multiple, Noe was able to demonstrate that implicit
in this calculation is an assumption Synapse will grow into perpetuity at a 10%
growth rate—far beyond the conventional limit of the long-term GDP growth rate.238
This is a highly questionable assumption because, although a company can be
expected to grow at least at the rate of inflation, “[i]t is theoretically impossible for
233
Tr. 755:15–20 (Noe); Tr. 656:16–657:13 (Sweet); Sweet DCF.
234
JX 342.
235
Tr. 609:1–2 (Sweet). Although Noe did not calculate an exit multiple, his model implies
a 7.5x EBITDA multiple. Tr. 755:6–11 (Noe).
236
Sweet DCF.
237
Noe Rebuttal Report ¶ 13; Tr. 759:4–760:8 (Noe).
238
See Noe Rebuttal Report Ex. 2. Sweet acknowledged that Noe’s calculation was correct
at trial. Tr. 718:12–719:3 (Sweet).
53
the sustainable perpetual growth rate for a company to significantly exceed the
growth rate in the economy. Anything over a 6–7% perpetual growth rate should be
questioned carefully.” 239
Sweet defends this choice on two grounds. First, he argues that a sale is in
tune with Synapse’s management’s strategy, and therefore an exit multiple is the
superior method to provide a terminal value in this case.240 Second, he maintains
that his model’s assumption of a long-term growth rate of 10% is justified by
historical nominal stock market returns “in the 11 ½ percent range.” 241
Neither justification is persuasive. Even assuming that an exit multiple better
reflects Synapse’s management’s strategy, 242 any future suitor would be well aware
of the impossibility of permanent 10% growth given the Company’s abysmal track
239
Pratt 113; see Asquith & Weiss 350 (“Growth rates many times faster than the overall
economy usually occur only for a short period, when a firm is starting and is either
increasing market share at the expense of its competitors or creating a new industry.”).
240
Tr. 657:1–13 (Sweet).
241
Tr. 671:4–7 (Sweet).
242
As Noe credibly explained at trial, because both methods seek to calculate the same
thing, there should not be company specific reasons for applying one or the other.
See Tr. 762:24–763:12 (Noe) (“Both the Gordon Growth methodology for estimating
terminal value, as well as using a[n exit multiple], be it EBITDA or some other measure of
financial performance, are both reasonable approaches to estimating terminal value. They
are not context specific. So one would not apply a multiple just because you’re envisioning
an exit strategy that would involve the sale of the company. Both methodologies are doing
the same thing, which is capitalizing those future cash flows beyond the terminal date into
a terminal value. And, therefore, they should provide you with consistent answers.”).
54
record, and would refuse to pay a purchase price that was based on such fantastical
assumptions.243 As to Sweet’s second point, nominal stock market returns are simply
not a relevant metric in a model, such as a DCF, that seeks to value a specific
company. As Noe made clear at trial, “the stock returns of companies in the stock
market reflect the returns of companies, some of which might be in a terminal growth
phase, but many others of which will be in [a] high growth phase[,]” making any
comparison between these returns and Synapse’s terminal growth rate
inappropriate.244 Because I find Noe’s calculation of Synapse’s terminal value to
be more credible, and have no basis in the credible evidence to alter it, I adopt it
wholesale.245
243
Pratt 113; Asquith & Weiss 350.
244
Tr. 761:14–20 (Noe).
245
M.G. Bancorporation, 737 A.2d at 526. Although Kruse never criticized Noe’s actual
terminal value calculation—instead focusing his fire on Noe’s decision to utilize a Gordon
Growth model—I discovered something puzzling in Noe’s calculation when reviewing the
evidence. Noe discusses calculating FCFs for a five year period following Synapse’s
management’s projections using an industry-wide 20% growth rate. See Noe Opening
Report ¶ 10. His DCF, however, only projects out for four years before calculating a
terminal value. See Noe Opening Report Ex. 3A. In calculating his terminal value, Noe
also appears to deviate from the well-accepted perpetual growth formula. When using a
perpetual growth model, a company’s terminal value = (FCFt * (1+g)) / (r-g), where FCFt
equals the last year of calculated free cash flows, g equals the chosen growth rate and r
equals the applicable discount rate. Pratt 25. For Noe’s chosen numbers, this should have
resulted in ($15,477,131 * (1+.031) / (.12-.031)), equaling a terminal value of
$179,291,265, with a net present value of $68,423,603. See Noe Opening Report Ex. 3A,
attached workbook. This is not what Noe did, however. His terminal value formula,
instead, was (FCFt * (1+i) / (r-g)), where i is the projected growth rate for the IoT industry
from 2021–25—which he used to calculate FCFs from 2021–24. Thus, his calculation was
($15,477,131 * (1+.20) / (.12-.031)) resulting in a terminal value of $208,680,424, with a
55
4. Cash available
Sweet makes an adjustment to Synapse’s value based on cash available of
$915,375,246 but Noe argues that this is budgeted and not actual cash. Accordingly,
he adopts a cash value of $828,280.247 The issue was not vetted at trial. Because
Sweet’s number is indeed a projection, and Noe’s is a reported number, I adopt
$828,280 as Synapse’s cash available for valuation purposes.248
5. Debt
The experts dispute how much debt Synapse carried on the date of the 2016
Merger. Noe lists the debt as $27.7 million as of December 2015, and Sweet lists it
as $29.343 million as of January 11, 2016. 249 Neither expert provided a debt figure
net present value of $79,636,499. Id. This is a long way of saying that I have found Noe’s
actual calculation (as opposed to his methodology) difficult to follow. Unfortunately, the
nuts and bolts of the calculation were not discussed at trial or in Sweet’s Rebuttal Report.
Corporate finance is not law, and my expertise is not in valuation. It is entirely possible
that I am mistaken in my analysis and Noe’s calculation was wholly appropriate. Given
that this aspect of Noe’s calculation was not addressed during this litigation, and it is
substantially more favorable to Kruse than my calculated alternative, I hesitate to fiddle
with the final calculation now. Instead, with these reservations noted, I adopt Noe’s
calculation as fact.
246
Sweet DCF.
247
Noe Rebuttal Report 8 n.19.
248
For reasons that were not discussed at trial, Noe did not add this cash available in his
calculation of Synapse’s value.
249
Noe Opening Report ¶ 13; Sweet Opening Report 16. As discussed, Sweet treats this
amount as contributed capital, not debt, in his DCF.
56
for the date that really matters—the date of the 2016 Merger. Although it would be
possible to estimate the value of the debt as of the 2016 Merger using an approximate
growth rate, given the imprecision of that exercise, I elect to use the calculated debt
value most temporally proximate to the 2016 Merger ($29.343 million).
6. The DCF Valuation
For the reasons stated above, I am satisfied that Noe has offered the most
reliable appraisal of Synapse’s fair value in one of his two DCF valuations. While
not perfect, Noe’s DCF valuation is far more credible than any of the valuations
proffered by Sweet, and far superior to any valuation I might endeavor to undertake
on my own.
III. CONCLUSION
For the foregoing reasons, I find that the fair value of Synapse’s equity as of
February 2, 2016, was $20,347,822, or $0.228 per share.250 The legal rate of interest,
compounded quarterly, shall accrue on Kruse’s appraisal award from the date of the
2016 Merger’s closing to the date of payment. The parties shall confer and submit
an implementing order and final judgment within ten days.
250
The precise calculation is attached as Appendix A.
57
APPENDIX A
Year 2016 2017 2018 2019 2020
Free Cash Flow $ (24,656,891) $ (22,046,813) $ (13,455,789) $ (6,826,545) $ 7,463,894
Present Value Factor 0.9449 0.8437 0.7533 0.6726 0.6005
NPV $ (23,298,572) $ (18,600,250) $(10,135,942) $ (4,591,323) $ 4,482,129
Terminal
2021 2022 2023 2024 Terminal (Court)
$ 8,956,673 $ 10,748,008 $ 12,897,610 $ 15,477,131 $208,680,424 $ 179,291,265
0.5362 0.4787 0.4274 0.3816
$ 4,802,281 $ 5,145,302 $ 5,512,823 $ 5,906,596 $ 79,639,499 $ 68,423,603
Noe Court
Firm Value $ 48,862,542 $ 48,862,542
Cash available $ - $ 828,280
Debt $ 27,747,500 $ 29,343,000
Equity value $ 21,943,322 $ 20,347,822
Shares outstanding 89,063,787 89,063,787
Per share equity
value $ 0.246 $ 0.228
Number of Kruse
shares 582,216 582,216
Value of Kruse shares $ 143,444.98 $ 133,015.09
Court's Alteration