|
Steiner Corporation v.
Benninghoff
5 F. Supp 2d 1117 (D. Nevada)
May 26, 1998
[spreadsheet
summary] [full
case]
Steiner Corporation was a privately-held
family business that supplied linen and rental
uniforms in the lucrative Las Vegas market.
Steiner also had subsidiaries in private-label
food canning and textile filter manufacturing.
Most of Steiner's stock was in family hands (93%),
but over time some company employees had acquired
minority holdings.
In July 1994,
facing the possibility of an IRS accumulated earnings
tax for not distributing its munificent profits
to shareholders, the Steiner family bought
out the company's minority shareholders in a cash-out
merger for $1200/share. Although a
majority of the minority approved the merger,
a group of minority shareholders (the Benninghoffs)
dissented from the merger and sought payment of
"fair value" under Nevada's appraisal statute.
See Nev.Rev.Stat. §§ 92A.315, 92A.320,
92A.360 to .500.
The Nevada statute
called for the company to pay its estimate of
"fair value" and then negotiate if the dissenters
wanted more. After being paid $840/share,
the dissenters claimed their shares were worth
at least $1950/share, but the company refused
to budge. Faced with this impasse, the company
(as required by the appraisal statute) sought
a judicial determination of the “fair value” of
its common stock as of the date of the merger.
The stock price
had been the subject of some study and negotiation
before the merger. Originally the Steiner
family had hired Houlihan, Lokey, Howard, and
Zukin [hereinafter HLHZ], which determined a value
of $1418/share. When the family offered
to buyout the minority for $975/share, the Steiner
board appointed a special committee to represent
the minority. The committee hired J.P. Morgan,
which came up with a range of $1100-$1500/share.
After some perfunctory negotiations, the Steiner
family and the special committee agreed on $1200/share.
The court divided
its task into the following components:
The Benninghoffs,
believing the value was more, elected dissenters’
rights and voted against the merger. They
retained David Nolte to perform a valuation.
Steiner then retained another expert, Dr. Kleidon,
who examined Nolte’s work for flaws.
I.
Standard of value - "fair value"
The court interpreted the nevada
statutory standard to require a determination
of
(1) the pre-merger
market value of the shares, discounted for illiquidity,
(2) the pre-merger
enterprise value of the corporation as a whole,
(3) the pre-merger
net asset value of the corporation, and
(4) any other
factor bearing on value.
For guidance,
the court held out the "valuation bible" -- Shannon
P. Pratt, Robert F. Reilly, & Robert
P. Schweihs, Valuing a Business: The Analysis
and Appraisal of Closely Held Companies (3d ed.1996).
But the case was not to be merely a battle of
experts: "Of course, it is the law that
we must follow, which the various experts in valuation
theory and economics who testified at trial may
not have completely realized."
The court,
as a preliminary matter, held that the Benninghoffs'
shares would not be subject to a minority discount
or a control premium. INstead, they would
receive a pro rata share of the company as a whole
(enterprise value).
The court then laid out the three
most-accepted valuation methods:
(1) discounted
cash flow analysis, or “DCF”;
(2) the “market
comparables” or “guidelines companies” approach;
and
(3) the “acquisitions”
method.
The court noted
that the DCF and acquisitions approach yield an
enterprise value, while the market comparables
approach yields a marketable minority value (what
Steiner’s shares would trade for if it were a
public company). But the court gushed its
preference for the DCF method: "The basic
premise of all valuation theory is that
what a company is worth today is a reflection
of what it can be expected to earn in the future."
There was one sticking point.
Should post-merger events be used in valuing the
Benninghoffs' shares? The company said the
DCF model should use a cost-of-capital figure
derived from the company's actual, low debt-to-equity
ratio (4 - 96). The Benninghoffs suggested
an industry figure (xx-yy), since Steiner's capital
structure was deviant. So the court said
it would construct a debt-to-equity ratio (and
chose 20 - 80) based on the capital structure
the market would "expect." This the court
said did not necessarily impound any post-merger
changes.
The company also argued that an
"acquisitions" valuation using comparable deals
would reflect gains (like synergy gains) expected
from those acquisitions and violate taking into
account post-merger events. The court agreed
that this would be a no-no under the Nevada statute
and the similar Delaware appraisal statute.
But the court imagined that somehow any synergy
gains could be disregarded, and these comparable
deals would be relevant.
A.
Market value of shares (with liquidity discount)
The court agreed with the Benninghoffs
that there was really no market value for their
shares. Although the company had bought
back some shares and some shareholders had sold
their shares, this means next-to-nothing.
Sometimes shareholders are desperate for immediate
cash!
But maybe, the court said, a
“hypothetical” market could be constructed
This involves analysis of similar, but public,
corporations to estimate the price at which
Steiner’s stock would have traded were it public.
Both of the parties' experts used a “market
comparables” approach to determine enterprise
value, but really their figures suggested a
marketable minority value. Both JPM and
Nolte used a segmented method, comparing Steiner’s
subsidiary to various food processing businesses,
while comparing the rest of Steiner to various
linen rental companies.
Once comparable
companies were identified, the experts calculated
Price-to-Earnings (PE) multiples and Total Invested
Capital (TIC) to Earnings Before Interest, Tax,
Depreciation, or Appreciation (EBITDA) multiples
for these comparables. Each multiple was
then applied to Steiner’s known values, EBITDA
and earnings, to determine the value of the Steiner’s
equity. Nolte’s value of $515,534,000 was
well within JPM and Kleidon’s calculated ranges,
and the court accepted that figure. Adding
this to Steiner’s excess cash (as found by Nolte)
and then dividing that sum by the total number
of outstanding shares produced per share market
value, $1589.31.
The court, for reasons that are
hard to fathom, discounted this "pre-merger
market value" to account for the illiquidity
of the close corporation's shares. The
court accepted the illiquidity discount could
range from 20% to 60%, and a low-end 25% discount
because it would be inconsistent to allow the
majority to cash-out a minority and then argue
the shares lacked a ready market. Thus,
the court said Steiner had a pre-merger market
value of $1191.98/share.
Final answer "market value":
$1198.98
B.
Pre-Merger Enterprise Value
1.
Discounted Cash Flow Method
The court focused
its attention on the DCF method -- which assumes
a company's value is the amount of cash it will
generate in the future, discounted to present
value. Citing Delaware cases, it broke the
DCF task into three components:
a.
Estimation of net cash flows
The court accepted
(as did the experts) that cash flow can be measured
by EBITDA --- the amount of sales, or revenue,
generated by the company, minus the cost of generating
those sales. Then the experts compute an
"EBITDA margin" for various years -- EBITDA divided
by that year’s sale’s figures. Then using
estimates of sales, the experts estimated future
EBITDA. Nolte (the dissenters' expert) estimated
future sales by starting with the latest actual
sales figure and projected 8% growth for the first
year (1995) decreasing to 6% by 1999, the last
year of the period. Remarkably, Kleidon (the company's
expert) agreed with Nolte's sales projections,
so the court did also. On the question
of EBITDA margin, Nolte used 15% and Kleidon used
14%, and the court pointed out that Steiner’s
own budget for 1995 projected an EBITDA margin
of 15.4% and actual margin was 14.6%. Given
Steiner's industry position, the court accepted
Nolte's 15%.
Multiplying estimated sales by
the EBITDA margin, the court had an EBITDA flow.
Next came taxes. Since
taxes are not part of the EBITDA, they were
computed first by calculating taxable income
(which is EBITDA minus depreciation, and amortization).
Then tax was computed assuming Steiner’s effective
tax rate of 43%. Then, since the court
said it was looking for net cash flow, depreciation
and amortization was added back. But there
was some controversy about interest expenses.
The court pointed out that these payment to
outside creditors reduced cash flow, but only
after-tax interest expenses should be
added back. So the court used a 26.7%
tax rate (reflecting that only 62% of the company's
interest was deductible) to assume that interest
payments reduced cash flow by 73.3% of payments.
There were
some final adjustments, on which the experts agreed.
Add net income from unconsolidated subsidiaries
and net cash from discontinued operations.
Subtract capital acquisitions and cost of acquisitions
(as equipment needs to be replaced). The
final figure is the holy grail of DCF - "net cash
flow."
|
Year |
Net cash flow |
|
1995 |
$8,637,219.53 |
|
1996 |
$6,777,487.53 |
|
1997 |
$9,155,006.53 |
|
1998 |
$22,502,877.53 |
|
1999 |
$26,167,265.53 |
Once cash flow
was calculated, the court was ready to discount
these amounts for each year to present value.
But first there was the question of terminal value.
b.
Estimating Terminal Value
What about
cash flow after 1999? The court explained
that "terminal value" is the value of cash flows
expected to be received by the company beyond
the “terminal year” discounted to the “present
value” of the terminal year. Once
determined, the terminal value is discounted again
to its value in 1994, the year of valuation.
The court explained
that the most critical step in calculating terminal
value is determining the “perpetual growth rate.”
The experts did not agreed -- JP Morgan used 4%,
Nolte 5% and Kleidon 4-5%. The court chose
Nolte's 5%, since the growth rate in 1999 (the
terminal year) had been 6%, closer to Nolte’s
than the other numbers. The court noted
that this growth rate included inflation.
What is the present value of
a “growing perpetuity”? The court used
the Gordon model:
|
TV = Pn(1
+ g) / (k - g)
|
TV
= terminal value
Pn
= projected cash flow in the year
"n"
n =
number of years in the discrete projection
period
k =
discount rate (the proper discount rate
to use will be discussed and calculated
in the following section)
g
= perpetual growth rate |
Filling in the equation with the
values obtained above results in:
TV = $26,167,265.53
(1 + .05) / (.0978 - .05)
TV = $27,475,628.81 / (.0478)
TV = $574,803,950
c.
Determining the Discount Rate
Where did the court get the discount
rate of 9.78%
The court explained
that the discount rate is essentially the cost
of capital, the interest rate at which the company
would have to invest the “present value” in order
to have the projected future value at the end
of the projection period. It was the most
hotly contested issue in the case.
The court divided
its task into three steps:
(1) calculate the company's
cost of equity;
(2) calculate the company's cost
of borrowing, or debt;
(3) calculate the weighted
average of the costs of equity and debt, with
the weights determined by the capital structure,
or ratio of debt to equity.
(i)
The Cost of Equity
To determine the cost of equity
(Ke ), the following equation is used:
|
Ke = R f
+ beta (RP m )
|
Rf
= risk free rate of return on 5-year
U.S. Treasury Notes
RPm
= market risk premium
beta
= "the non-diversified risk associated
with the economy as a whole as it affects
this firm," or as Steiner put it in its
trial brief, "an adjustment to the general
risk premium to reflect the particular risk
of the company." |
|
The Risk Free Rate |
Here the court looked at
five-year U.S. Treasury Notes ("risk free"
because there is little risk of the U.S.
defaulting). Which period? Treasury
Note rate had fluctuated between February
and July of 1994. In February
1994, when JPM conducted its initial analysis,
the rate on 5-year Treasury Notes was 5.45%,
By the time of the merger in July, the rate
had risen to 6.9%. Thus when Nolte simply
adopted JPM's WACC, which had been calculated
in February using February data, to discount
cash flow figures projected using July data,
he was essentially using the wrong risk-free
rate. Evidence was
presented at trial that, leaving everything
else constant, the rise in the risk-free
rate alone reduced the estimated present
value of Steiner from February to July
by nearly 26%. The court was incredulous,
but decided the correct risk-free rate
was the rate from July -- 6.9%. |
|
The Market Risk Premium |
The market risk premium
is a measure of the additional return needed,
on average, to convince investors to invest
in the stock market rather than in risk-free
Treasury Notes. Both JPM and Nolte
used a 5% market risk premium. Kleidon used
a market risk premium of 7.4%. The difference
appears to be based on a legitimate split
in the valuation profession. Some analysts
recommend using a "time horizon" of only
the past twenty years or so, which results
in a risk premium of 5%, while others recommend
using all available data going back as far
as 1926, when the necessary data began to
be accumulated and processed. The court
simply found that the 5% figure is appropriate
to use in this case, given that experts
from both sides testified to that effect
(i.e, Engel and Nolte). |
|
Beta |
In addition to the general
market risk premium, the cost of equity
for a company is influenced by that company's
particularized degree of riskiness. The
factor used to represent this concept is
generally referred to as "Beta" ("beta").
Technically, a beta is the covariance of
a company's rate of return against the market
rate. Betas for many public companies are
calculated and published periodically by
several sources. If the betas for Steiner's
comparable companies are known, then the
beta for Steiner can be estimated.
The experts were vague.
Steiner's attempts at trial to portray
itself as a risky little private company
beset by various financial problems was
not very successful. Therefore, the court
held that an appropriate beta for Steiner
is approximately 0.75. (Note that a beta
of less than 1.00 indicates that a company
is less risky than the market in general.
This, the court found, was true in Steiner's
case. Given a market risk premium of 5%;
and a beta of 0.75, Steiner is thus 3.75%
riskier than investing in risk free Treasury
Notes.) |
We can now proceed to calculate
the cost of equity. Plugging the values we have
obtained into the formula set forth at the beginning
of this section:
Ke = R f
+ beta (RP m )
Ke = 6.9% + 0.75 (5.0%)
Ke = 10.65%
Thus Steiner's cost of equity is
10.65%.
(ii)
The Cost of Debt
The cost of debt equals the amount
of interest a bank would charge on a long-term
loan. Because Steiner should be able to
borrow money at only 90-125 basis points above
the risk free rate, the cost of debt for Steiner
is approximately 8.57%.
(iii)
The Weighted Average Cost of Capital
Although the cost of equity and
debt have been determined, the amount of weight
each deserves has not. The weighting is
established by the ratio of debt to equity in
the company’s capital structure. Both
JPM and Nolte support a 20%/80% ratio.
Thus the discount rate, or Weighted
Average Cost of Capital (WACC) is:
WACC = K subd
(%D) [1-TR (.62)] + K sube (%E)
WACC
= 8.58% (.20) [1-(.43)(.62)] + 10.65% (.80)
WACC
= 9.78%
d.
Discounting to Present Value
Thus, with the WACC calculations,
the net present value of Steiner is the sum of
the discounted cash flow figures and the discounted
terminal value:
NPV =
P1/(1+k).5+P2/(1+k)1.5+P3/(1+k)2.5
+P4/(1+k)3.5+P5
/(1+k)4.5 + TV/(1 + k)5
|
k = WACC
P1...P5
= net cash flow projected for years
1 - 5
TV = terminal value |
|
Year |
NPV |
NPV |
NPV |
|
1 |
$8,637,219.53 / (1 + .0978)
.5 |
($8,637,219.53 x 0.9544) |
$8,243,362.32 |
|
2 |
$6,777,487.53 / (1 + .0978)1.5 |
($6,777,487.53 x 0.8694) |
$5,892,347.66 |
|
3 |
$9,155,006.53 / (1 + .0978)
2.5 |
($9,155,006.53 x 0.7919) |
$7,249,849.67 |
|
4 |
$22,502,877.53 / (1 +
.0978) 3.5 |
($22,502,877.53 x 0.7214) |
$16,233,575.85 |
|
5 |
$26,167,265.53 /
(1 + .0978) 4.5 |
($26, 167,265.53
x 0.6571) |
$17,194,510.18 |
|
Terminal |
$574,803,950.00
/ (1 + .0978)5 |
($574,803,950.00 x 0.6272) |
$360,517,037.44 |
| |
Total - NPV |
$415,330,683.12 |
|
Excess cash |
$68,300,000 |
|
Interest-bearing debt |
($15,758,859) |
|
Total equity |
$467,872,094.12 |
|
Equity/share |
$1273.64
|
Why did the court use a half-year
factor for the discount rate? The court
said this was "mid-year discounting convention,"
which assumes that cash flows are not all received
at the end of the year. Why was excess
cash added? The court did not explain,
but presumably because it was not being used
to generate cash flow. Why was interest-bearing
debt subtracted. The court said that both Nolte's
and Kleidon's reports had done this.
Final answer "discounted cash
flow": $1273.64
2.
Acquisitions Method
Steiner believed
the Acquisitions Method was inappropriate because
it assumes the company is being sold to a third
party. While data for this method are collected
from transactions in which whole companies are
sold, the court pointed out this method does not
depend on the company being sold. It just
tells you how much the company would have fetched.
Therefore, this method is not inappropriate.
Using Nolte’s multiple of 1.3 for the “deal value
to last twelve month earnings,” the court accepted
a per share value of $1958.95.
What companies were compared?
We don't know. What were their P-E ratios?
Neither. Why was a 1.3 multiplier appropriate,
apparently after figuring out value based on P-E
comparisons? No idea.
Nonetheless, the court said the
"acquisitions" method was not to be given as
much weight as the DCF method. It used
a weighted average of the two methods -- a 70/30
ratio, which yielded an average per share enterprise
value of Steiner of $1479.22.
Final answer (enterprise value):
$1479.22.
C.
Pre Merger Net Asset Value of Steiner
The case law supports the view
that this factor should receive meaningful weight
in only a few types of cases, usually those
involving companies owning a lot of land or
other natural resources. Steiner is primarily
a service company and its main assets are non-tangible
and difficult to value – such as good will,
a trained workforce, or client lists.
Therefore this factor was given no weight in
the analysis.
D.
Any Other Factor Bearing on Value
Nine members of the Steiner family
tendered their shares to Steiner at $1200 per
share. This is evidence that $1200 is
certainly not outrageously unfair. However,
there are many reasons why a shareholder might
decide to sell their shares when it is likely
that the opportunity may not arise again.
Therefore, this factor also deserves no weight.
E.
Weighting the Various Measures of Value
Given the stress on the DCF method
at trial, the enterprise value component should
be weighted significantly more heavily than
the market component. Using 25% of market value
($1191.98) and 75% of enterprise value ($1479.22),
a final per share value of $1407.42 is reached.
| |
Court finding |
Weight |
Weighted value |
| Market value |
$1191.98 |
25% |
$298.00 |
| Enterprise value |
$1479.22 |
75% |
$1109.42 |
|
DCF |
$1273.64 |
70%(75%)=52.5% |
$668.66 |
| Acquisitions |
$1958.95 |
30%(75%)=22.5% |
$440.76 |
| Net asset value |
n/a |
0% |
$0 |
| Final fair value |
|
|
$1407.42 |
Final answer
"fair value": $1407.22
|
5.6.2 Business valuation in corporate appraisal 