<<

. 51
( 100 .)



>>




CHAPTER 7 Adjusting for Levels of Control and Marketability 233
were several other inconsistencies in the results of the two
regressions.
3. The log“log form of regression that Phillips and Freeman used
can have the effect of making large variations look small. The
standard errors of their regressions were very high. The
standard error of the Mergerstat regression was 0.925. Two
standard errors is 1.85. Exponentiating, the 95% con¬dence
interval is approximately equal to multiplying the (value/sales)
estimate by two standard errors on either side of the regression
estimate. The high side of the 95% con¬dence interval is e1.85
6.36 times the regression estimate, and the low side is e 1.85
0.157 times the regression estimate. Let™s put some speci¬c
numbers into their equation to see what the con¬dence intervals
look like. Let™s assume we are forecasting the value of the
common stock as a percentage of sales for a ¬rm over $100
million in value that is neither a bank, a private placement, nor
a subsidiary. Their regression equation is ln(Value/Sales)
3.242 0.56 ln net margin 0.45 ln (1/PE of the S&P 500).
Let™s assume a 5% after-tax margin and an average PE for the
S&P 500 of 15, so 1/PE 0.067. Then, ln(Value/Sales) 3.242
(0.56 ln 0.05) (0.45 ln 0.067) 3.242 1.678 1.219
e0.345
0.345. Thus, the regression estimate of (Value/Sales)
1.413, or value is approximately 1.4 times sales, which seems
high. If sales are $100, then net income after taxes is $5, which
when multiplied by a PE ratio of 15 leads to a value of $75,
which implies value should be 0.75 Sales, not 1.4. The
reliability of the forecast is low. The 95% con¬dence interval is
approximately: 0.22 Sales Value 8.99 Sales.
4. There were fairly few transactions with a private seller. In the
Mergerstat database, private targets were 18 out of 416
transactions, and in the SDC database, private targets were 33
out of 445 targets. In total, private targets were approximately
6% of the combined databases.
The small number of transactions with privately held sellers is not
necessarily worrisome in itself, but combined with the limitations of the
results in 1, the inconsistent results in 2, and the very wide con¬dence
intervals in 3, the results of this study are insuf¬cient to reject DLOM for
control interests of privately held ¬rms.


Kasper™s BAS Model
Larry Kasper (Kasper 1997, p. 106) uses an econometric equation devel-
oped by Amihud and Mendelson (Amihud and Mendelson 1991) to cal-
culate the bid-ask spread (BAS). Their equation is: r 0.006477 0.01012
0.002144 ln BAS, where r is the excess monthly returns on a stock
portfolio over the 90-day Treasury Bill rate and the BAS is multiplied by
100, i.e., a BAS of 25% is denominated as 25, not 0.25.
Kasper says that most business brokers would not list a business that
had to be discounted more than 25%. Substituting 25 into the above equa-



PART 3 Adjusting for Control and Marketability
234
tion, the excess return required for a BAS of 25% is 0.0069 per month, or
approximately 8.28% per year. One would then seek out business brokers
(or through IBA, Pratt™s Stats, BIZCOMPS, etc.) for actual BASs. Anyone
interested in using Kasper™s model must read his outstanding book, as
this summary is inadequate for understanding his work.
A number of differences in the environment of NASDAQ and pri-
vately held business can weaken the applicability of this regression equa-
tion from the former to the latter:
1. The BAS in NASDAQ compensates the dealer for actually taking
possession of the stock. The dealer actually stands to gain or
lose money, whereas business brokers do not.
2. It takes much longer to sell a private business than stock on
Nasdaq.
3. The market for privately held ¬rms is much thinner than it is
with Nasdaq.
4. Transactions costs are far higher in privately held business than
in Nasdaq.
Note that items 2 through 4 are the components of the economic
components approach, which we will cover shortly in my model. Also,
the reservation in 1 also applied in the Menyah and Paudyal results ear-
lier in the chapter, where the BAS depends on the number of market
makers. Again, business brokers are not market makers in the same sense
that dealers are. Additionally, as Kasper points out, the regression coef-
¬cients will change over time. Kasper also presents a different model, the
discounted time to market model (Kasper 1997, pp. 103;“04) that is worth
reading. Neither of his models considers transactions costs or the effects
of thin markets.40


Restricted Stock Discounts
We will now discuss DLOM for restricted stocks as a preparation for our
general model for DLOM. We use two valuation methodologies in cal-
culating the restricted stock discount. The ¬rst is based on my own mul-
tiple regression analysis of data collected by Management Planning, Inc.
(MPI),41 an independent valuation ¬rm in Princeton, New Jersey. The sec-
ond method involves using a Black“Scholes put option as a proxy for the
discount.

Regression of MPI Data
Ten studies of sales of restricted stocks have been published.42 The ¬rst
nine appear in Pratt, Reilly, Schweihs (1996, chap. 15) and Mercer (1997);


40. That is not to say that I downgrade his book. It is brilliant and a must read for anyone in the
profession.
41. Published in Chapter 12 of Mercer (1997). I wish to thank MPI for being gracious and helpful
in providing us with its data and consulting with us. In particular, Roy H. Meyers, Vice
President, was extremely helpful. MPI provided us with four additional data points and
some data corrections.
42. See Mercer (1997, p. 69) for a summary of the results of the ¬rst nine studies.




CHAPTER 7 Adjusting for Levels of Control and Marketability 235
in those studies, the authors did not publish the underlying data and
merely presented their analysis and summary of the data. Additionally,
only the Hall/Polacek study contains data beyond 1988 (through 1992).
The Management Planning study, which Mercer justi¬ably accords a sep-
arate chapter and extensive commentary in his book, contains data on
trades from 1980“1996 and thus is superior to the others in two ways:
the detail of the data exists and the data are more current.
Table 7-5 is two pages long. The ¬rst page contains data on 53 sales
of restricted stock between 1980“1996. Column A is numbered 1 through
53 to indicate the sale number. Column C, our dependent (Y) variable, is
the restricted stock discount for each transaction. Columns D through J
are our seven statistically signi¬cant independent variables, which I have
labeled X1, X2, . . ., X7. Below is a description of the independent variables:


# Independent Variable

1 Revenues squared.
2 Shares Sold”$: the discounted dollar value of the traded restricted shares.
3 Market capitalization price per share times shares outstanding, summed for all classes
of stock.
Earnings stability: the R 2 of the regression of net income as a function of time, with time
4
measured as years 1, 2, 3, etc.
Revenue stability: the R2 of the regression of revenue as a function of time, with time
5
measured as years 1, 2, 3, etc.
6 Average years to sell: the weighted average years to sell by a nonaf¬liate based on SEC
Rule 144. I calculated the holding period for the last four issues (DPAC, UMED, NEDI,
and ARCCA) based on changes in Rule 144, even though it was not effective yet,
because the change was out for review at that time and was highly likely to be
accepted.43 These transactions occurred near the beginning of March 1996, well after
the SEC issued the exposure draft on June 27, 1995. This was approximately 14
months before the rule change went into effect at the end of April 1997. The average
time to resale for the shares in these four transactions was determined based on the
rule change, resulting in a minimum and maximum average holding period of 14
months and 2 years, respectively.44
7 Price stability: This ratio is calculated by dividing the standard deviation of the stock
price by the mean of the stock price”which is the coef¬cient of variation of price”
then multiplying by 100. The end-of-month stock prices for the 12 months prior to the
valuation date are used.




I regressed 30 other independent variables included in or derived
from the Management Planning study, and all were statistically insignif-
icant. I restrict our commentary to the seven independent variables that
were statistically signi¬cant at the 95% level.
The third page of Table 7-5 contains the regression statistics. In re-
gression #1 the adjusted R 2 is 59.47% (B9), a reasonable though not stun-
ning result for such an analysis. This means that the regression model
accounts for 59.47% of the variation in the restricted stock discounts. The


43. According to John Watson, Jr., Esq., of Latham & Watkins in Washington, D.C., the securities
community knew the rule change would take place. In a telephone conversation with Mr.
Watson, he said it was only a question of timing.
44. In other words, I assumed perfect foreknowledge of when the rule change would become
effective.




PART 3 Adjusting for Control and Marketability
236
T A B L E 7-5

Abrams Regression of Management Planning Study Data


A B C D E F G H I J

4 Y X1 X2 X3 X4 X5 X6 X7
Rev2
6 Discount Shares Sold-$ Mkt Cap Earn Stab Rev Stab AvgYrs2Sell Price Stab

7 1 Air Express Int™l 0.0% 8.58E+16 $4,998,000 25,760,000 0.08 0.22 2.84 12.0
8 2 AirTran Corp 19.4% 1.55E+16 $9,998,000 63,477,000 0.90 0.94 2.64 12.0
9 3 Anaren Microwave, Inc. 34.2% 6.90E+13 $1,250,000 13,517,000 0.24 0.78 2.64 28.6
10 4 Angeles Corp 19.6% 7.99E+14 $1,800,000 16,242,000 0.08 0.82 2.13 8.4
11 5 AW Computer Systems, Inc. 57.3% 1.82E+13 $1,843,000 11,698,000 0.00 0.00 2.91 22.6
12 6 Besicorp Group, Inc. 57.6% 1.57E+13 $1,500,000 63,145,000 0.03 0.75 2.13 98.6
13 7 Bioplasty, Inc, 31.1% 6.20E+13 $11,550,000 43,478,000 0.38 0.62 2.85 44.9
14 8 Blyth Holdings, Inc. 31.4% 8.62E+13 $4,452,000 98,053,000 0.04 0.64 2.13 58.6
15 9 Byers Communications Systems, Inc. 22.5% 4.49E+14 $5,007,000 14,027,000 0.90 0.79 2.92 6.6
16 10 Centennial Technologies, Inc. 2.8% 6.75E+13 $656,000 27,045,000 0.94 0.87 2.13 35.0
17 11 Chantal Pharm. Corp. 44.8% 5.21E+13 $4,900,000 149,286,000 0.70 0.23 2.13 51.0
18 12 Choice Drug Delivery Systems, Inc. 28.8% 6.19E+14 $3,375,000 21,233,000 0.29 0.89 2.86 23.6
19 13 Crystal Oil Co. 24.1% 7.47E+16 $24,990,000 686,475,000 0.42 0.57 2.50 28.5
20 14 Cucos, Inc. 18.8% 4.63E+13 $2,003,000 12,579,000 0.77 0.87 2.84 20.4
21 15 Davox Corp. 46.3% 1.14E+15 $999,000 18,942,000 0.01 0.65 2.72 24.6
22 16 Del Electronics Corp. 41.0% 4.21E+13 $394,000 3,406,000 0.08 0.10 2.84 4.0
23 17 Edmark Corp 16.0% 3.56E+13 $2,000,000 12,275,000 0.57 0.92 2.84 10.5
24 18 Electro Nucleonics 24.8% 1.22E+15 $1,055,000 38,435,000 0.68 0.97 2.13 21.4
25 19 Esmor Correctional Svces, Inc. 32.6% 5.89E+14 $3,852,000 50,692,000 0.95 0.90 2.64 34.0
26 20 Gendex Corp 16.7% 2.97E+15 $5,000,000 55,005,000 0.99 0.71 2.69 11.5
27 21 Harken Oil & Gas, Inc. 30.4% 7.55E+13 $1,999,000 27,223,000 0.13 0.88 2.75 19.0
28 22 ICN Paramaceuticals, Inc. 10.5% 1.50E+15 $9,400,000 78,834,000 0.11 0.87 2.25 23.9
29 23 Ion Laser Technology, Inc. 41.1% 1.02E+13 $975,000 10,046,000 0.71 0.92 2.82 22.0
30 24 Max & Erma™s Restaurants, Inc. 12.7% 1.87E+15 $1,192,000 31,080,000 0.87 0.87 2.25 18.8
237
238




T A B L E 7-5 (continued)

Abrams Regression of Management Planning Study Data


A B C D E F G H I J

4 Y X1 X2 X3 X4 X5 X6 X7
Rev2
6 Discount Shares Sold-$ Mkt Cap Earn Stab Rev Stab AvgYrs2Sell Price Stab
31 25 Medco Containment Svces, Inc. 15.5% 5.42E+15 $99,994,000 561,890,000 0.84 0.89 2.85 12.8
32 26 Newport Pharm. Int™l, Inc. 37.8% 1.10E+14 $5,950,000 101,259,000 0.00 0.87 2.00 30.2
33 27 Noble Roman™s Inc. 17.2% 8.29E+13 $1,251,000 11,422,000 0.06 0.47 2.79 17.0
34 28 No. American Holding Corp. 30.4% 1.35E+15 $3,000,000 79,730,000 0.63 0.84 2.50 22.1
35 29 No. Hills Electronics, Inc. 36.6% 1.15E+13 $3,675,000 21,812,000 0.81 0.79 2.83 52.7
36 30 Photographic Sciences Corp 49.5% 2.70E+14 $5,000,000 44,113,000 0.06 0.76 2.86 27.2
37 31 Presidential Life Corp 15.9% 4.37E+16 $38,063,000 246,787,000 0.00 0.00 2.83 17.0
38 32 Pride Petroleum Svces, Inc. 24.5% 4.34E+15 $21,500,000 74,028,000 0.31 0.26 2.83 18.0
39 33 Quadrex Corp. 39.4% 1.10E+15 $5,000,000 71,016,000 0.41 0.66 2.50 44.2
40 34 Quality Care, Inc. 34.4% 7.97E+14 $3,150,000 19,689,000 0.68 0.74 2.88 7.0
41 35 Ragen Precision Industries, Inc. 15.3% 8.85E+14 $2,000,000 22,653,000 0.61 0.75 2.25 26.0
42 36 REN Corp-USA 17.9% 2.85E+15 $53,625,000 151,074,000 0.02 0.88 2.92 19.8
43 37 REN Corp-USA 29.3% 2.85E+15 $12,003,000 163,749,000 0.02 0.88 2.72 36.1
44 38 Rentrak Corp. 32.5% 1.15E+15 $20,650,000 61,482,000 0.60 0.70 2.92 30.0
45 39 Ryan™s Family Steak Houses, Inc. 8.7% 1.02E+15 $5,250,000 159,390,000 0.90 0.87 2.13 13.6
46 40 Ryan™s Family Steak Houses, Inc. 5.2% 1.02E+15 $7,250,000 110,160,000 0.90 0.87 2.58 14.4
47 41 Sahlen & Assoc., Inc. 27.5% 3.02E+15 $6,057,000 42,955,000 0.54 0.81 2.72 26.1
48 42 Starrett Housing Corp. 44.8% 1.11E+16 $3,000,000 95,291,000 0.02 0.01 2.50 12.4
49 43 Sudbury Holdings, Inc. 46.5% 1.39E+16 $22,325,000 33,431,000 0.65 0.17 2.96 26.6
50 44 Superior Care, Inc. 41.9% 1.32E+15 $5,660,000 50,403,000 0.21 0.93 2.77 42.2
51 45 Sym-Tek Systems, Inc. 31.6% 4.03E+14 $995,000 20,550,000 0.34 0.92 2.58 13.4
52 46 Telepictures Corp. 11.6% 5.50E+15 $15,250,000 106,849,000 0.81 0.86 2.72 6.6
53 47 Velo-Bind, Inc. 19.5% 5.51E+14 $2,325,000 18,509,000 0.65 0.85 2.81 14.5
54 48 Western Digital Corp. 47.3% 4.24E+14 $7,825,000 50,417,000 0.00 0.32 2.64 22.7
55 49 50-Off Stores, Inc. 12.5% 6.10E+15 $5,670,000 43,024,000 0.80 0.87 2.38 23.7
56 50 ARC Capital 18.8% 3.76E+14 $2,275,000 18,846,000 0.03 0.74 1.63 35.0
57 51 Dense Pac Microsystems, Inc. 23.1% 3.24E+14 $4,500,000 108,862,000 0.08 0.70 1.17 42.4
58 52 Nobel Education Dynamics, Inc. 19.3% 1.95E+15 $12,000,000 60,913,000 0.34 0.76 1.74 32.1
59 53 Unimed Pharmaceuticals 15.8% 5.49E+13 $8,400,000 44,681,000 0.09 0.74 1.90 21.0
60 Mean 27.1% 5.65E+15 $9,223,226 $78,621,472 0.42 0.69 2.54 25.4
4 Regression #1

6 Regression Statistics
7 Multiple R 0.8058
8 R square 0.6493
9 Adjusted R square 0.5947
10 Standard error 0.0873
11 Observations 53

<<

. 51
( 100 .)



>>