Estate of Hendrickson v. Commissioner

T.C. Memo. 1999-278

UNITED STATES TAX COURT

ESTATE OF JAMES WALDO HENDRICKSON, DECEASED,
MARK HART HENDRICKSON, EXECUTOR, Petitioner v.
COMMISSIONER OF INTERNAL REVENUE, Respondent

Docket No. 21225-97. Filed August 23, 1999.

Brett J. Miller and Megan J. Kight, for petitioner.

Russell D. Pinkerton, for respondent.

MEMORANDUM FINDINGS OF FACT AND OPINION

BEGHE, Judge: Respondent determined a deficiency in

petitioner's Federal estate tax of $2,465,624 and an accuracy-

related penalty of $477,113 for negligence under section

6662(b)(5).
 - 2 -

Unless otherwise noted, all section references are to the

Internal Revenue Code in effect at decedent's death, and all Rule

references are to the Tax Court Rules of Practice and Procedure.

After concessions, including respondent's concession of the

penalty, the sole issue for decision is whether, for purposes of

computing the gross estate of James Waldo Hendrickson (decedent),

the fair market value of 1,499 common shares of Peoples Trust and

Savings Bank of Boonville, Indiana (Peoples), was $4,497,000

($3,000 per share) as petitioner contends, $8,938,912 ($5,963.25

per share) as respondent contends, or some other amount. We hold

that the fair market value was $5,757,296 ($3,840.76 per share).

FINDINGS OF FACT

Decedent died testate on May 20, 1993 (the valuation date),

survived by his two sons: Mark Hart Hendrickson (Mark) and

Vinson Eric Hendrickson. Decedent and Mildred A. Hendrickson,

the mother of decedent's sons, had been divorced in 1986, after

more than 45 years of marriage.

When the petition was filed, Indiana was the residence and

principal place of business of Mark and Peoples, respectively,

the co-personal representatives of decedent's estate.1

1
Under Indiana law, the term "personal representative"
includes "executor". Ind. Code Ann. sec. 29-1-1-3 (Michie
Supp. 1998). In accordance with decedent's will, we use the
term "co-personal representative" exclusively in this case.
 - 3 -

A. Peoples Trust and Savings Bank of Boonville, Indiana

Peoples is an independent bank chartered by the State of

Indiana. It was originally chartered in 1895 and has been in

continuous operation ever since. Peoples has only one office,

located in Boonville, Indiana, which has a population of

approximately 6,000. Boonville is the county seat of Warrick

County, Indiana, and is located near the southwest corner of the

State, approximately 14 miles east of Evansville, Indiana. The

primary customer base of Peoples is Warrick County, which does

not include Evansville. Warrick County is bordered on the west

by Vanderburgh County, and on the east by Spencer County.

Decedent's branch of the Hendrickson family has played the

leading role in the management of Peoples for more than 50 years.

James W. Hendrickson (J.W.), decedent's father, began working at

Peoples in 1939, while maintaining a law practice, and eventually

became its president. At his death in 1951, J.W. owned very few

shares of stock in Peoples; decedent, who inherited fewer than 10

shares of Peoples stock from J.W., thereupon took over as

president, and served as president for 40 years, until 1991, when

Mark took over as president. Like his father, decedent was a

practicing lawyer and maintained a solo probate practice during

the years he was employed by Peoples.

In 1957, while decedent was away on vacation, members of his

extended family attempted to gain control of Peoples. In order

to thwart the attempted hostile takeover, decedent borrowed money
 - 4 -

from Old National Bank in Evansville (Old National) to finance

his acquisition of sufficient shares of Peoples stock to maintain

control. Although decedent acquired small blocks of Peoples

stock over the years, he made his major purchases in response to

the 1957 takeover attempt.

Mark began his employment with Peoples as a teller in June

1972, following his graduation from college, and was employed by

Peoples until his admission to law school in 1974. Following

graduation from law school in 1977, Mark returned part-time to

Peoples, assisting in the trust department, where he did legal

work and handled collection matters. In the mid-1980's, he

became a part-time trust officer, a position he maintained until

he became president in 1991, after decedent's health started to

deteriorate. After Mark became president, decedent assumed the

honorary title of Chief Executive Officer (CEO), which was not

provided for in the bylaws. Mark's relationship with decedent

became increasingly tense and difficult because of their

differences of opinion on how Peoples should be managed. While

decedent had managed Peoples very conservatively, Mark wanted to

adopt a more progressive approach that was favored by younger

employees of Peoples.

At the time of his death, decedent was CEO and a director of

Peoples and the owner of 1,499 shares of Peoples common stock

(the estate shares), representing 49.97 percent of the shares

outstanding.
 - 5 -

B. Balance Sheet and Capitalization

As of March 31, 1993 (the reporting date), Peoples had

reported total assets of $90.7 million, total liabilities of

$70.8 million, and stockholder's equity of $19.9 million. On the

valuation date, the capital structure of Peoples included no

debt, other than deposits and other short-term liabilities.

1. Assets

As of the reporting date, Peoples' assets consisted

primarily of net loans of $29.9 million and marketable securities

of $54.3 million. Marketable securities thus represented

approximately 60 percent of Peoples' assets, while net loans

represented approximately 33 percent.

The bulk of Peoples' investment portfolio consisted of 5-

year treasury bonds that had been purchased in the high interest

rate environment of the late 1980's and early 1990's. As of May

1993, a substantial number of those bonds were scheduled to

mature in 12 to 24 months, subjecting Peoples to interest rate

risk in the lower interest rate environment prevailing at that

time.

2. Stock

a. Control

On the valuation date, there were 3,000 shares of Peoples

common stock outstanding. The two largest shareholders were

decedent (1,499 shares) and Mildred Hendrickson (610 shares).
 - 6 -

Mark had 85 shares. The remaining shares were held by 29

shareholders, each of whom held at least 3 shares.

On the valuation date, the estate shares were 49.97 percent

of Peoples' outstanding shares, and no other shareholder held an

interest of similar size. Although the estate shares were

numerically a minority interest, they were a controlling interest

in substance. The estate shares had effective control of

Peoples, regardless of who owned them.2 There would be few

circumstances in which the estate shares would not determine the

outcome of any particular vote, because unless every other

shareholder voted against the estate shares, the estate shares

would always win. Thus, over time, the holder of the estate

shares would in all likelihood be able to determine all, or

substantially all, the members of Peoples' board of directors

(the board).3

2
Because every shareholder owned at least 3 shares, any
existing shareholder who acquired the estate shares would
automatically acquire actual control, because he or she
would acquire a majority interest (1,499 + 3 = 1,502/3,000).
3
The articles of incorporation of Peoples do not provide
for cumulative voting for directors. Although the Indiana
general corporate law permits the certificate of
incorporation to provide for cumulative voting for
directors, Ind. Code Ann. sec. 23-1-30-9(b) (Michie 1999),
the Indiana corporate law applicable to financial
institutions does not appear to permit cumulative voting.
See Ind. Code Ann. sec. 28-13-6-9 (Michie 1996).
 - 7 -

b. Stock Transactions

There was no regular market for Peoples common stock, and

transfers were infrequent. On the valuation date, Peoples did

not have an employee stock ownership plan nor any history of

repurchasing shares. Within the 24 months before and after the

valuation date, there was only one arm's-length sale of Peoples

stock. In that transaction, which occurred after the decedent's

death, in 1994, Julia Raibley, assistant secretary of Peoples,

purchased 3 shares at $800 per share from the Stone estate. In

February 1990, after being named to the board, Victor Bowden

purchased 10 shares from the Toole Estate in order to comply with

the requirement in Peoples' bylaws that members of the board own

at least 10 shares of Peoples stock. Mr. Bowden purchased the

shares for $700 per share using funds he borrowed from decedent.

Also in February 1990, the Toole Estate sold 35 Peoples shares to

Mark for $700 per share.4

In June 1990, decedent transferred by gift 2 shares of

Peoples stock to Mark, reducing his interest in Peoples from

1,501 shares (50.03 percent) to 1,499 shares (49.97 percent).

The only reason given by decedent to Mark for the gift was to

"round off" Mark's holdings from 83 shares to 85 shares. Mark

4
The fair market value per share for Peoples stock reported
on the Form 706, United States Estate (and Generation-
Skipping Transfer) Tax Return, of the Ella Wright Estate in
1989 or 1990 was also $700.
 - 8 -

and decedent had not previously discussed a control-breaking

transfer such as this one; some years earlier, when Mark had

suggested estate planning to decedent, decedent had not expressed

any interest.

Although there was no active market for Peoples shares,

after Mark became president he occasionally received informal

purchase inquiries from representatives of other banks in the

area. Peoples did not receive any purchase inquiries from

business brokers or investment banking firms.

There was not much of a market for Peoples stock, and shares

could not always be sold. For example, sometime prior to Mark's

becoming president of Peoples in August 1991, Mark and decedent

received telephone solicitation from a representative of

Hilliard-Lyons, a Louisville investment banking firm, about

purchasing shares from Charlotte Marsh. Mark and decedent both

declined. The same shares were still available for purchase in

1993 and again in 1996 or 1997, when Mark was personally

solicited by the son of Mrs. Marsh.

c. Dividend Payments

The board did not declare any dividends between 1984 and

1995. During this time period, Mrs. Hendrickson and her attorney

made a demand for dividend payments at an annual shareholders'

meeting. Mrs. Hendrickson had received her approximately 20-

percent interest in Peoples from decedent as part of her divorce

property settlement. The board apparently took no action on her
 - 9 -

demand, as no dividends were paid until 1996, when the board

declared a $4 million dividend. In 1998, the board declared a

dividend of around $1.5 to $2 million.

d. Excess Capital

As a result of not having paid dividends, Peoples, on the

valuation date, was overcapitalized, as measured by the ratio of

book equity to total assets. On the reporting date and the

valuation date, Peoples had an equity-to-asset ratio of

approximately 22 percent; at that time, the average equity-to-

asset ratio for Midwestern banks and thrifts with assets less

than $150 million was between 7 and 9 percent. On the valuation

date and the reporting date, a 9-percent equity-to-assets ratio

would have been a reasonable level of capitalization for Peoples.

On the reporting date, Peoples had equity of $19,918,000, of

which $12,919,000 was excess capital.

C. Governance and Management

As of the valuation date, Peoples was still incorporated in

the State of Indiana and was subject to applicable Indiana

corporate and banking law.

1. Shareholder Approval

The articles of incorporation and bylaws of Peoples contain

no provisions concerning shareholder voting requirements for

mergers, acquisitions, sales of assets, or liquidation.

Accordingly, under Indiana corporate law, a plan of merger or
 - 10 -

sale of substantially all the assets would have required the

approval of no more than a majority of the shares entitled to

vote. See Ind. Code Ann. sec. 23-1-40-3(e), 23-1-41-2(e) (Michie

1999); see also Ind. Code Ann. sec. 28-1-7-5 (Michie Supp.

1998).5

2. Board of Directors

On the valuation date, the board of directors comprised

eight individuals, the majority of whom were also employed by

Peoples. The bylaws of Peoples then in effect provided that two-

thirds of the board would constitute a quorum and prohibited the

transaction of any business without a quorum. The bylaws also

required all directors to own at least 10 shares of stock in

Peoples; thus, all the directors were shareholders. In addition

to decedent and Mark, the board included the following

individuals:

Name Position
Alan Bender Executive V.P., Senior Loan Officer
Alan Bennett Decedent's friend
Victor Bowden Decedent's godson
Florence Davis V.P., Trust Officer
John Farrell V.P., Loan and Security Officer
Richard Johnson Outside director

Most of the board members had been selected by the decedent

because of their longstanding, personal relationships with him,

rather than on account of any relevant expertise. Victor Bowden,

5
The provision for a vote of two-thirds of the outstanding
shares to approve a sale of assets by an Indiana financial
institution applies only to such institutions organized
after Dec. 31, 1992. See Ind. Code Ann. 28-1-8-4 (Michie
1996).
 - 11 -

for example, was a young man whom decedent had supported much

like a foster child; John Farrell was a friend from military

service in World War II; and Alan Bender used to drive a laundry

truck and had the decedent on his route. None of the members of

the board had any formal training in banking,6 and only half the

members of the board had attended college. Despite the close

relationships of the other board members to decedent, however,

they were not a rubber stamp for decedent and Mark. The other

directors were strong-willed independent thinkers who contributed

to the board's deliberations. Moreover, those directors who were

also officers could use their positions as employees to subvert

any board decision with which they were in disagreement. But,

while the board was not a rubber stamp, many of the directors

thought similarly to decedent and were resistant to change.

3. Financial Management and Reporting

a. Outside Accountant

Peoples' longstanding regular accountant and auditor was

Ronnie Robinson, a sole practitioner in Evansville. Mr. Robinson

had no other financial institutions as clients.

6
Mark completed graduate banking training at the University
of Wisconsin in 1996. After the valuation date, Tony
Aylsworth, chief operating officer of Peoples since February
1998 (but not a member of the board), also completed
graduate banking training at the University of Colorado,
Boulder.
 - 12 -

b. Budgeting

Peoples did not prepare budgets until 1991, after being

forced to do so by the FDIC. The budgets were prepared one

quarter at a time by Julia Raibley.

c. Earnings

In the years 1991-93, Peoples enjoyed an unusually high

yield curve, because of the spread between the rates of interest

received on securities and loans over interest paid on savings

accounts and certificates of deposit.

By the time of decedent's death, Peoples had begun

performing interest rate sensitivity analyses; however, its

methods of doing so were criticized as unreliable by the FDIC

examiners.

Peoples did not follow the practice of forecasting its

earnings.

D. Operations

1. Employees

On the valuation date, Peoples had approximately 30

employees, most of whom were in their mid-50's or older. Peoples

had no retirement plan, but it also had no mandatory retirement

age. As a result, Peoples had employees in their 70's and 80's.

In general, Peoples had a good relationship with its employees,

resulting in very little turnover and a median term of employment

of 17-1/2 years. However, there was some friction between
 - 13 -

tellers and bookkeepers on the one hand, and loan officers on the

other, resulting in little mingling between the two groups.

2. Facilities

The operations of Peoples were housed in a two-story

building built in 1964; Peoples had no branches. Peoples'

facilities and equipment were in average condition and considered

adequate for its operations.

When Mark became president, Peoples was for the most part

operating with 1960's technology. Most operations in the bank

were done manually, and there were no information systems to keep

track of customers' credit, payments, or deposits. Peoples did

not have any ATM machines. Peoples had desktop PC's for use by

loan officers and customer service representatives, but only to

print out loan applications, signature cards, and other

operational documents. Under Mark's direction, however, Peoples

had entered into a contract with Old National to provide

electronic data processing (EDP) applications for deposits,

loans, and accounting. Although many of the directors were

hesitant about learning the new technology, they reluctantly

cooperated. During 1991 and 1992, Peoples converted its system

of accounting for its loan portfolio from a manual system to a

computerized system and began having its loans serviced by Old

National; however, on the valuation date, the complete conversion

of Peoples to the Old National EDP applications had not been

completed.
 - 14 -

Security at Peoples was antiquated. Despite a number of

bank robberies in the area, Peoples had little protection from

armed robbery or burglary. Security at the bank consisted of

glass doors that could be locked and two 1960's vintage still

image security cameras. Peoples did not employ security guards

or arm its employees and did not have a central alarm system. In

the event of a robbery, Peoples did have a system in place to

activate a silent alarm that would notify the sheriff if bait

money should be removed from a teller bay. However, at the time

of decedent's death, the physical cash was not carefully managed,

and tellers frequently kept as much as $80,000 in their cash

drawers.

3. Franklin Loan and Savings Association

On the valuation date, the Franklin Loan and Savings

Association (Franklin) was affiliated with Peoples. Franklin did

not have its own facilities or employees and used those of

Peoples. Franklin paid Peoples for employee time used, along

with an annual rent of $600.

4. Depository Accounts

a. Checking Accounts

On the valuation date, Peoples offered "free" checking

accounts that really were free. There were no monthly charges,7

7
Peoples charged a fee on dormant accounts as provided by
State law.
 - 15 -

minimum balances, or charges for transactions. In fact, until

shortly before decedent's death, Peoples provided checking

account customers with free printed blank checks; even after

Peoples decided to charge for checks, it charged customers only

for actual printing costs with no markup. As a result of

Peoples' generous checking account policy, there were

approximately 4,000 checking accounts open at Peoples on the

valuation date; however, approximately 1,200 of those accounts

had an average daily balance of less than $200 and total checking

account deposits amounted only to approximately $1 million. As a

result, most checking accounts were unprofitable for Peoples,

inasmuch as the costs of servicing the accounts exceeded earnings

on the deposits.

b. Savings Accounts

On the valuation date, Peoples had $7 to $8 million in

deposits in savings accounts.

c. Certificates of Deposit

The overwhelming majority of Peoples' deposits were

attributable to certificates of deposit. Those deposits included

"hot money", large deposits that come from a wide geographic area

in search of the best interest rate. Hot money is a volatile

source of deposits because it is likely to be withdrawn anytime a

better interest rate is being offered by another depository

institution. Peoples was not interested in pursuing or retaining

hot money, because, as discussed infra, it did not have
 - 16 -

sufficient loan demand to pay the yields demanded by hot money

depositors.

5. Lending Activities

a. Loan Products

As of the reporting date, Peoples' loan portfolio was

heavily concentrated in real estate. Over 90 percent of Peoples'

loan portfolio was in real estate loans, of which 76.59 percent

were loans on 1 to 4 family residential real estate. Most real

estate loans made by Peoples were for existing properties, rather

than new construction. Peoples' concentration of loans in 1 to 4

family residential real estate placed it in the 98th percentile

in comparison to its peer institutions.

Peoples offered few choices in the way of loan products to

real estate borrowers. It did not offer adjustable rate

mortgages8 (ARM's) or home equity loans9 and did not participate

in any FHA or VA mortgage programs. Fixed-rate mortgages were

offered only with a 15- or 20-year term; a 30-year term, often

favored by first-time home buyers because of the lower monthly

payments, was not available. Also discouraging to first-time

home buyers, Peoples' loan-to-value (LTV) and loan limit

8
Peoples did not have sufficient information technology to
comply with regulations concerning the disclosures required
to be made to consumers with respect to adjustable rate
mortgage borrowers.

9
Peoples now offers home equity loans.
 - 17 -

restrictions required significantly greater downpayments than

competing lenders. While its competitors were lending on an 80-

percent LTV, or as high as 90 to 95 percent with private mortgage

insurance (PMI), Peoples required an LTV ratio of 70 percent.

Although Peoples raised its LTV requirement to 80 percent in

1993, discussed infra, it did not offer any programs using PMI to

lower the borrower's downpayment. Finally, Peoples' $250,000

lending limit required proportionately larger downpayments on

more expensive homes than competitors. With a maximum mortgage

loan of $250,000, any house with a purchase price of more than

$312,500 would require a downpayment greater than 20 percent.

Peoples generally avoided making other types of consumer

loans, such as credit cards, automobile leasing, and automobile

financing. Although it technically offered automobile financing,

Peoples set rates above market because it was not interested in

making automobile loans, due to concerns over whether it had

sufficient personnel to track automobile documentation

(insurance, titles, etc.) and deal with collections. Such

automobile loans as were made were mainly to Peoples' employees.

As of the reporting date, only 2.13 percent of the Peoples

loan portfolio was in commercial and industrial loans, placing

Peoples in the 4th percentile (very low) in comparison to its

peers. Peoples generally did not make commercial loans, had only

one revolving line of credit open, and did not offer letters of

credit. After Mark became president, and before decedent's
 - 18 -

death, Peoples tried to start a Small Business Administration

(SBA) loan program, but was unsuccessful in submitting loan

applications to the SBA because of inadequate documentation.

In sum, although Peoples' charter permitted it to offer a

broad range of lending products to consumer and commercial

borrowers, Peoples in practice generally failed to serve all but

a select class of home buyers who were not discouraged by

Peoples' LTV, loan limit, 20-year maximum term, and lack of

ARM's.

b. Loan-To-Asset Ratio

The effects of Peoples' narrowly defined market and

conservative lending practices are readily apparent. During the

4-year period ending on the valuation date, Peoples' net loans as

a percentage of assets averaged 30 percent. Peoples' peers, in

comparison, averaged 52 to 55 percent. After Mark became

president of Peoples, he tried to raise Peoples' loan-to-asset

ratio significantly and set a goal of 60 to 70 percent for 1993,

a rate that had not been achieved as of the date of trial. Aside

from adopting a more aggressive stance in the market, Peoples

could also have achieved an increased loan-to-asset ratio by

reducing its assets by declaring a substantial dividend.

On March 1, 1992, Joe Melhiser was hired as an assistant

vice president from another bank in an effort to increase

lending. As part of the effort to increase lending, Peoples

raised its LTV to 80 percent in 1993. Although the increased LTV
 - 19 -

triggered concerns at Peoples over increased risks of default or

delinquency, Peoples did not raise its interest rates.

c. Loan Underwriting and Documentation

Loan decisions were made by a four-person loan committee on

the basis of a loan application. Peoples' underwriting

procedures were criticized by the FDIC for a number of

shortcomings. Until early 1993, for example, Peoples did not

require credit reports or title insurance for residential

mortgage loans and did not check whether flood insurance was

required on the property being financed.10 Contrary to standard

industry practice, Peoples made loans without verifying the

market value of the underlying collateral through independent

appraisal. Instead, Peoples used less reliable in-house

appraisals that did not conform to the format used by independent

appraisers. In some cases, Peoples even recycled old appraisals

of a property, rather than obtaining a new appraisal that would

reflect current market values.

Unlike most mortgage lenders, Peoples did not sell any

mortgage loans in the secondary mortgage market. Peoples had

unsuccessfully tried to sell mortgage loans in the past to Fannie

Mae and Freddie Mac but did not have underwriting practices and

documentation sufficient to comply with the standards of the

10
There was a high likelihood of flooding in some areas
served by Peoples; the Ohio River is the southern boundary
of Warrick County.
 - 20 -

secondary market. In any event, Peoples needed to retain

ownership of the mortgage loans it made, because its supply of

funds exceeded the demand for its mortgage loans--selling the

mortgages would have further compromised Peoples' net interest

margin and earnings.

d. Loan Monitoring

Peoples monitored loans using paper ledger cards that were

stored in pockets that tracked the day of the month on which each

loan was due, so that if the loan was past due, the card would

remain in what would become a "past due" pocket instead of being

put in a current pocket.

Meetings concerning delinquent loans were held by Peoples'

loan committee. Peoples also had watch lists of problem loans as

required by the FDIC examiners, but they were not used by Peoples

to monitor loans.

6. Fee Income

Income from fees can make a significant contribution to the

income of a bank. Banks may earn income or fees from points and

origination fees on loans, ATM fees, trust fees, credit card

fees, servicing agreements, and insurance sales. Peoples

generally charged no points or fees, however, and had only

minimal fee income from its activities. Peoples did not service

loans made by other depository institutions.
 - 21 -

E. Regulation and Monitoring

1. Classification, Regulation, and Insurance

The banking industry includes both commercial banks and

thrifts. While there is some overlap, commercial banks generally

provide a wider range of banking services than thrifts, whose

principal mission is financing home ownership. There are also

some differences in ownership. Commercial banks are always owned

as stock companies, while thrifts may be stock-owned or mutually

owned. Commercial banks and stock-owned thrifts are frequently

owned by a bank holding company owning one or more institutions.

Although most thrifts are mutually owned, stock companies account

for more than two-thirds of the assets of thrifts.

Both banks and thrifts can be federally or State chartered

and are generally insured by either the Bank Insurance Fund (BIF)

or Savings Association Insurance Fund (SAIF) of the Federal

Deposit Insurance Corporation (FDIC). BIF insures all federally-

chartered commercial banks, most State-chartered commercial

banks, and some State-chartered thrifts. SAIF insures all

federally-chartered and some State-chartered, thrifts. Since

July 1, 1991, SAIF- and BIF-insured institutions have paid the

same level of deposit insurance premium (0.23 percent of

deposits).

On the valuation date, Peoples was chartered by the State of

Indiana with the powers of a commercial bank and was BIF-insured.

As a result, Peoples was subject to regulation at the Federal
 - 22 -

level by the FDIC and at the State level by the Department of

Financial Institutions of the State of Indiana (IDFI). The FDIC

and the IDFI share regulatory responsibilities and exchange

documents, reports and examinations, and correspondence

concerning the institutions that they both supervise and

regulate.

Although Peoples was chartered with the powers of a

commercial bank, it displayed the characteristics of a thrift.

Its loan portfolio was heavily concentrated in residential real

estate loans that were held until maturity; Peoples did not sell

or service loans. The asset structure of Peoples was heavily

weighted toward investment securities, and the loan-to-deposit

ratio was below 50 percent. Despite the powers granted under its

charter, Peoples was not equipped to operate as a commercial bank

without significant expenditures to update its systems and

procedures to process commercial lending and deposit products.

Accordingly, for purposes of valuing the estate shares, Peoples

should be treated as a thrift, rather than a commercial bank.

2. Safety and Soundness

Both FDIC and IDFI conduct regular compliance and safety and

soundness examinations of Peoples using a composite ratings

system based on six areas of concern: (1) Capital adequacy,

(2) asset quality, (3) management competency, (4) earnings level

and trend, (5) level of liquidity, and (6) interest rate

sensitivity. FDIC conducted examinations of Peoples as of the
 - 23 -

close of business November 10, 1990, and December 11, 1992; IDFI

conducted an examination as of March 14, 1992. In all three

cases, Peoples received a uniform composite rating of one (1),

the highest possible rating. Institutions earning a uniform

composite rating of one (1) are basically sound in every respect

and are considered to be resistant to external economic and

financial disturbances and more capable of withstanding the

vagaries of business conditions than institutions with lower

ratings. Nevertheless, in their reports with respect to the 1992

examinations, both FDIC and IDFI expressed concern over Peoples'

liability sensitive position and cautioned that earnings could be

negatively impacted during a period of rising interest rates.

The FDIC also expressed concern over shortcomings in various

policies and procedures of Peoples, including financial

reporting, underwriting, and budgeting, although apparently not

of sufficient magnitude to affect Peoples' top rating.

3. Community Reinvestment Act

The Community Reinvestment Act (CRA), title VII of the

Housing and Community Development Act of 1977, Pub. L. 95-123,

sec. 802, 91 Stat. 1147, currently codified at 12 U.S.C. sec.

2901 (1994), was enacted by Congress to encourage depository

institutions to help meet the credit needs of the communities in

which they operate, including low- and moderate-income

neighborhoods, consistent with safe and sound banking operations.

CRA requires that each insured depository institution’s record in
 - 24 -

helping meet the credit needs of its entire community be

evaluated periodically. That record is taken into account in

considering an institution’s application for deposit facilities,

including mergers and acquisitions. CRA examinations are

conducted by the Federal agencies that are responsible for

supervising depository institutions, such as FDIC and the Office

of Thrift Supervision (OTS). Inasmuch as Peoples was insured by

FDIC, it was subject to FDIC examination with respect to its CRA

compliance. Peoples' designated CRA area included all of Warrick

County. Peoples reduced the size of its CRA area in August 1992

at the suggestion of an FDIC examiner because of its compliance

difficulties. If Evansville had been included in Peoples'

designated CRA area, Peoples would have had significantly greater

difficulty complying, because it would have had to prove it was

fairly serving Evansville's low- and moderate-income

neighborhoods.

Peoples received CRA performance ratings of "needs to

improve" on two occasions. In its CRA Performance Report as of

the close of business on November 10, 1990 (first CRA report),

FDIC cited three principal CRA compliance issues: (1) Lack of

adequate effort by Peoples to determine the credit needs of the

community, (2) lack of adequate effort by Peoples to publicize

the types of loans it offered, and (3) generally conservative

lending practices. In a memorandum of understanding dated

March 13, 1991 (first MOU), and signed by all directors of
 - 25 -

Peoples, Peoples agreed to address the CRA compliance issues

raised by the FDIC in the first CRA report.

In a CRA Performance Report as of the close of business

August 31, 1992 (second CRA report), FDIC identified additional

CRA compliance problems, including: (1) Lack of formal training

by Peoples of its employees on CRA compliance, (2) failure by

Peoples to monitor its own performance in complying with the CRA,

(3) failure by Peoples to review its lending patterns for

evidence of discriminatory lending practices, (4) loan-to-deposit

ratio, (5) low volume of farm and business credit extended, and

(6) lobby hours significantly below that of the competition. In

response to the second CRA report, the directors of Peoples

signed another Memorandum of Understanding on October 7, 1992

(second MOU).

Within a month after signing the second MOU, in an attempt

to improve its CRA compliance, Peoples created a new position--

compliance officer--and hired Thomas Krochta, a local attorney,

to do the job. Nevertheless, on December 31, 1992, as Mr.

Krochta was settling into his new position, Simona L. Frank,

Chicago regional director of FDIC's Division of Supervision,

wrote to the board expressing concern over Peoples' continued

noncompliance and requesting immediate corrective action. Ms.

Frank said that she planned to recommend to the national office

of the Division of Supervision that Peoples continue to be

designated a "problem bank" under the CRA. As a result of the
 - 26 -

"problem bank" designation, Peoples was subjected to a very high

level of scrutiny in future examinations by the FDIC;

examinations were conducted more frequently; greater

correspondence between Peoples and the FDIC was required; and

Peoples could have been prohibited from any merger, acquisition,

divestiture, or expansion activity. Designation as a "problem

bank" also excluded Peoples from bidding on the assets of failed

financial institutions. Continued failure to comply with the CRA

could have resulted in a cease-and-desist order by the FDIC, with

the ultimate sanction of forced closure.

Peoples' CRA compliance costs were significant for a bank

its size. In addition to the additional demands on management

imposed by the CRA requirements, Peoples was forced to hire two

full-time attorneys to oversee its CRA compliance. In February

1993, shortly after hiring Mr. Krochta, Peoples hired Tony

Aylsworth, another local attorney as an assistant vice president

and assistant compliance officer. Mr. Aylsworth's first

assignment at Peoples was "scrubbing" the loan files.

"Scrubbing" as used in this context, refers to the process of

reviewing loan files to see if they contain adequate

documentation, and correcting any documentation problems

discovered, such as missing documents or signatures. Inadequate

loan documentation was a contributing factor in Peoples' CRA

compliance difficulties, because Peoples was unable to show why

and where it was denying certain loans. Until decedent's death
 - 27 -

on May 20, 1993, he worked closely with Mr. Aylsworth on the

"scrubbing" project.

On July 22, 1993, the FDIC notified Peoples that it had

demonstrated compliance with only about half of the terms and

conditions of the second MOU. Not until early 1994 did Peoples

finally achieve compliance with the second MOU. In October 1994,

Mr. Aylsworth ceded his position as assistant compliance officer

and became a trust officer, a position he held until February

1998, when he became chief operating officer.

4. Third-Party Monitoring

Peoples was highly regarded by regulators and the banking

community for its safety and soundness and was recognized by the

following bank rating services:

Rating service Rating

Veribank Inc. Blue ribbon bank

Bauer Financial Reports Five star bank

Sheshunoff Information Listed in highest rated
Services banks in America

F. Market Conditions

1. Banking Industry

In 1993, the banking industry was still recovering from the

recession of the early 1990's, and--in the case of thrifts--from

the savings and loan crisis. The first quarter of 1993 was the

first time since 1989, when the savings and loan industry bailout

was introduced, that no savings banks had failed. Prospects for
 - 28 -

the industry were encouraging as a result of low interest and

inflation rates that led to increased consumer borrowing and

spending, increased production and expansion, and lower

unemployment.

2. Competition

Peoples’ main competitors in Boonville were Old National,

Warrick Federal Credit Union (the Credit Union), and Boonville

Federal Savings Bank. Old National and the Credit Union both had

competitive positions superior to that of Peoples. Among its

Boonville competitors, Peoples was considered to be something of

a dinosaur. Peoples had been losing customers since the Credit

Union opened. In comparison to its competitors, Peoples' only

advantage was in the low costs and fees it offered on home

mortgages, along with a competitive interest rate.

Peoples also faced competition from large commercial banks

in Evansville that had presences in other parts of southern

Indiana: (1) Old National; (2) CNB Bancshares; and (3) National

City Bancshares. However, all three Evansville banks were trying

to grow through acquisition, so while they may have posed a

competitive threat, they were also a potential acquirer of

Peoples. Old National's Boonville branch, for example, had

resulted from Old National's acquisition of a local bank.
 - 29 -

3. Local Economy

On the valuation date, the local economy served by Peoples

was growing slowly. In Boonville, the largest employer was the

School Corporation. In Warrick County, the two largest

employers, Alcoa and Peabody Coal, were both significantly

downsizing. Alcoa, which previously employed 3,500 to 4,000

local workers, was reducing its local workforce by one-third,

while Peabody Coal was reducing its local workforce by two-

thirds, to fewer than 200 employees. Another coal mining

concern, Amax, had ceased local mining operations entirely. The

area served by Peoples also contained a Whirlpool manufacturing

plant and some farming activity. Despite the negative

developments in the Warrick County economy, many residents of

Warrick County worked in Evansville, which was experiencing

modest growth.

On April 16, 1987, the U.S. District Court for the Southern

District of Indiana entered a consent judgment in an action filed

by the U.S. Environmental Protection Agency (EPA) against the

State of Indiana and the City of Boonville. EPA had initiated

the action because Boonville's sewage treatment facilities were

overloaded, resulting in the discharge of insufficiently treated

wastewater, in violation of the Clean Water Act11 and other

Federal laws. The consent judgement required Boonville to

11
Water Quality Act of 1987, Pub. L. 100-4, sec. 1(a), 101
Stat. 7, currently codified as 33 U.S.C. sec. 1251 (1994).
 - 30 -

prohibit "all new sewer connections" (sewer tap ban) until it

could demonstrate compliance with all remedial provisions of the

consent judgment. EPA could grant waivers from the sewer tap

only under very limited circumstances:

Waivers from this sewer ban for new sources may be
granted * * * only if EPA determines that the proposed
connection will eliminate an existing health hazard and
the resulting public health benefit outweighs the
adverse impact of any reduction in wastewater effluent
quality.

The sewer tap ban made it very difficult to obtain commercial or

residential building permits in Boonville and was considered a

high priority issue by the Boonville Board of Public Works,12 but

it did not bar new construction elsewhere in Warrick County. On

the valuation date, Newburgh, a suburb of Evansville, was one

area in Warrick County where new home construction was active,

primarily single-family homes priced between $300,000 to

$400,000. However, because of its conservative lending practices

(described supra pp. 16-19), Peoples could not serve the needs of

younger cash-poor borrowers who could otherwise afford to

purchase a $300,000 to $400,000 house.

Although Peoples was located only 14 miles from Evansville,

Peoples generally chose not to engage in lending activities in

Evansville, or its surrounding county, Vanderburgh County.

12
Tony Aylsworth, an officer of Peoples since February
1993, served on the Boonville three-person Board of Public
Works from January 1996 until February 1998.
 - 31 -

Peoples avoided making loans in Vanderburgh County because its

conservative-minded loan officers were unfamiliar with the area

and its values and because adding Evansville to Peoples'

designated CRA area would have increased Peoples' CRA compliance

requirements.

G. Estate Tax Return

Decedent's co-personal representatives executed and timely

filed a United States Estate (and Generation-Skipping Transfer)

Tax Return, Form 706, on August 8, 1994.13 The Form 706 had been

prepared by Jeffrey B. Baker, a certified public accountant and

certified financial planner. Decedent's 1,499 shares of Peoples

common stock were included on Schedule B of the Form 706. Of the

1,499 shares, the Form 706 reported 1,486 shares at a value of

$3,159,726 ($2,126 per share), and 13 shares at a value of

$10,400 ($800 per share).

The $2,126-per-share value reported on the Form 706 was

based on a valuation of the estate shares by Harding Shymanski &

Company, P.C. (HSC), an Evansville, Indiana, public accounting

firm. HSC valued the Peoples' total equity on a minority basis

using the weighted average value under three different valuation

methods. The three methods and their weightings were: (1)

Capitalized earnings (30 percent), (2) price/earnings multiple

13
Petitioner requested and received a timely extension of
time for filing the Form 706.
 - 32 -

(30 percent), and (3) price/book multiple (40 percent). After

computing the weighted average value of total equity on a

minority basis, HSC applied a 30-percent marketability discount,

and divided the result by 3,000, the total number of shares of

Peoples common stock outstanding, to determine the fair market

value of the estate shares on a per share basis. Petitioner did

not use HSC as an expert witness in this case.

On July 25, 1997, respondent timely mailed petitioner a

notice of deficiency with respect to its estate tax liability.

The notice of deficiency determined a value of $8,938,91214 for

the estate shares, based upon an appraisal prepared by David F.

Fuller of Business Valuation Services, Inc., who acted as

respondent's expert witness in this case.

H. Lack of Marketability

On the valuation date: (1) Peoples had few opportunities

for growth; (2) Peoples' earnings were subject to significant

interest rate risk; (3) Peoples had no employee stock option plan

or history of repurchasing shares; and (4) there was no readily

available public or private market for Peoples stock. Each of

these conditions contributed to a lack of marketability of

Peoples stock.

14
The redetermined value of the estate shares in the notice
of deficiency and the amount asserted at trial by respondent
are essentially the same; they differ only because
respondent's expert rounded certain figures in his report.
See discussion infra of respondent's expert.
 - 33 -

ULTIMATE FINDING OF FACT
On the valuation date, the fair market value of the estate

shares was $5,757,296 ($3,840.76 per share).

OPINION

The issue for decision is the fair market value on the

valuation date (May 20, 1993) of 1,499 shares of Peoples common

stock included in decedent's gross estate.

Valuation is a question of fact, and the trier of fact must

weigh all relevant evidence to draw the appropriate inferences.

See Commissioner v. Scottish Am. Inv. Co., 323 U.S. 119, 123-125

(1944); Helvering v. National Grocery Co., 304 U.S. 282, 294-295

(1938); Anderson v. Commissioner, 250 F.2d 242, 249 (5th Cir.

1957), affg. in part and remanding in part on another ground T.C.

Memo. 1956-178; Estate of Newhouse v. Commissioner, 94 T.C. 193,

217 (1990); Skripak v. Commissioner, 84 T.C. 285, 320 (1985).

Fair market value is defined for Federal estate and gift tax

purposes as the price that a willing buyer would pay a willing

seller, both having reasonable knowledge of all the relevant

facts and neither being under compulsion to buy or to sell. See

United States v. Cartwright, 411 U.S. 546, 551 (1973) (citing

sec. 20.2031-1(b), Estate Tax Regs.); see also Snyder v.

Commissioner, 93 T.C. 529, 539 (1989); Estate of Hall v.

Commissioner, 92 T.C. 312, 335 (1989). The willing buyer and the

willing seller are hypothetical persons, rather than specific

individuals or entities, and the peculiar characteristics of
 - 34 -

these hypothetical persons are not necessarily the same as the

individual characteristics of an actual seller or an actual

buyer. See Estate of Curry v. United States, 706 F.2d 1424,

1428-1429, 1431 (7th Cir. 1983); Estate of Bright v. United

States, 658 F.2d 999, 1005-1006 (5th Cir. 1981); Estate of

Newhouse v. Commissioner, supra at 218; see also Estate of Watts

v. Commissioner, 823 F.2d 483, 486 (11th Cir. 1987), affg. T.C.

Memo. 1985-595. The hypothetical willing buyer and willing

seller are presumed to be dedicated to achieving the maximum

economic advantage. See Estate of Curry v. United States, supra

at 1428; Estate of Newhouse v. Commissioner, supra at 218. This

advantage must be achieved in the context of market and economic

conditions at the valuation date. See Estate of Newhouse v.

Commissioner, supra at 218.

For Federal estate tax purposes, the fair market value of

the subject property is generally determined as of the date of

death of the decedent; ordinarily, no consideration is given to

any unforeseeable future event that may have affected the value

of the subject property on some later date. See sec. 20.2031-

1(b), Estate Tax Regs.; see also First Natl. Bank v. United

States, 763 F.2d 891, 893-894 (7th Cir. 1985); Estate of Newhouse

v. Commissioner, supra at 218; Estate of Gilford v. Commissioner,

88 T.C. 38, 52 (1987).

Special rules apply to the valuation of the stock of a

closely held corporation. While listed market prices are the
 - 35 -

benchmark in the case of publicly traded stock, recent arm’s-

length transactions generally are the best evidence of fair

market value in the case of unlisted stock. See Estate of

Andrews v. Commissioner, 79 T.C. 938, 940 (1982); Duncan Indus.,

Inc. v. Commissioner, 73 T.C. 266, 276 (1979); Estate of Branson

v. Commissioner, T.C. Memo. 1999-231. Where the value of

unlisted stock cannot be determined from actual sale prices,

value is best determined by taking into consideration the value

of listed stock in comparable corporations engaged in the same or

a similar line of business, as well as all other factors bearing

on value, including analysis of fundamentals. See sec. 2031(b);

Estate of Newhouse v. Commissioner, supra at 217; Estate of Hall

v. Commissioner, supra at 336. The factors that we must consider

are those that an informed buyer and an informed seller would

take into account. See Hamm v. Commissioner, 325 F.2d 934, 940

(8th Cir. 1963), affg. T.C. Memo. 1961-347. Rev. Rul. 59-60,

1959-1 C.B. 237, "has been widely accepted as setting forth the

appropriate criteria to consider in determining fair market

value", Estate of Newhouse v. Commissioner, supra at 217; it

lists the following factors to be considered, which are virtually

identical to those listed in section 20.2031-2(f), Estate Tax

Regs.:

(a) The nature of the business and the history of the
enterprise from its inception.

(b) The economic outlook in general and the condition
and outlook of the specific industry in particular.
 - 36 -

(c) The book value of the stock and the financial
condition of the business.

(d) The earning capacity of the company.

(e) The dividend-paying capacity.

(f) Whether or not the enterprise has goodwill or
other intangible value.

(g) Sales of the stock and the size of the block of
stock to be valued.

(h) The market price of stocks of corporations
engaged in the same or a similar line of business
having their stocks actively traded in a free and open
market, either on an exchange or over-the-counter.
[Rev. Rul. 59-60, 1959-1 C.B. at 238-239.]

As is customary in valuation cases, the parties rely

primarily on expert opinion evidence to support their contrary

valuation positions. We evaluate the opinions of experts in

light of the demonstrated qualifications of each expert and all

other evidence in the record. See Anderson v. Commissioner,

supra; Parker v. Commissioner, 86 T.C. 547, 561 (1986). We have

broad discretion to evaluate "'the overall cogency of each

expert's analysis.'" Sammons v. Commissioner, 838 F.2d 330, 334

(9th Cir. 1988) (quoting Ebben v. Commissioner, 783 F.2d 906, 909

(9th Cir. 1986), affg. in part and revg. in part T.C. Memo. 1983-

200), affg. in part and revg. in part on another ground T.C.

Memo. 1986-318. Expert testimony sometimes aids the Court in

determining values; sometimes it does not, particularly when the

expert is merely an advocate for the position argued by one of

the parties. See, e.g., Estate of Halas v. Commissioner, 94 T.C.
 - 37 -

570, 577 (1990); Laureys v. Commissioner, 92 T.C. 101, 129

(1989). We are not bound by the formulas and opinions proffered

by an expert witness and will accept or reject expert testimony

in the exercise of sound judgment. See Helvering v. National

Grocery Co., 304 U.S. at 295; Anderson v. Commissioner, 250 F.2d

at 249; Estate of Newhouse v. Commissioner, 94 T.C. at 217;

Estate of Hall v. Commissioner, 92 T.C. at 338. Where necessary,

we may reach a determination of value based on our own

examination of the evidence in the record. See Lukens v.

Commissioner, 945 F.2d 92, 96 (5th Cir. 1991) (citing Silverman

v. Commissioner, 538 F.2d 927, 933 (2d Cir. 1976), affg. T.C.

Memo. 1974-285); Ames v. Commissioner, T.C. Memo. 1990-87, affd.

without published opinion 937 F.2d 616 (10th Cir. 1991). Where

experts offer divergent estimates of fair market value, we decide

what weight to give these estimates by examining the factors they

used in arriving at their conclusions. See Casey v.

Commissioner, 38 T.C. 357, 381 (1962). We have broad discretion

in selecting valuation methods, see Estate of O'Connell v.

Commissioner, 640 F.2d 249, 251 (9th Cir. 1981), affg. on this

issue and revg. in part T.C. Memo. 1978-191, and in determining

the weight to be given the facts in reaching our conclusion

because “finding market value is, after all, something for

judgment, experience, and reason”, Colonial Fabrics, Inc. v.

Commissioner, 202 F.2d 105, 107 (2d Cir. 1953), affg. a

Memorandum Opinion of this Court. Moreover, while we may accept

the opinion of an expert in its entirety, Buffalo Tool & Die
 - 38 -

Manufacturing Co. v. Commissioner, 74 T.C. 441, 452 (1980), we

may be selective in the use of any part of such opinion, or

reject the opinion in its entirety, Parker v. Commissioner, supra

at 561. Finally, because valuation necessarily results in an

approximation, the figure at which we arrive need not be directly

attributable to specific testimony if it is within the range of

values that may properly be arrived at from consideration of all

the evidence. See Silverman v. Commissioner, supra at 933;

Alvary v. United States, 302 F.2d 790, 795 (2d Cir. 1962).

1. Respondent's Expert

Respondent relies on the expert report of David N. Fuller, a

principal in the Dallas, Texas, office of Business Valuation

Services, Inc. (BVS). Mr. Fuller has worked in business

valuation since he graduated from Southern Methodist University

in 1989 with an M.B.A. in finance. He was a Manager in the

Valuation Group at Deloitte & Touche from 1989 until 1992, when

he became associated with BVS. Mr. Fuller is an Accredited

Senior Appraiser and Chartered Financial Analyst. Mr. Fuller

valued the estate shares at $8,939,000 ($5,963.25 per share)

using the weighted average value of the estate shares under an

income approach and a market approach, reduced by a marketability

discount of 10 percent.
 - 39 -

a. Discounted Cash-Flow

The income approach employed by Mr. Fuller was the

discounted cash-flow method (DCF). A DCF analysis attempts to

measure value by forecasting a firm's ability to generate cash

and discounting the flows to present value using the firm’s cost

of capital. There are three components to the DCF analysis: (1)

The cash-flow projections over the forecasted period; (2) the

terminal value; and (3) the appropriate discount rate. Using

DCF, a firm's value is calculated as the discounted present value

of the forecasted cash-flow from operations plus the discounted

present value of the terminal value. See Brealey & Myers,

Principles of Corporate Finance 30, 64, G4 (4th ed. 1991).

Before performing his DCF analysis, Mr. Fuller reduced the

operating assets shown on Peoples' balance sheet in order to

project Peoples' free cash-flow from operations (FCF). Mr.

Fuller made these adjustments because he considered Peoples to be

overcapitalized, as measured by its ratio of book equity to

assets. With total equity of $19,918,000 and total assets of

$90,689,000 on the reporting date, Peoples had a book equity-to-

assets ratio of 22 percent.15 In comparison, according to Mr.

15
The pro forma balance sheet prepared by Mr. Fuller as of
May 20, 1993, showed total equity of $20,772,000 and total
assets of $94,948,000, resulting in a similar book equity-
to-assets ratio.
 - 40 -

Fuller, the average of Peoples' "peer group"16 was only 7.9

percent. Mr. Fuller considered a 9-percent book equity-to-assets

ratio to be a reasonable level of capitalization for Peoples,17

and, accordingly, he reduced the balances of Peoples' total

assets and total equity accounts by $12,919,000--the amount

necessary to lower Peoples' book equity-to-asset ratio to 9

percent, resulting in adjusted total equity and total assets of

$6,999,000 and $77,770,000, respectively. Mr. Fuller then

treated the $12,919,000 in assets removed in his adjustments as

16
The depository institutions comprising Peoples' peer
group are determined by The Federal Financial Institutions
Examination Council (FFIEC), an entity established by
Congress in 1978 to promote consistent examination and
supervision of financial institutions. Members in the FFIEC
include the Comptroller of the Currency, the Chair of the
FDIC, and a member of the Federal Reserve Board of
Governors.

Peer group data is used by the FFIEC in Uniform Bank
Performance Reports, which are issued by the FFIEC for every
insured bank on a quarterly basis. The FFIEC assigns each
bank or holding company to a particular peer group based
upon asset size and number of branches or banks. Mercer,
Valuing Financial Institutions 61, 143 (1992). Peoples'
peer group consisted of 43 commercial banks and 6 thrifts
operating in Indiana, Illinois, Ohio, and Kentucky, with
average total assets of approximately $82 million.

17
We note that the 9-percent figure used by Mr. Fuller
was very close to the average equity-to-assets ratios for
the guideline companies selected by petitioner's expert,
James E. Magee, of Alex Sheshunoff & Co. Investment Banking,
discussed infra. Mr. Magee used two groups of guideline
company data: One that was based on controlling interest
transactions, and one that was based on minority interest
transactions. The average equity-to-asset ratios for the
two groups were 9.05 percent and 8.57 percent, respectively.
 - 41 -

nonoperating assets, whose income would not be included in

Peoples' FCF.

Mr. Fuller forecasted Peoples' FCF for 5 years forward (the

valuation horizon), and computed a terminal value using the

Gordon dividend growth model (Gordon model). The Gordon model is

a model for estimating the terminal value of a going concern,

which assumes that FCF will continue indefinitely and grow at a

constant rate. For purposes of computing the terminal value

under the Gordon model, Mr. Fuller assumed that Peoples' FCF

would continue indefinitely, growing at a rate of 1.5 percent

annually. In forecasting FCF for the valuation horizon and the

terminal value, Mr. Fuller took into account the earnings impact

of removing $12,919,000 from operating assets. Mr. Fuller

assumed that such a reduction would reduce Peoples' net interest

income, rather than loan income, because Peoples could readily

dispose of marketable securities, while its loans had proven to

be unmarketable. Accordingly, Mr. Fuller forecasted Peoples' net

interest income at approximately $1.1 million less than Peoples'

reported net interest income for the calendar year 1992--a

sufficient amount to reflect the loss of an approximately 8-

percent return on the nonoperating assets.

Mr. Fuller estimated Peoples' cost of capital using a

weighted average cost of capital (WACC) formula and calculated

Peoples' cost of equity using the standard capital asset pricing

model (CAPM) formula. The cost of equity was calculated using a
 - 42 -

7-percent risk-free rate, a risk premium of 7.3 percent, and a

beta of "about 1.0, approximately equal to the overall market

average of 1.0".

Mr. Fuller calculated beta using the average unlevered

beta18 for the 23 publicly traded Midwestern banks tracked by the

Value Line Investment Survey (4th ed. Apr. 9, 1993) (the VL

list). The VL list included the leading full-service commercial

banks in the Midwest, such as Banc One Corp., First Chicago

Corp., National City Corp., and Norwest Corp. The VL list did

not contain any small, single-location banks such as Peoples; all

the banks on the VL list were substantially larger. The market

capitalizations of the banks on the VL list ranged from

approximately $700 million to $15 billion, with mean and median

capitalizations of approximately $3.76 billion and $2.88 billion,

respectively.

After calculating unlevered betas for each of the companies,

Mr. Fuller calculated an average unlevered beta of 0.9 and an

average relevered beta of 1. Despite the fact that Peoples was

unleveraged, Mr. Fuller chose a beta of 1, the same as the

average relevered beta.

18
An unlevered beta measures the business risk of a company
by removing the effect of financial leverage. This permits
the betas of comparison companies to be considered so that
business risk can be isolated and evaluated apart from the
risks associated with financial leverage. Copeland et al.,
Valuation: Measuring and Managing the Values of Companies
331 (2d ed. 1994).
 - 43 -

Using the 14.3-percent discount rate he calculated under

CAPM, Mr. Fuller calculated discounted present values of

$3,480,000 and $4,660,000 for Peoples' FCF for the valuation

horizon, and its terminal value, respectively, for a value from

operations of $8,140,000. He then added back the book value of

the nonoperating assets, net of a 10-percent minority discount,

to arrive at a total equity value of $19,770,000 ($6,590 per

share).

There are significant shortcomings in Mr. Fuller’s

application of CAPM in this case that highlight our doubts over

the appropriateness of its application to the valuation of small,

closely held companies. As we said recently, "CAPM is a

financial model intended to explain the behavior of publicly

traded securities", and we "do not believe that CAPM and WACC are

the proper analytical tools to value a small, closely held

corporation with little possibility of going public." Furman v.

Commissioner, T.C. Memo. 1998-157, 75 T.C.M. (CCH) 2206, 2214,

1998 T.C.M. (RIA) par. 98,157, at 868-98. Unlike the market

contemplated by CAPM, the market for Peoples stock, to the extent

one even exists, is not efficient, liquid, or free of significant

transaction costs. Moreover, in relation to other closely held

corporations, the liquidity of a financial institution is even

further reduced by the fact that acquisitions and dispositions of
 - 44 -

its stock are subject to regulatory approval.19 Finally, CAPM

assumes that investors hold, or have the ability to hold,

diversified portfolios that eliminate, on a portfolio basis, the

effects of unsystematic risk--the elements of risk that are

specific to the asset held. Consequently, because CAPM assumes

that an investor holding a diversified portfolio will encounter

only systematic risk, the only type of risk for which an investor

can be compensated is systematic or market risk, which represents

the sensitivity of the future returns from a given asset to the

movements of the market as a whole. See id. (citing Brealey &

Myers, Principles of Corporate Finance 137-138, 143-144 (4th ed.

1991); Pratt et al., Valuing a Business 166 (3d ed. 1996)).

In calculating Peoples' discount rate, Mr. Fuller followed

the principles of CAPM and did not make any provision for

Peoples' unsystematic risk, based on the assumption that such

risk was diversifiable. Yet respondent and Mr. Fuller have

overlooked the difficulties in diversifying an investment in a

block of stock they argued is worth approximately $8.94 million.

Construction of a diversified portfolio that will eliminate most

unsystematic risk requires from 10 to 20 securities of similar

value. See Brealey & Myers, supra at 137-139. Thus, proper

diversification of an investment in the Peoples shares owned by

19
An acquisition of greater than a 24.9-percent interest
requires Federal regulatory approval.
 - 45 -

petitioner, as valued by respondent, would require a total

capital investment of at least $89 million. We do not think the

hypothetical buyer should be limited only to a person or entity

that has the means to invest $89 million in Peoples and a

portfolio of nine other securities.

As illustrated by Mr. Fuller's valuation, the selection of

beta is another problem inherent in the application of CAPM to

the valuation of closely held companies. See Furman v.

Commissioner, supra. Beta, a measure of systematic risk, is a

function of the relationship between the return on an individual

security and the return on the market as a whole. See Pratt et

al., supra at 166. The betas of public companies are frequently

published or can be calculated using historical pricing data on

the company's stock. Thus, a beta cannot be calculated for the

stock in a closely held corporation--it can only be estimated

based on the betas of comparable publicly traded companies.

However, because the betas for small corporations tend to be

larger than the betas for larger corporations, it may be

difficult to find suitable comparables when valuing a small,

closely held corporation. See Ibbotson Associates, Stocks,

Bonds, Bills & Inflation, 1993 Yearbook (Ibbotson) at 159;

Copeland et al., Valuation: Measuring & Managing the Value of

Companies 265-266 (2d ed. 1994). In this case, Mr. Fuller used 1
 - 46 -

as the beta, which equaled the relevered20 average beta of the

banks on the VL bank list. As discussed, supra, there are

substantial differences in size and operations between Peoples

and the banks on the VL bank list; we do not believe that their

betas are representative of the greater business risks faced by

Peoples. For example, in comparison to the large banks on the VL

list, Peoples had limited opportunities for growth; less ability

to diversify risk, because of limited product offerings and

dependence on the economic conditions of a few counties; lacked

the ability to create the economies of scale available to large

banks; had greater interest rate risk because it could not sell

mortgages on the secondary market; had less control over credit

risk due to inadequate underwriting standards and a lack of

information technology support; and could not afford to employ

the personnel and technology used by large banks to protect and

pursue earnings through the management of interest rate risk.

Mr. Fuller did not otherwise adequately support his

selection of a beta of 1, a figure he admits is "approximately

equal to the overall market average of 1 based on the S&P 500."21

That statement, if anything, suggests that Mr. Fuller's beta is

20
Mr. Fuller did not explain why he used the relevered
beta, rather than the unlevered beta, when Peoples was not
leveraged.
21
The S&P 500 stock index includes 500 of the largest
stocks (by market value) in the United States.
 - 47 -

unreasonably low; using a beta greater than 1 would increase the

discount rate used in the Fuller analysis, thereby decreasing the

value otherwise computed. We do not believe that an investment

in Peoples, a small, single-location bank, whose earnings were

susceptible to impending interest rate mismatches and sluggish

local economic conditions, presents the same systematic risk as

an investment in an index fund holding shares in 500 of the

largest corporations in the United States.

In calculating the discount rate, Mr. Fuller used an equity

risk premium of 7.3 percent, "based on the average share of

common stock of publicly traded companies", and cited Ibbotson.

We think that Mr. Fuller meant Ibbotson's long-horizon equity

risk premium, which represents the total returns of large company

stocks, less the long-term risk-free rate, which is widely used

in calculating a cost of capital under CAPM.

Although Mr. Fuller cited Ibbotson as his source for equity

risk premium, in his initial report he ignored a crucial aspect

of the Ibbotson approach to constructing a cost of capital--the

small stock premium. In his rebuttal report, Mr. Fuller

unsuccessfully tried to persuade us that the small stock premium

is not supported by financial theory, characterizing the risk

associated with a firm's size as unsystematic risk, for which the

market does not compensate. The relationship between firm size

and return is well known. Size is not an unsystematic risk

factor and cannot be eliminated through diversification. "On
 - 48 -

average, small companies have higher returns than large ones."

Ibbotson at 125 (citing Banz, The Relationship Between Returns

and Market Value of Common Stock, 9 J. Fin. Econ., 3-18 (1981)).

We have already alluded to the likelihood that small stocks will

have higher betas than larger stocks, because of greater risk.

See Ibbotson at 126. However, it has been found that the greater

risk of small stocks is not fully reflected by CAPM, in that

actual returns may exceed those expected based on beta. See id.

Consequently, when calculating a cost of capital under CAPM on a

small stock22, it is appropriate to add a small stock premium to

the equity risk premium, to reflect the greater risk associated

with an investment in a small stock in comparison to the large

stocks from which the equity-risk premium is calculated. Based

on Peoples' size, a microcapitalization equity size premium of

3.6 percent should have been added. See Ibbotson at 161.

Consequently, even if we accepted Mr. Fuller's beta of 1, which

we do not, Peoples' cost of capital should have been at least 18

percent.

b. Guideline Company Method

The market approach used by Mr. Fuller was the guideline

company method (guideline method). Under the guideline method,

22
There are actually three different premiums: (1) The mid-
capitalization equity size premium (capitalization between
$696 and $3,015 million); the low-capitalization equity size
premium ($171 million to $696 million); and (3) the
microcapitalization equity size premium (capitalization
below $171 million).
 - 49 -

value measures are developed using the stock prices of similar

companies (guideline companies) that are publicly traded. The

value measures are then compared to the subject company’s

fundamental data to reach an estimate of value for the subject

company or its shares. Because value under the guideline method

is developed from the market data of similar companies, the

selection of appropriate comparable companies is of paramount

importance.

Mr. Fuller’s principal criterion for selecting guideline

companies was geography, rather than size, financial, or

operating characteristics. All seven of the guideline companies

selected operated primarily in Indiana, Illinois, and Ohio.

As in the DCF analysis, Mr. Fuller adjusted the values of

Peoples' equity and assets to adjust the book equity-to-assets

ratio to 9 percent and made an adjustment to earnings. Mr.

Fuller then calculated the median price-to-earnings23 multiple

(10.4), price-to-assets ratio (12.1 percent), and price-to-book

equity ratios of the guideline companies (110.3 percent). After

calculating a value from operations using the ratios, Mr. Fuller

added back the excess equity value, reduced by a 10-percent

minority discount, to find the total value of Peoples' equity.

Applying the ratios to the adjusted equity, assets, and earnings

23
The price-to-earnings multiple used by Mr. Fuller was
based on the most recent four quarters' earnings for each
corporation.
 - 50 -

figures of Peoples, Mr. Fuller determined the following total

equity values:

Ratio Total Value

Price-to-earnings $24,751,000

Price-to-book equity 19,344,000

Price-to-assets 21,021,000

Mr. Fuller then used the mean of the values determined using the

price-to-book equity and price-to-assets ratios to determine a

total equity value of $20,200,000. He did not include the value

determined using the price-to-earnings ratio, as he thought the

“unusually high earnings reported for the period may result in

the value of Peoples being overstated.” Finally, Mr. Fuller

applied a 10-percent marketability discount.

Mr. Fuller supported his finding of a 10-percent

marketability discount in his discussion of both marketability

and control premium factors. He concluded that little or no

marketability discount was appropriate, because the estate shares

carried significant elements of control and might command a

control premium. Mr. Fuller failed to focus on the fact that two

conceptually distinct adjustments were involved, one a discount

for lack of marketability and the other a premium for the

benefits of control. See Estate of Andrews v. Commissioner, 79

T.C. 938, 952-953 (1982). Although there may be some overlap,

because control, or lack of it, is a factor that may affect

marketability, even controlling shares in a nonpublic corporation
 - 51 -

can suffer from lack of marketability, because of the absence of

a ready private placement market and the costs of floating a

public offering. See id. at 953.

We agree with Mr. Fuller's use of the guideline method and

his adjustments to reflect excess capital; however, we do not

think that his selection of guideline companies was appropriate,

in light of Peoples' thriftlike operations and earnings base.

Five of the seven guideline companies selected by Mr. Fuller were

bank holding companies engaged in a broad range of personal and

commercial banking services. Only two of the guideline companies

chosen were thrifts, and like most of the other guideline

companies, they were multibranch institutions that had

significantly greater assets than Peoples, though not by the same

order of magnitude as the banks on the VL bank list. On its

December 31, 1992, balance sheet, Peoples reported total assets

of $90.6 million; in comparison, the mean and median total asset

values of the guideline companies were $303.1 million and $323.3

million, respectively, for the comparable period.24

2. Petitioner's Expert
Petitioner relies on the expert report of James E. Magee, a
director and senior associate of Alex Sheshunoff & Co. Investment
Banking (ASC). Headquartered in Austin, Texas, ASC is nationally

24
Mr. Fuller provided asset values for six of the guideline
companies as of Dec. 31, 1992, and for the seventh as of
Dec. 31, 1993.
 - 52 -

known for its valuation and mergers and acquisitions expertise in
the financial services industry and has been recognized as an
expert by Federal banking regulators, including FDIC, Federal
Reserve Bank (FRB), Office of the Comptroller of the Currency
(O.C.), and OTS. ASC has completed over 300 merger and
acquisition transactions and over 3,500 stock valuations
involving regional and community banks and thrifts.
Mr. Magee has over 30 years of experience in the banking
industry. In the first half of his career, Mr. Magee worked in
management positions at two New York banks, including one money
center bank where he was a vice president, and as a regulator
employed by the board of Governors of the Federal Reserve System,
Division of Supervision and Regulation. The latter half of Mr.
Magee's career has been spent as an appraiser and consultant
serving the banking industry exclusively. Mr. Magee holds an
M.B.A. in finance from Adelphi University in New York.
Using the guideline method, Mr. Magee valued the estate
shares at $4,497,000 ($3,000 per share). While employing the
same general approach as Mr. Fuller, there are a number of
differences in Mr. Magee's report that account for their
substantial differences of opinion regarding the fair market
value of the estate shares.
Mr. Magee's methodology for selecting guideline companies
was significantly more exacting than Mr. Fuller's. As discussed
supra, Mr. Fuller's guideline companies included five banks and
two thrifts, most of which were significantly larger than
Peoples. In contrast, Mr. Magee's selection criteria were
 - 53 -

limited to thrifts comparable in size to Peoples. The emphasis
on thrifts, rather than banks, is in accordance with our finding
that Peoples, while legally chartered as a bank, more closely
resembled a thrift in its operations.
As discussed supra, the guideline company data used by Mr.
Fuller was based on publicly traded minority interests; Mr.
Magee, in contrast, used two groups of guideline companies, one
based on mergers and acquisitions of private companies, the other
based on publicly traded minority interests like that used by Mr.
Fuller. Mr. Magee looked at both minority and control
transactions because he conceded that the estate shares had
effective control.
To examine thrift pricing on a control basis, Mr. Magee
selected six thrifts (the control group) meeting the following
criteria: (1) Thrifts that sold in the Midwest, (2) return on
average assets greater than 1 percent, (3) total assets less than
$100 million, and (4) transactions that were pending or completed
between January 1 and December 31, 1992. In order to examine
thrift pricing on a minority basis, Mr. Magee selected 10 thrifts
(the minority group) meeting the following criteria: (1) Thrift
organizations in the United States, (2) total assets less than
$150 million, (3) not subject to announced or rumored
acquisition, and (4) publicly traded securities as evidenced by
listing on a major exchange [or trading market].
3. Comparison of the Experts' Reports
In performing their analyses under the guideline method,
Messrs. Fuller and Magee both focused on the same three ratios:
 - 54 -

(1) Price-to-earnings, (2) price-to-book equity, and (3) price-
to-assets. However, they disagreed to some extent on the weight
to be accorded each of the three ratios. As discussed supra, Mr.
Fuller used an equal weighting of the values derived using the
price-to-book and price-to-assets ratios, while rejecting the use
of the price-to-earnings ratio over concerns that it would
overstate value. Mr. Magee used an equal weighting of the values
found using the price-to-earnings multiple and the price-to-book
ratio. Mr. Magee did not use the price-to-assets ratio in
reaching his valuation conclusion and described it as a “check
point” for the other two ratios, rather than as the “principal
determinant of the value of a controlling interest.” However,
Mr. Magee noted, the price-to-assets ratio does provide
“additional stability” to the analysis by removing the effects of
variability in earnings and book equity.
We agree with Mr. Fuller that the use of the price-to-
earnings ratio may overstate the value of the estate shares, due
to the fact that a large portion of Peoples earnings was
attributable to investments in high yielding Treasury securities.
We also think the weighted average of the price-to-book and
price-to-asset ratios will be more likely to cancel out any
anomaly in the data for either ratio. Accordingly, in valuing
the estate shares under the guideline method, we look to the
price-to-book and price-to-asset ratios.
The mean, median, high, and low values for the guideline
companies examined by Messrs. Fuller and Magee are as follows:
 - 55 -

Guideline method ratios

Price/earnings Price/book Price/assets
multiple ratio ratio
Fuller
Mean 11.20 120.50% 11.90%
Median 10.40 110.30 12.10
High 15.40 171.50 16.30
Low 6.70 72.10 6.40

Magee

Control
Mean 9.16 124.14 10.38
Median 8.64 124.87 9.95
High 16.40 167.76 15.00
Low 2.74 76.19 5.41

Minority
Mean 6.42 54.73 n/a
Median 6.45 52.74 18.52
High 6.47 82.05 37.60
Low 6.35 32.38 9.50

Based on the data of their respective guideline companies,

Messrs. Fuller and Magee chose the following ratios to value the

estate shares:

Guideline method ratios used
in valuing the estate shares

Price/earnings Price/book Price/assets

Fuller 14.13* 153.7% 12.3%

Magee 5.5 65.0% not used

* Calculated, but not used in actual valuation

Differences in perception are common in questions of

valuation. The differing ratios chosen by Messrs. Fuller and

Magee to value the estate shares reveal an extreme divergence of

views. All three ratios used by Mr. Fuller exceed the respective
 - 56 -

mean and median of the guideline companies; both the price-to-

earnings multiple (used but ultimately ruled out) and the price-

to-book equity ratio are near the highest values in the guideline

company data. In contrast, the price-to-earnings multiple and

price-to-book ratios selected by Mr. Magee are comparable to the

mean values from the minority group data.

While there is little difference in Mr. Fuller's guideline

company data, and Mr. Magee's control group data, we think that

Mr. Magee's criteria for the selection of comparable companies

produced a group of companies that more closely resembled the

size and operating characteristics of Peoples than Mr. Fuller's

guideline companies. Accordingly, in determining the value of

the estate shares under the guideline method, we rely on the data

supplied by Mr. Magee.

Mr. Magee did not, however, address Peoples'

overcapitalization and, unlike Mr. Fuller, did not make any

normalizing adjustments. In contrast, as discussed supra, Mr.

Fuller removed excess equity, valued equity from operations, and

then added back the excess equity. Adjustments to equity were

necessary to value Peoples properly, and we think Mr. Fuller used

a sensible approach in so doing. Accordingly, in valuing the

estate shares, we use adjusted equity and assets of $6,999,000

and $77,770,000, respectively, for purposes of the guideline

method, and $12,919,000 in excess equity before discounts.
 - 57 -

We also differ with Mr. Magee on his use of ratios that more

closely resemble the minority group data than the control group

data. Inasmuch as the estate shares had effective control, we

think that they should be valued as a controlling, rather than

minority, interest. Accordingly, we value the estate shares

using the control group data.

We disagree with Mr. Fuller's optimistic assessment of

Peoples' standing among comparable institutions (or the less-

than-comparable institutions he used). Had we not removed the

excess equity in performing the guideline method, then perhaps

Peoples would be more attractive than its financials would

otherwise suggest, due solely to the value of excess equity,

which could be paid out as an extraordinary dividend. However,

when using the guideline method to value Peoples' equity from

operations (and adding back the excess capital), we think that

the attractiveness of Peoples, and of the estate shares, takes a

dramatic nosedive.

As an institution, Peoples was financially sound, but

offered an investor little hope of meaningful growth in revenues

or earnings. A number of negative factors have been discussed

supra, such as a limited market, limited product offerings,

aggressive competitors, and outdated technology. Peoples was

also hindered by its employees, who on average were at least in

their midfifties, and tended to resist change. In sum, we think

that Peoples showed little potential to be much more than what it
 - 58 -

was on the valuation date; we therefore think it appropriate to

value Peoples as an enterprise at the low end of the control

group. Accordingly, for purposes of valuing the estate shares,

we use a price-to-book ratio of 77 percent and price-to-assets

ratio of 5.5 percent. Based on adjusted shareholder equity of

$6,999,000 ($2,333 per share) and adjusted assets of $77,770,000

($25,923 per share), we find a value of equity used in operations

of $5,389,230 ($1,796.41 per share) using the price-to-book

ratio, and $4,277,350 ($1,425.78 per share) using the price-to-

assets ratio. Averaging the two ratios, we find a value of

equity from operations of $4,833,290 ($1,611.10 per share). We

agree with Mr. Fuller's application of a 10-percent minority

discount to the excess equity, and, accordingly, add $11,627,100

($3,875.70 per share) of excess equity from nonoperating assets

to the value of equity from operations, producing a fair market

value of total equity of $16,460,390 ($5,486.80 per share).

Accordingly, we hold that the fair market value of 1,499 shares

of Peoples stock, before consideration of marketability concerns,

is $8,224,713.25

Mr. Magee distinguished the estate shares from the shares of

publicly traded companies due to their lack of marketability in

support of his application of a 30-percent marketability

discount. Mr. Magee cited several empirical studies that, on

25
$5,486.80 x 1,499 = $8,224,713
 - 59 -

average, support the application of marketability discounts in

the range of 30 to 45 percent. Mr. Magee analyzed several

factors with respect to the marketability of the estate shares,

such as earnings quality, dividend payment history, size of the

block of stock, prices of comparable investment substitutes,

management's stock redemption policies, capitalization of the

firm, and economic factors. In his analysis he noted certain

factors in support of a discount, including Peoples' interest

rate risk, the sewer tap ban, and the modest economic growth in

Peoples' market area. Mr. Fuller also emphasized the lack of

liquidity in Peoples stock, which was not subject to any

repurchase or employee stock option plan and was not easily sold,

as evidenced by the fact that a block of 100 shares had been

offered and available for sale for more than 5 years without

eliciting any expressions of interest.

While we recognize that elements of control may enhance

marketability, we do not think that the estate shares were

rendered marketable by virtue of their effective control.

No matter who was in control, Peoples was still a small,

community bank with limited growth opportunities, capitalized

with common stock that was not publicly traded and not easily

sold privately. A buyer of the estate shares would either have

to sell the block privately, cause Peoples to make a public

offering, or seek an acquiror. Any of those three options could

take a number of months, and require significant transaction
 - 60 -

costs for the services of accountants, lawyers, and investment

bankers.

Accordingly, although we recognize the estate shares'

effective control in valuing Peoples equity from operations, we

do not think that a 49.97-percent interest in a small, closely

held bank, is a readily marketable interest. Accordingly, we

apply a 30-percent marketability discount to the fair market

value of the estate shares, and hold that on the valuation date,

the estate shares had a fair market value of $5,757,296

($3,840.76 per share).

To give effect to the concessions of the parties and our

determination of the fair market value of the estate shares,

Decision will be entered

under Rule 155.