Correspondence from Winner to Morrill; Annotated Memorandum from Winner to Williams

    CHAMBERS. FERGUSON. WAT—I". WALLAS. ADKINS 8: FULLER. F’.A.

ATTORNEYS AT LAW
SUITE 730 EAST INDEPENDENCE PLAZA

951 SOUTH INDEPENDENCE BOULEVARD

JULIUS LEVONNE CHAMBERS CHARLOTTE, NORTH CAROLINA 28202
JAMES E FERGUSON “ TELEPHONE (704) 375-846!

MELVIN L. WATT

JONATHAN WALLAS

KARL ADKINS

JAMES c. FULLER. JR. September 25 , 1981

C. YVONNE MIMS
JOHN W. GRESHAM
RONALD L. GIBSON
GILDA F. GLAZER
LESLIE J. WINNER

Mr. Richard MOrrill
Department of Geography
University of Washington
Seattle, Washington 98195

Re: Reapportionment of North
Carolina General Assembly

Dear Mr. Morrill:
This letter is written pursuant to our conversation of September

25, 1981 concerning apportionment of the North Carolina General
Assembly.

As we discussed, I anticipate needing the answers to the following

questions in order to be able to prevail in the litigation which

seeks to require reapportionment of the General Assembly:

1. Using the current rules which prohibit dividing
counties, how can the State be apportioned to divide the
population of the State more perfectly so that there is
lower deviation from the ideal population per representa-
tive? In answering this question, we probably need to limit
the number of representatives in a multi—representative, multi-
county district to the largest number in the current plan.

That number for both the House and the Senate is four.

2. If the rules are changed to allow division of
multi-member single county districts, what effect does
that have on (a) the number of districts in which the major-
ity of the population is black; and (b) the degree of racial
concentration in the plan?

3. ' If county lines are disregarded and the only re-
quirement is contiguity of census tracts or enumeration
districts, what effect does that have on (a) the number of
districts in which the majority of the population is black;
and (b) the degree of racial concentration in the plan?

Mr. Richard Morrill
September 25, 1981

Page 2

With regard to this question, Jerry Ingalls said that since
there are about 1000 census tracts and enumeration districts,
it would not be feasible to answer this without using smaller
subdivisions than the whole State to disaggregate. He had
suggested making a plan that maximizes racial concentration
keeping counties intact, and then subdividing those represen—
tative districts. He also suggested that you may know a better
solution to this technical problem.

4. DOes the adopted 1981 plan unnecessarily dilute black
vote? Bipartflnflar GD how does racial concentration in this plan
compareto racial concentration in the 1970 plan; and (b) is there
a plan that concentrates black population more without dividing
counties?

There are a couple of other considerations. First Jerry Ingalls
proposed in addition to analyze the compactness of the districts.
While compactness is not our primary goal, I think that whatever plans
we suggest must be able to withstand the criticism that they are too
spread out. I think that to do this we only need to be able to show
that our plans are not any less compact that the adopted plan is.

Second, North Carolina has previously considered point contiguity
as contiguous. If it is helpful to us, we can do that too.

Third, down the road we will need to be able to analyze whatever new
plans they propose along the same criteria as we have analyzed our
own, that is, racial concentration and population deviation.

We would like to be able to go to Court in mid—December or early
January to prevent an election from being held pursuant to the current
plan. We, therefore, will need the answers to these questions before
that. In addition, and since I spoke with you, I have learned that
the State has just submitted the North Carolina Constitutional pro—
visions to the Department of Justice for pre—clearance. Thus, the
sooner we can get evidence of the racially dilution effect of prohi-
biting division of counties, the better.

We are looking for someone with the technical skills to give an
objective answer to these questions and with the credentials to
qualify as an expert who can testify about the answers in Court.

Mr. Richard Morrill
September 25, 1981

Page 3

My conversation with you lead me to believe that you are interested
in‘the project and are capable of doing it. I do need the answers
to these questions:

1. What is the fastest that the analysis can be completed?

2. How much will it cost? You suggested over the telephone
that you thought you could do it for $10,000. I assume this does
not include travel costs. Are there other costs that it does not

include, for example, witness fees?

3. You mentioned on the telephone that you would ask
graduate students to do part of the analysis. We need to be sure
that you will be able to testify about the entire result and that
the graduate students will not need to testify separately. How do
you anticipate dividing the responsibilities?

4. For my own curiousity, could you send me a copy of any
relevant articles you have written?

For your information I am enclosing a copy of the Complaint in the
law suit, a copy of the maps of the current plan of apportionment,
and a copy of the relevant constitutional provisions.
I enjoyed talking with you and I look forward to hearing from you.
Sincerely,
@4241. Cam
Leslie J. Winner

LJW:ddb
cc: Mr. Napoleon Williams
Mr. Steve Suitts

{EA 9“ /

Mr. Napoleon Williams
NAACP Legal Defense Fund71mu-
Suite 2030
gflflrlo Columbus Circle
(fi' / New York, New York 10019
t

Re: Gingles, et al. v. Edmisten,
et al.; Computer Analysis

 

Dear Napoleon:

I have gotten a final estimate for the folks at the University
of North Carolina at Charlotte to do the necessary analysis
for the reapportionment case. Before I describe it and its
costs, I would like to say that it has been a pleasure to
work with them. Not only did they seem to know what they
are talking about, but also they seem to have a genuine
interest in the result of the study and the result of the
lawsuit.

The following is my layperson's understanding of what Jerry
and Paul propose to do for us. I am going to run this
letter by Jerry before I mail it to you so that I can be
sure that he and I-are on the same wave length. At that
point, I will also alert him to expect a telephone call from
you and your computer person to answer any questions which

you may have.

First of all Jerry and Paul will obtain the 1980 Census data
in a form that is capable of being used on their computer
and will also obtain whatever data is necessary to put the
map of census tracts and enumeration districts onto their
computer. They will use this data-fa a program which is
called Location Allocation Problem as the basic tools of
their analysis. It is my understanding that Loca tion
Allocation Problem is a well-developed program for doing

this sort of problem and is the most reliable one in the

country. I} I4» wom_ueed x,7lt Sfi41«7 “Mohv7buté a.
SWW On IWA ‘

 

does not do us any good to come up with a plan with 80

V single member districts if the populatxg deviation is 60 or
70 percent. We need to limit the population deviation to

“934* ‘ .

the latest amount which the Supreme Court has approved for a
State legislature in order for the results of this part of
the analysis to be :Erxnuch use to us.q?The next part of the
analysis will be the part that changes the rules to allow
the subdivision of counties. The analysis would be essentially
the same as the first part analysis except that the rule of
keeping counties inyfiact would be disregarded. This would
be done in two ways. The first would be to make a rule that
a multi-member single county districtscan be be split into
single member districts but there will be no district that
consists of pieces of more than one county. The second way

to do it is to disregard county lines and put together

districts consisting of any census gract or enumeration
v/district as long as they are contigkpus. These would be
4

/selected to optighize low variance, high compactness, and

high racial concentration and would be analyzed in the same

- A
magtér as the first partIand will be compared to each other

and to the results of the first part.

Because of the exceeding%harge number of possibilities of
districts, if you totally disregard county lines and have no
other rules, some artificial limits will have to be created
in order to make it practical to run the program. The
problem is picking which sub ets of the state to use as
(imafingpto Qv" " ;
units to desegrate £55? single,§ember districts. The most
obvious choices are to take the best solutions from first
part of the analysis which maximized racial contration and
devide those districts even further to further maximize
minority concentration. Another idea is to disaggregate
congressional districts. In any event, this part of the

program can be run several different ways and we will have a

choice of final solutions to choose from. We will probably

v

The first part of their analysis will be to analyze the
current plans; They will analyze it with respect to the
respgee—ef variance from the ideal district Size, the compactness
of each district,and the between versus within district
variance of minority population. The percent variance from
the ideal population, of course, relates to the one person——
one vote. The compactness anaylsis is necessary in order to
show that any plan which we propose does not have districts
that areAless compact than the plan which they adopted. Of
course, the between versus within district variancéZminority
population will.tend to show whether black people are concentrated
within districts allowing them to prevail in elections or if
they are evenly disbursed among districts which would tend
to dilute the vote of black citizens. They will then analyze
the implications of these resutls with respect to the constitutional
‘ mmdulfi M
one personrone vote requirements,.germaiadifigva campaign
a~4

efficiency, in the likely miniority composition of the

legislature.

The next step will be to optimize the current plan using
approximately the same total number of districts and number
of representatives per district. That is, they will produce

. WWW. .rsam . WW“ 0) .
plans which haveaa minimize of population de+vaaee, a maXimum
of compactness, andfg maximum concentration of minority
population. The next step will be to use the same ground
rules, that is,do not divide counties, but will vary the
number of districts and the npmber of representatives per
district in order to try to produce equal population districts
and in order to try to minimize the number of multi—representative
districts. Each of these results will be analyzed for the
same three factors: variance from target, compactness, and

racial concentration. It is my suggestion in this phase

that in doing the portion that minimizes the number of

r
multi-representative districts, they simultaneou§:%ht a cap

on the variation from the target population. That is, it

 

need to discuss more how to divide up the state as the
starting point for this piece of the analysisfq¥The final
part of the product will be a summary comparison of the
analysis and identification of the best solution on each
criteria and the implications and trade offs in selecting
any "best" solution. The rest of their wrap up really will
not have too much to do with us. There-a;e considerabie
alternatives to the lgcation Bllocation problem/and discuss
implications o€?%%€iJ~mizing criteria/and:fh:%§3licy implications
of the North Carolina districting procedure.fi?Thus, what we
get, in my opinion, is the answers to the following questions:
1. Could they have drawn the lines for apportioning

the State legislature using the current rules in a

way that more successfully complied with one

person 435 one vote criteria and/or in a way that

was more successful in concentrating minority

votes?

2. What are the implications for minority concentration;
of dividing within counties but not dividing and

crossing county lines at the same time?

\

but

3. How can youAmaximize black concentration disregarding

county lines?

4. How do these solutions compare with the current
LUI
method which regard to other legitimate government
concerns, that is, compactness of districts and

variation in the population of districts?

What this will costs us is $20,000.00. That is computed as
0 WM “3 .’ ,
follws: $3,000.00 for typing, supplies, equipmenfi, travel, kflqflfi

d

11 Cash)

 

 

f“! a

 

document preparation, etc.}$5,000.00 for computer time; Z;;:~
. F s

$10,000.00 for salaries} ' {he salaryggf-Paul L
. «“5

//€;;ith‘the computervp czuoivcyquzflu/C , Jerry Ingles, who
/ U “

b is a political geographer, some consultation with other

I people in their field, and some student labor to do some of

I .the routine tasks. Jerry and Paul are anticipating

  
  

working for 20 hours a week for 8 to 10 weeks on the PrOject
Ixr>(L‘\,/'I-.;I_.‘..) ,

CL
:\,'Q—\ ‘b . ~
.1 and hag*built inA$%,000.00 a piece for their salariiésand

I

 

\§z7666166“f6£ university overhead totalling $20,000.00.
_7 1;, I {Wu—VI"

V’analyEflng this amount it is important to note that they
estimate that they will actually use approximately $16,000.00
of computer time,but because they will be using the computer
only in the middle of the night at its off peak time,they

I believe they can get the computer center to agree to sell
them the time for a flat rate of $5,000.00. It is also
important to note that the_university's usual rule for
overhead is 50 percent of the project amount. They believe
they can negotiate for a 7 to 10 percent overhead instead.

fIThis amount assumes that they will answer all questions in
the outline which they gave me on September 13, 1981 as we
modified it in our discussions and also includes time in
court if the testimony is before December 15. They will
need to have their travel expenses covered if they need to
attend court out of Charlotte. After December 15 they would

charge us $30.00 per hour for actual in court time and will

need to have their travel expenses covered.

 

 

 

My feeling is they will do a very thorough job and will be

easy to work with and that their price is not out of line.

Of course, I am not the person holding the purse. Why don‘t

you give me a call and we can discuss this. If you would

like to talk to Jerry his telephone number is 704/5&2=%9%.
577-17. ‘73.

I am not sure that I did a very good job of explaining all

of this. I hope you can make heads or tails of it.
Sincerely yours,

Leslie J. Winner

LJW:ofh