RECEIVE11
inr22 3 46Pi!‘00
paLgAL fl.iTE ~~Cl4’ilLi;iiiH
,,,cr,~E r,F T,iZ SECiiETAHI
BEFORE THE
POSTAL RATE COMMISSION
WASHINGTON, D.C. 20268-0001
POSTAL RATE AND FEE CHANGES, 2000
Docket
No. R2000-1
DIRECT TESTIMONY
OF
JAMES T. HEISLER, Ph.D.
ON BEHALF OF
PITNEY BOWES
Communications
regarding this document
should be served on
Ian D. Volner
N. Frank Wiggins
Venable, Baetjer, Howard & Civiletti, LLP
1201 New York Avenue, N.W.
Suite 1000
Washington, DC 200053917
Counsel for Pitney Bowes Inc.
Dated:
May 22, 2000
PB-T-3
1
2
3
4
AUTOBIOGRAPHICAL
5
Research
6
Opinion Research
7
President and Manager of its Washington
8
Vice President in 1988 and Executive Vice President in 1993. I relocated to the
9
Princeton
10
company’s
11
directors.
12
practices serving the IT/Telecommunications
13
issues area.
14
SKETCH
My name is James T. Heisler.
Industry for 32 years.
Corporation
headquarters
in the Marketing
I currently am Executive Vice President of
International.
I joined ORC International
as Vice
DC office in 1982, was named Senior
in 1993. My current duties include Director of the
Interactive Services.
I also am a member of the corporate
At various times, I have also been responsible
Opinion Research Corporation,
15
larger survey research organizations.
16
corporation
17
I have been employed
board of
for professional
industries and Market Assessment
founded in 1938, is one of the world’s
It has been has been an independent
since 1991 and publicly traded since 1993.
I hold a Ph.D. in Social Psychology
1
from Illinois Institute of Technology.
1
Purpose
and Scope of Testimony
The purpose of my testimony
2
is to sponsor and explain market research
3
conducted
on behalf of Pitney Bowes Inc. that measures
4
household
customer
5
First-Class
Mail for which postage
6
personal
computer
7
research
results
8
information,
9
rate First-Class
reactions to possible discounts
been
11
execution
12
results.
by Pitney
to shape its request for discounted
Bowes,
postage
meter or by a
as PC postage.)
together
with
The
other
for such single piece
Mail that complies with metering requirements.
I present the conclusions
10
is metered, either by postage
used
and non-
for certain single piece rate
via the Internet (referred to hereafter
have
household
of the research
from this research then describe the design and
and then discuss
and characterize
the research
13
14
Conclusions
15
The results of this study indicate that there is a substantial
16
in PC postage
17
postage
18
evidenced
19
not using meters.
20
and postage
is associated
meters when a one cent discount
with the use of these services.
by mailers, in households
21
postage,
22
Class Mail currently
23
referencing
with a one cent discount
using stamps.
PC Postage
portion
on First-Class
This interest
and in small businesses,
The data from the household
market interest
of this study
has been
who are currently
indicate
that
PC
will attract up to 1.4 billion pieces of FirstThe non-household
portion
of the study
indicates that 2.6 billion pieces from small businesses
2
1
will divert from stamps.
The non-household
2
stand-alone
3
Class Mail prepared
4
pieces of mail from stamps.
portion of the study referencing
meters indicates that implementation
of a one cent discount for First-
with postage meters is calculated
to divert up to 3.5 billion
5
6
The Research
Design
7
A.
8
ORC International
9
Overview
representatives
conducted
of qualifying
10
was with representatives
11
Study).
two telephone
households
studies.
(the Household
of qualifying
One study was with
Study). The other study
small businesses
(the Non-Household
12
13
1.
The Household
14
To qualify for the Household
15
households
16
and inkjet or laser printers.
17
responsible
18
19
Study
Study, respondents
that use stamps and have personal computers
for preparing
Representatives
The respondent
and Internet access
had to be the individual most
First-Class Mail for the household.
of
qualifying
households
were
asked
about
their
reactions to discounts on PC postage for First-Class Mail.
20
2.
21
To qualify for the Non-Household
22
had to come from
The Non-Household
Study
Study, a business
had to use stamps
and spend at least ten dollars on postage per month. Respondents
3
to the Non-
1
Household
2
mail.
3
Study were those
Approximately
4
their reaction
5
discounts
responsible
half the non-household
to discounts
for PC postage
for decisions
respondents
were
and half about
regarding
asked
about
their reaction
to
for metered postage.
6
B.
The Samples
7
1.
The Household
8
The underlying
9
individuals
U.S.‘telephone
Study
sample for the Household
households.
Study was a random sample of
ORC International
is a licensee
10
Sampling
11
Washington
12
telephone
13
states was drawn using the Genesys Sampling System.
System created and maintained
PA. ‘This
households.
by Marketing
is one of a few standard
A sample of households
of the Genesys
Systems Group of Fort
sources
for samples
in the forty-eight
14
2. The Non-Household
15
Pitney Bowes, using their licensed copy of the Dun & Bradstreet
of business
17
step, Pitney-Bowes
18
users of Pitney Bowes postage
19
made to remove
20
universe
21
25 employees
22
interviews was set at 200 for each of these sub-populations,
customers
generated
the non-household
sample.
were removed from the database,
meters from the population.
non-household
or less and 246,604
As a first
thus removing
was
The resulting
locations of which 9,008,956
had 26-50 employees.
4
database
No attempt
users of other meters from the population.
count was 9,255,550
contiguous
Study
16
establishments,
of U.S.
had
The target number of
or 400 interviews
in
1
total.
The samples
2
database..
3
C.
4
The questionnaires
5
similar,
6
household
for the study were then systematically
with
minor
for the household
differences
settings.
included
current First-Class
9
was read a concept statement.
described
PC postage.
11
either
12
determination
13
the sample selection
14
each concept.
postage
household
questions
and
and a question
Once a respondent
In the Household
In the Non-Household
meters
the
Study were
are included in Library Reference
qualifying
Mail piece volume.
10
and Non-Household
to accommodate
The questionnaires
8
qualified,
process.
were
Approximately
then asked
Study the statement
half the respondents
how likely they would
described
17
one-cent
discount,
18
would be extremely
19
at a given price level, they were not asked about higher price levels.
D.
21
Interviewing
on First-Class
Once a respondent
likely to use the service described
during
be to use the
described
20
at three price levels: no discount
the
were to see
concept
discount.
about
Study,
16
and a two-cent
PB-2.
Study the concept statement
In the Non-Household
or PC postage.
non-
he or she
of which concept would be used was made systematically
Respondents
15
from the
The Questionnaires
Each questionnaire
7
selected
postage,
a
said she or he
in the concept statement
Interviewing
22
Telephone
Center
23
(CATI) was utilized.
by telephone
in Tucson
AZ.
was
conducted
Computer
The CATI system
5
from
Assisted
ORC
International’s
Telephone
can be programmed
Interviewing
to facilitate
the
1
handling
2
next question while the software follows simple or complex instructions.
interviews,
encourage
respondent
Interviewing
6
E.
7
Data were transferred
known population
population
Standard
to the table
processing
Results
17
1.
were used to bring results
into line with
sample
contacted
for the study was balanced
against
gender, age, income, and region proportions.
samples were weighted to population
Industrial Classification
Weighted
16
cross tabulations
The Household
current estimate
20
households.
counts by broad
(SIC) groupings.
were produced.
Study
Of the households
19
21
from the CATI system
procedures
The non-household
15
18
and Weighting
parameters.
The household
13
14
Data Processing
Simple weighting
11
12
took place between February 15 and March 6,200O.
software.
9
10
see only thee appropriate
cooperation.
5
8
letting the interviewer
The average interview length was under ten minutes, a target length set to
3
4
of complex
contacted,
41% qualified
of 99 million U.S. households,
for the interview.
this corresponds
The following concept was read to qualifying household
6
Using a
to 40.8 million
respondents:
As you may know, there are ways to affix postage to mail you send
through the US Postal Service besides using postage stamps.
One
way to do it is electronically via the Internet using standard wordprocessing programs and an inkjet or laser printer.
With Internet
postage..
. You can refill postage over the Internet or by phone 24 hours a
day, 7 days a week, so you no longer have to make trips to the
Post Office to purchase stamps
. Postage can be printed’ directly from your printer onto
envelopes or labels as they are being addressed using standard
word-processing
programs
. You don’t have to worry about keeping track of single stamps
. Your addresses are checked and corrected against the US
Postal Service database
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Then the respondents
were asked their likelihood of subscribing
to the PC
postage service, with no mention of a postage discount.
Ql
If you could have access to all the features I’ve just described for a
monthly access fee of just $5.00, plus the regular cost of first-class
postage you actually use (IF NECESSARY:
33 cents for a regular
first-class letter), how likely would you be to subscribe to this
service? Please use a scale where “5” means you would be
extremely likely to subscribe and “1” means you would be not at all
likely to do so. Of course, you may use any number between 1 and
5.
Unweighted
Weighted
5
4
3
Base
200
Base
205
Extremely likely
7%
7
12
10
65
0
L
1 Not at all likely
6 Don’t know/Refused
28
7
Those who did not say they would be “extremely
1
2
service were asked a similar question
3
piece discount of one cent on First-Class postage.
Unweighted
Weighted
which now included
reference
10%
9
11
1
55
0
intent data.
If a great deal of
5
historical data are available for the product or service in question,
6
tailor an adjustment
7
necessary
9
to a per
192
Base
There are a variety of ways to handle
8
to the
189
Base
5 Extremely likely
4
3
2
1 Not at all likely
6 Don’t know/Refused
4
likely” to subscribe
it is possible to
that is specific for the product or service in question.
The
historical data are not available in this case
I have chosen
estimate
to use 80% of the extremely
of the likely user population.
10
used
in consumer
11
adjustment
12
followed
13
are other options.
14
(5) responses,
15
approaches
research
also offers
when
some
in this questionnaire
likely respondents
This is an adjustment
no historical
computational
data
as an
that is regularly
This
are available.
advantages
for the approach
when multiple price levels are investigated.
There
For example, .we might use 60 percent of the extremely
40% of the (4) responses
and 20% of the (3) responses.
generally produce recognizably
8
similar results
likely
The two
The adjusted intent estimates can then be converted
1
2
by including
reported
mail volume for appropriate
3
The total estimated
4
actual calculation
5
1. The Non-Household
6
The computations
7
the Household
8
consists
9
establishments
respondents
pieces affected by a one cent discount
in the equation.
is 1.436 billion.
The
is appended.
Study
associated
with the Non-Household
Study, with two important additions.
of two populations;
establishments
with 26-50 employees.
different rates and, therefore, the calculations
11
population.
12
concerning
reactions
to possible discounts
13
concerning
possible
discounts
14
were systematically
15
drawn.
16
found in each of the two studies.
17
made separately.
Moreover,
Study are like those for
The Non-Household
with 25 employees
These two populations
10
18
to volume estimates
the Non-Household
19
by a respondent.
20
the concepts,
21
of the original
22
concept
Calculations
23
qualifying
population
were sampled at
for use of postage
respondents,
The calculations
Potential
respondents
when the sample was
approximately
half would be
of potential effect need to be
is considered
at the point at which they begin to consider one of
the subset of the respondents
sample,
one
meters and one
of the study assumes that only one concept
Therefore,
for each
Study is, in reality two studies;
assigned to one of these populations
The structure
or less and
must be made separately
for use of PC postage.
Of the total of four hundred
Study
but they represent
concerning
for one concept may represent
half
all of the sample considering
that
potential impact are thus projected
for each concept.
The concept estimates
9
to the full
are not additive,
1
they are independent.
2
these
3
available.
4
studies,
Since each concept
no means
Of the population
of establishing
is alone in the world described
overlap
between
of 9.008 million establishments
5
less, 5.415 million (60%) qualified for this study.
6
establishments
7
The PC Postage concept statement for the Non-Household
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
with 26-50 employees,
in
the concepts
is
with 25 employees
or
Of the population
of 246,604
123,006 (50%) qualified.
Study was:
As you may know, there are ways to affix postage to First-Class Mail you
send through the US Postal Service besides using postage stamps. One way
to do it is electronically from your computer via the Internet using standard
With Internet
word-processing
programs and an inkjet or laser printer.
postage..
You can refill postage over the Internet or by phone 24 hours a day, 7
days a week, so you no longer have to make trips to the Post Office to
purchase stamps
Postage can be printed directly from your printer onto envelopes or
labels as they are being addressed using standard word-processing
programs
An Internet metering system keeps track of the postage you use, so
you have an accurate record of postage expenses for tax purposes
Barcoding can be applied to your mail simultaneously, ensuring faster
and more accurate mail processing
Your mail projects a more professional “business” image, and each
piece is dated
10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
.
And the postage meter concept statement was:
As you may know, there are ways to affix postage to First-Class Mail you
send through the US Postal Service besides using postage stamps. One way
to do it is to lease a postage meter. With a postage meter..
. You can refill postage on your meter via modem or phone 24 hours a
day, 7 days a week, so you no longer have to make trips to the Post
Office to purchase stamps
l
The meter handles small and large envelopes and prints pressuresensitive postage tapes for packages
l
A postage meter keeps track of the postage you use, so you have an
accurate record of postage expenses for tax purposes
l
Metered mail projects a more professional “business” image and dates
each piece
The intent question
was in much the same form as for the Household
Study, but adjusted to match the concept that was read.
01
If you could have all the features I’ve just described for a monthly
(lease cost of less than $20/access fee of $5.00) plus the regular
cost of first-class postage (IF NECESSARY: 33 cents for a regular
first-class letter), how likely would you be to (lease a postage
meter/subscribe
to the Internet postage service)?
Please use a
scale where “5” means you would be extremely likely to (lease a
postage meter/subscribe to the service) and “1” means you would
be not at all likely to do so. Of course, you may use any number
from 1 to 5.
11
1
The responses
to the intent questions were as follows:
Number of Employees
25 or less
26-50
93
97
66
99
PC Postage
Unweighted Base
Weighted Base
5 Extremely likely
4
3
2
1 Not at all likely
6 Don’t know/Refused
5%
5
13
12
63
3
7%
3
13
6
67
4
Number of Employees
25 or less
26-50
Postage Meters
Unweighted Base
Weighted Base
5 Extremely likely
4
3
2
1 Not at all likely
6 Don’t know/Refused
107
114
0%
1
10
9
103
101
80
2%
3
7
4
82
1
1
n
L
3
As in the Household
Study,
those
who
did not say they
4
“extremely
5
included reference to a per piece discount of one cent on First-Class
would
be
likely” to subscribe to the service were asked a similar question which
12
postage.
1
Number of Employees
26-50
25 or less
92
88
92
82
PC Postage
Unweighted Base
Weighted Base
5 Extremely likely
4
3
2
1 Not at all likely
6 Don’t know/Refused
25%
8
12
5
49
1
29%
13
12
10
32
4
Number of Employees
25 or less
26-50
Postage Meters
Unweighted Base
Weighted Base
5 Extremely likely
4
3
2
1 Not at all likely
6 Don’t know/Refused
107
114
23%
6
18
8
41
3
100
99
30%
11
9
8
40
2
2
3
4
The detailed
calculations
of the potential
affected resulted in the following estimates.
13
First-Class
Mail volume to be
1
Calculation
2
of Pieces of First-Class Mail Affected
No discount
One cent discount
Total
PC Postage
25 employees or less
26-50 employees
216 million
29 million
2.3 billion
71 million
2.5 billion
100 million
No discount
One cent discount
Total
Postage Meter
25 employees or less
26-50 employees
0
7 million
3.4 billion
1 1 1 million
3.4 billion
1 18 million
The detailed calculations
are available in the Appendix.
3
14
APPENDIX
1
2
PC Postage (Households)
X
- No Discount
Sampled households extremely likely to subscribe to
the PC postage service at no discount (7%)
13 respondents
.8 intent estimate adjustment
10 respondents
5% of sampled
households
Qualifying households
40.8 million
X
Adjusted percent extremely likely to subscribe to the
PC postage service at no discount (5%)
X
Average pieces of mail sent per household
per year
Total estimated pieces affected
2.0 million
168
336 million
3
Those who did not say they were extremely
4
were those asked the intent question with the addition of a one cent per piece
5
discount.
15
likely to subscribe
to the service
1
The volume estimates for a one-cent discount are shown below.
X
Sampled households extremely likely to subscribe to
the PC postage service at a one-cent discount (10%)
19 respondents
.8 intent estimate adjustment
15 respondents
8% of sampled
households
+ Assumption that sampled households who would
subscribe at no discount would do so at one cent
Qualifying
Households
5% of sampled
households
40.8 million
X
Adjusted percent extremely likely to subscribe to the
PC Postage at a one-cent discount (13%)
X
Weighted average pieces of mail sent per household
per year (168 pieces sent by those who would convert
at no discount and 324 pieces sent by those who
would convert at one-cent discount but not at no
discount)
Total estimated pieces affected
2
16
5.3 million
271
1 ,436 billion
1
2
PC Postage (25 Employees
X
or Less) - No Discount
Sampled businesses extremely likely to subscribe to
the PC postage service at no discount (5%)
4 respondents
.8 intent estimate adjustment
3 respondents
4% of sampled
businesses
Qualifying
5.4 million
businesses
X
Adjusted percent extremely likely to subscribe to the
PC Postage at no discount (4%)
X
Average pieces of mail sent per business per year
Total estimated pieces affected
3
PC Postage (26-50 Employees)
X
216,000
1,000
216 million
- No Discount
Sampled businesses extremely likely to subscribe to
the PC postage service at no discount (7%)
6 respondents
.8 intent estimate adjustment
5 respondents
5% of sampled
businesses
123,000
Qualifying businesses
X
Adjusted percent extremely likely to subscribe to the
PC Postage at no discount (5%)
6,150
X
Average pieces of mail sent per business per year
4,750
Total estimated pieces affected
17
29 million
1
PC Postage (25 Employees
X
or Less) - One-Cent Discount
Sampled businesses extremely likely to subscribe to
the PC postage service at a one-cent discount (25%)
21 respondents
.8 intent estimate adjustment
16 respondents
19% of sampled
businesses
+ Assumption that sampled businesses who would
subscribe at no discount would do so at one cent
4% of sampled
businesses
Qualifying businesses
5.4 million
X
Adjusted percent extremely likely to subscribe to the
PC Postage at a one-cent discount (23%)
1.2 million
X
Weighted average pieces of mail sent per household
per year (1,000 pieces sent by those who would
convert at no discount and 2,250 pieces sent by
those who would convert at one-cent discount but not
at no discount)
Total estimated pieces affected
2
18
2,103
2.5 billion
1
PC Postage (26-50 Employees)
X
- One-Cent biscount
Sampled businesses extremely likely to subscribe to
the PC postage service at a one-cent discount (29%)
27 respondents
.8 intent estimate adjustment
22 respondents
23% of sampled
businesses
+ Assumption that sampled businesses who would
subscribe at no discount would do so at one cent
5% of sampled
businesses
Qualifying businesses
123,000
X
Adjusted percent extremely likely to subscribe to the
PC Postage at a one-cent discount (28%)
34,440
X
Weighted average pieces of mail sent per business
per year (4,750 pieces sent by those who would
convert at no discount and 2,500 pieces sent by
those who would convert at one-cent discount but not
at no discount)
2,904
Total estimated pieces affected
19
100 million
1
Postage Meter (25 Employees
or Less) - No Discount
Sampled businesses extremely likely to subscribe to
the PC postage service at no discount (0%)
X
0 respondents
.8 intent estimate adjustment
Qualifying businesses
X
Adjusted percent extremely likely to subscribe to the
PC Postage at no discount (4%)
X
Average pieces of mail sent per business per year
Total estimated pieces affected
2
Postage Meter (26-50 Employees)
X
0
- No Discount
Sampled businesses extremely likely to subscribe to
the PC postage service at no discount (2%)
2 respondents
.8 intent estimate adjustment
2 respondents
2% of sampled
businesses
Qualifying businesses
123,000
X
Adjusted percent extremely likely to subscribe to the
PC Postage at no discount (2%)
2,460
X
Average pieces of mail sent per business per year
2,750
Total estimated pieces: affected
3
20
6.8 million
1
Postage Meter (25 Employees
X
or Less) - One-Cent
Discount
Sampled businesses extremely likely to subscribe to
the PC postage sewice at a one-cent discount (23%)
26 respondents
.8 intent estimate adjustment
21 respondents
18% of sampled
businesses
+ Assumption that sampled businesses who would
subscribe at no discount would do so at one cent
Qualifying businesses
0% of sampled
businesses
5.4 million
X
Adjusted percent extremely likely to subscribe to the
PC Postage at a one-cent discount (18%)
X
Average pieces of mail sent per business per year
Total estimated pieces affected
2
21
972,000
3,500
3.4 billion
1
Postage Meter (26-50 Employees)
X
- One-Cent
Discount
Sampled businesses extremely likely to subscribe to
the PC postage service at a one-cent discount (30%)
30 respondents
.8 intent estimate adjustment
24 respondents
24% of sampled
businesses
+ Assumption that sampled businesses who would
subscribe at no discount would do so at one cent
4% of sampled
businesses
Qualifying businesses
123,000
X
Adjusted percent extremely likely to subscribe to the
PC Postage at a one-cent discount (28%)
34,440
X
Weighted average pieces of mail sent per business
per year(2,750 pieces sent by those who would
convert at no discount and 3,750 pieces sent by
those who would convert at one-cent discount but not
at no discount)
3,430
Total estimated pieces affected
2
22
1 18 million