DPI Secuprint Inc.

National Labor Relations Board - Board Decisions

Aug 20, 2015

362 NLRB No. 172 (N.L.R.B. 2015)

362 NLRB No. 172
NOTICE: This opinion is subject to formal revision before publication in the
bound volumes of NLRB decisions. Readers are requested to notify the Ex-
ecutive Secretary, National Labor Relations Board, Washington, D.C.
20570, of any typographical or other formal errors so that corrections can
be included in the bound volumes.
DPI Secuprint, Inc. and Graphic Communications
Conference/International Brotherhood of Team-
sters, Local 503-M, Petitioner. Case 03–RC–
012019
August 20, 2015
DECISION ON REVIEW AND ORDER
BY CHAIRMAN PEARCE AND MEMBERS HIROZAWA
AND JOHNSON
On May 20, 2011, the Acting Regional Director for
Region 3 issued a Decision and Direction of Election, in
which he found that the petitioned-for unit of hourly pre-
press, digital press, offset bindery, digital bindery, and
shipping and receiving employees at the Employer’s
Rochester, New York facility, was appropriate. Thereaf-
ter, in accordance with Section 102.67 of the Board’s
Rules and Regulations, the Employer filed a timely re-
quest for review. The Employer contends that the Acting
Regional Director erred in approving the petitioned-for
unit because it excludes hourly offset-press employees—
the press operators and feeder-tenders. The Petitioner
filed an opposition.
On June 16, 2011, the Board granted the Employer’s
request for review. Thereafter, the Employer and Peti-
tioner filed briefs on review, as did amicus the Graphic
Communications Conference of the International Broth-
erhood of Teamsters (GCC/IBT). The Employer also
filed a brief in response to GCC/IBT’s amicus brief.
The Board has delegated its authority in this proceed-
ing to a three-member panel. We have carefully consid-
ered the entire record in this proceeding, including the
briefs on review. For the reasons set forth in the Acting
Regional Director’s decision and the additional reasons
set forth below, we affirm the Acting Regional Director’s
finding that the petitioned-for unit is appropriate and that
the offset-press employees need not be included in the
unit.
Facts
The Employer is engaged in commercial printing, with
an emphasis on security printing products such as vehicle
titles and birth certificates. The Employer employs 12
salaried and 20 hourly employees. Two nonsupervisory
hourly employees work in the prepress department, one
in digital press, seven in offset press, two in the digital
bindery, four in the offset bindery, and three in shipping
and receiving. For both offset and digital printing, the
process begins in prepress, where images and text sup-
plied by the customer are used to create proofs. For off-
set printing, once the customer approves the proofs, pre-
press employees create plates and place them in a rack in
the offset department. The offset-press employees pick
up the plates left by the prepress employees, hang the
plates on one of the two offset presses the Employer
owns (the Komori and the GTO), use the presses to print
text and images on blank paper, and prepare printed
sheets for pickup by offset-bindery employees. Offset-
bindery employees cut, fold, or stitch the printed paper,
depending on the specifications of the job. Shipping and
receiving employees shrink wrap and palletize completed
jobs, fill out packing slips, and load finished product
onto delivery trucks. For digital printing, once the cus-
tomer approves the proofs, prepress employees transmit a
digital file to the digital press department. The digital-
press employee prints on paper from the digital file, and
production is then completed in the digital-bindery de-
partment before the product goes to shipping and receiv-
ing.
The entire facility is arranged to accommodate work-
flow. The prepress department is located at one end of
the building, with doors connecting it to the digital room
(which houses both the digital press and digital bindery
equipment) and the offset press room (which houses both
the offset press and offset-bindery equipment). A door
connects the digital room and the offset press room; an-
other door leads from the offset press room to the ship-
ping and receiving room.1
The departments in the proposed unit operate Monday
through Friday. The prepress and digital-bindery de-
partments operate on two shifts, offset bindery and ship-
ping and receiving operate only on the first shift, and the
digital press operates only on the second shift. Although
there is some variation in the exact hours, all but one of
the first-shift employees in the proposed unit begin work
between 7 and 8:30 a.m.; that employee, in offset bind-
ing, begins work at 5 a.m. The record does not specify
the exact hours of the second shift, but it appears to end
at midnight. The offset-bindery and the shipping and
receiving employees report directly to Jason Colline, the
Employer’s vice president of operations. The first-shift
prepress and digital bindery employees report to other
supervisors, who in turn report to Colline. All second-
shift employees report to the second-shift plant manager,
Mike Remmerly, who reports to Colline.
The offset-press department operates on a different,
two-shift schedule. The first shift runs from 5 a.m. to 3
p.m., and the second from 3 p.m. to 1 a.m. The offset

1 Emp. Exh. 2, showing the layout of the facility, is attached as an
appendix.
2 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
press also has a “weekend shift” that runs from 6 a.m. to
6 p.m. on Fridays, Saturdays, and Sundays.
There are a total of three offset-press operators and
three feeder-tenders (or press helpers) who work on the
Komari press, plus an operator for the GTO press. Press
operators set up the presses (by, among other things, se-
lecting the proper paper stock, inks, and plates, and mix-
ing ink colors as needed), operate the presses, ensure
proper print quality, make adjustments to the presses
during printing, and perform maintenance on the presses.
Feeder-tenders load paper into the press and assist the
press operators as needed. According to Vice President
Colline, feeder-tenders are capable of operating the
presses, because press operator and feeder-tender respon-
sibilities overlap. Colline also testified that “a lot of”
offset-press employees “come from that feeder tender
background.” Offset-press employees working the first
shift report to Colline, while those on the second report
to Remmely. It is not clear who employees report to on
the weekend shift, but it is either Colline or Remmely.
All company employees are trained on the job, alt-
hough some employees come with prior experience.
Offset-press operators generally come with prior experi-
ence; feeder-tenders and bindery employees may or may
not have past experience. It takes longer for the Employ-
er to train offset-press operators than any other employ-
ees.2
Apart from testimony that some feeder-tenders have
become offset-press operators, there is no evidence that
other employees have ever permanently or temporarily
transferred from one job classification to another. Em-
ployees from one department do, however, assist em-
ployees in other departments on an ad hoc basis. Within
the petitioned-for unit, employees in each of the depart-
ments have assisted in at least one other department.3
There is no evidence, however, that employees from oth-
er departments have worked in the offset-press depart-
ment, and the Employer testified that shipping and bind-
ery employees do not possess the requisite skill to per-
form offset-press work.
Employees from the various departments regularly
come into contact with each other. Press employees
(both digital and offset) will discuss particular jobs with

2 Offset-bindery employee Robert Schultz testified that it can take
“months” to learn to operate offset-bindery equipment. Colline testi-
fied that an inexperienced but “really, really sharp” employee could
learn to operate one of the offset presses in 6 months. The record does
not indicate the amount of training necessary for employees with no
past experience to be trained to do prepress, digital press, digital bind-
ery, feeder-tender, or shipping and receiving work.
3 In recent years, the Employer’s welcome letter to newly hired em-
ployees has stated that they are expected to assist in other departments
as needed.
prepress employees (when there are questions about a
particular job, or when a plate or digital file appears to
have a problem), and will consult with bindery employ-
ees (to see if a problematic printing run can be salvaged,
for instance); such interactions take place on a daily ba-
sis. Similarly, shipping and receiving employees deliver
supplies and materials to all the various departments, and
employees from other departments may go to shipping
and receiving to retrieve necessary materials.
There is no specific pay rate for any position, but the
record indicates that digital and offset-press operators are
paid about $20 per hour, feeder-tenders about $16, bind-
ery employees between $16 and $18, shipping and re-
ceiving employees between $10 and $17, and prepress
employees between $15 and $20. All employees receive
the same health benefits, holiday pay, and 401(k) plan,
and all employees are subject to the same general poli-
cies and operating procedures manual. All hourly em-
ployees use the same entrance, timeclock, and lunch-
room.4 The Employer periodically holds meetings for all
employees, e.g., to discuss production problems.5 On
slow days, shipping and receiving employees, bindery
employees, and feeder-tenders have been sent home ear-
ly, but offset-press operators are not, unless they ask to
be.
The Acting Regional Director’s Decision
The Acting Regional Director found that the peti-
tioned-for employees shared a sufficient community of
interest to constitute an appropriate unit. Specifically, he
determined that there is functional integration among
their departments, as each handles an aspect of producing
a single product; that there is a high degree of contact
among the petitioned-for employees, and some ad hoc
job interchange; that although the skills and functions of
the various petitioned-for employee classifications differ,
none requires any prior training; that all petitioned-for
employees have at least a common second-level supervi-
sor (Colline); and that the petitioned-for employees share
roughly similar wages, hours, benefits, and working con-
ditions.
The Acting Regional Director then considered but re-
jected the Employer’s argument that the offset-press em-
ployees must be included in the unit. The Acting Re-
gional Director reasoned that the offset-press operators
are more highly skilled than the other employees, that it
takes longer to train an offset press operator than it does

4 The offset press, bindery, and shipping and receiving employees
also share a locker room.
5 In the past, the Employer has also hosted social occasions (such as
Christmas parties or summer picnics) to which all employees are invit-
ed, but Colline testified that the last such occasion was several years
ago.
DPI SECUPRINT, INC. 3
other employees, that offset-press employees work dif-
ferent schedules than the petitioned-for employees, and
that offset-press operators are treated differently when
work is slow: they are allowed to stay on the job while
other employees are sent home. Based on the foregoing
considerations, the Acting Regional Director found that
the offset-press employees do not share an overwhelming
community of interest with the petitioned-for employees
such that the petitioned-for unit is inappropriate without
them.
The Acting Regional Director also rejected the Em-
ployer’s argument that AGI Klearfold, LLC, 350 NLRB
538 (2007), and Moore Business Forms, Inc., 216 NLRB
833 (1975), require that the offset-press employees be
included in the unit. In both cases, the Board held that an
offset-press-only unit was inappropriate because it ex-
cluded prepress employees. The Acting Regional Direc-
tor acknowledged that, in keeping with those cases, the
Board generally finds units of offset-press and prepress
employees—which the Board refers to as the “traditional
lithographic unit”—appropriate. But he observed that
AGI Klearfold states only that “appropriate weight” be
given to the traditional lithographic unit, and that the
Board has found both press/prepress and press-only units
appropriate in particular cases. See, e.g., NTA Graphics,
Inc., 307 NLRB No. 224 (1992) (not reported in Board
volumes), enfd. 996 F.2d 1216 (6th Cir. 1993), judgment
vacated on other grounds 511 U.S. 1124 (1994); Conti-
nental Web Press, 262 NLRB 1395 (1982), enf. denied
742 F.2d 1087 (7th Cir. 1984).6 The Acting Regional
Director also pointed out that in AGI Klearfold and
Moore Business Forms, the pressmen regularly entered
the prepress room for various reasons. The Acting Re-
gional Director stated that, in the present case, by con-
trast, there is no evidence of comparable contact between
offset-press employees and the petitioned-for employees,
or of offset-press employees performing any duties in
other departments, and the offset-press employees work
different shifts from the other employees. The Acting
Regional Director also observed that the Employer was
not arguing that the prepress employees should be in-
cluded with the offset-press employees, but rather was
seeking a unit comprising all of its employees. He there-
fore found that, even according the appropriate weight to
the traditional lithographic unit, the offset-press employ-
ees need not be included in the unit.

6 The Supreme Court vacated the judgment in NTA Graphics be-
cause the union had disclaimed interest in representing the employees
at issue. See NTA Graphics, 316 NLRB 25 (1995).
For the reasons that follow, we agree that the peti-
tioned-for unit is appropriate and need not include offset-
press employees.
Analysis
The Board’s decision in Specialty Healthcare & Reha-
bilitation Center of Mobile, 357 NLRB No. 83 (2011),
enfd. sub nom. Kindred Nursing Centers East, LLC v.
NLRB, 727 F.3d 552 (6th Cir. 2013), which issued sub-
sequent to the Acting Regional Director’s decision, sets
forth the principles that apply in cases like this one, in
which a party contends that the smallest appropriate bar-
gaining unit must include employees or job classifica-
tions not included in the petitioned-for unit. As ex-
plained in that decision, when a union seeks to represent
a unit of employees “who are readily identifiable as a
group (based on job classifications, departments, func-
tions, work locations, skills, or similar factors), and the
Board finds that the employees in the group share a
community of interest after considering the traditional
criteria, the Board will find the petitioned-for unit to be
an appropriate unit. . . .” See id., slip op. at 12.
If the petitioned-for unit satisfies that standard, the
burden is then on the proponent of a larger unit to
demonstrate that the additional employees it seeks to
include share an “overwhelming community of interest”
with the petitioned-for employees, such that there “is no
legitimate basis upon which to exclude certain employ-
ees from” the larger unit because the traditional commu-
nity of interest factors “overlap almost completely.” Id.,
slip op at 11–13 and fn. 28 (quoting Blue Man Vegas,
LLC v. NLRB, 529 F.3d 417, 422 (D.C. Cir. 2008));
Odwalla, Inc., 357 NLRB No. 132, slip op. at 4 (2011);
Northrop Grumman Ship Yard, 357 NLRB No. 163, slip
op. at 3 (2011); DTG Operations, Inc., 357 NLRB No.
175, slip op. at 4 (2011); Macy’s, Inc., 361 NLRB No. 4,
slip op. at 7 (2014).
7

7 The dissent argues that the Specialty Healthcare standard is flawed.
We need not address that argument at length. Specialty Healthcare was
enforced by the United States Court of Appeals for the Sixth Circuit,
and the “overwhelming community of interest” standard has been en-
dorsed by the District of Columbia Circuit. See Kindred Nursing Cen-
ters East, LLC v. NLRB, 727 F.3d 552 (6th Cir. 2013); Blue Man Ve-
gas, supra at 421–422. Both of those courts, as well as the Board, have
also squarely rejected the dissent’s argument that the overwhelming
community-of-interest test contravenes Sec. 9(c)(5) of the Act, which
provides that, in determining an appropriate unit, “the extent to which
the employees have organized shall not be controlling.” See Kindred,
supra at 563–565 (discussing the issue at length); Blue Man Vegas,
supra at 423; Macy’s, supra, slip op. at 18 fn. 71; Specialty Healthcare,
supra, slip op. at 8–9. As the court stated in Blue Man Vegas, “[a]s
long as the Board applies the overwhelming community-of-interest
standard only after the proposed unit has been shown to be prima facie
appropriate, the Board does not run afoul” of Sec. 9(c)(5). 529 F.3d at
423. Here, the Regional Director made an initial determination, and we
4 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
In any particular workplace, there may be a number of
ways in which the employees could be appropriately
grouped for collective bargaining. See Overnite Trans-
portation Co., 322 NLRB 723 (1996). Therefore,
“demonstrating that another unit containing the employ-
ees in the proposed unit plus others is appropriate, or
even that it is more appropriate, is not sufficient to
demonstrate that the proposed [smaller] unit is inappro-
priate.” Specialty Healthcare, supra, slip op. at 10.8
Applying these principles, we find that the employees
in the petitioned-for unit are a readily identifiable group
who share a community of interest, and that the Employ-
er has not demonstrated that the offset-press employees
share an overwhelming community of interest with the
petitioned-for employees.9

agree, that the unit was prima facie appropriate based on traditional
community-of-interest factors; indeed, the Employer does not dispute
that the employees in the petitioned-for unit share a community of
interest.
Similarly, the Board and the courts have rejected the dissent’s argu-
ment that Specialty Healthcare is “irreconcilable” with Sec. 9(b), which
provides that the Board decide the appropriate unit “in each case . . . in
order to assure to employees the fullest freedom in exercising” their
rights under the Act. Macy’s, supra, slip op. at 18 fn. 72; see American
Hospital Assn. v. NLRB, 499 U.S. 606, 610–614 (1991) (“in each case”
simply means that whenever parties disagree over union appropriate-
ness, the Board shall resolve the dispute, and the imposition of a rule
defining appropriate units in acute care hospitals does not run afoul of
the “in each case” command so long as the Board applies it in each
case).
8 See also American Hospital Assn., supra at 610 (“employees may
seek to organize ‘a unit’ that is ‘appropriate’—not necessarily the sin-
gle most appropriate unit”) (emphasis in original; quoting Sec. 9(b));
Morand Bros. Beverage Co., 91 NLRB 409, 418 (1950) (“There is
nothing in the statute which requires that the unit for bargaining be the
only appropriate unit, or the ultimate unit, or the most appropriate unit;
the Act requires only that the unit be ‘appropriate.’”) (emphasis in
original), enfd. 190 F.2d 576 (7th Cir. 1951), cert. denied 346 U.S. 909
(1953).
9 Specialty Healthcare clarified that it is the nonpetitioning party—
here, the Employer—who bears the burden of demonstrating an over-
whelming community of interest among employees in a group larger
than an otherwise appropriate petitioned-for unit. Id., slip op. at 12–13
fn. 28. And it is the Board’s usual rule in representation cases to apply
its decisions retroactively, including to all pending cases. See SNE
Enterprises, 344 NLRB 673, 673–674 (2005). Even assuming, howev-
er, that Specialty Healthcare changed the law in this respect, we find
that imposing the burden of proof on the Employer here is not a retro-
active change that “work[s] a ‘manifest injustice.’” Id. at 673 (citations
omitted); see also Northrop Grumman, supra, slip op. at 3 fn. 8. First,
there was no “significant departure from a well-settled area of the law.”
SNE Enterprises, supra at 674. Second, as in SNE Enterprises, there is
no evidence that the Employer relied on contrary precedent; indeed, the
Employer presented extensive evidence aimed at demonstrating the
extent of the community of interest between the offset-press employees
and the employees in the petitioned-for unit. Finally, the Employer’s
brief in response to GCC/IBT’s amicus brief expressly argues that the
Employer met the burden articulated in Specialty Healthcare. Accord-
ingly, although the Acting Regional Director’s decision may not have
To begin, the petitioned-for employees are “readily
identifiable as a group.” They are all the hourly employ-
ees in the prepress, digital press, bindery, and shipping
and receiving departments—in short, all the hourly em-
ployees who do not work on the offset presses. Thus, the
petitioned-for employees are readily identifiable as a
group based on departments and functions.10 See Berg-
dorf Goodman, 361 NLRB No. 11, slip op. at 2 (2014);
Specialty Healthcare, supra, slip op. at 12.
As the Acting Regional Director found, the petitioned-
for employees also share a community of interest. In
making this determination, the Board examines
whether the employees are organized into a separate
department; have distinct skills and training; have dis-
tinct job functions and perform distinct work, including
inquiry into the amount and type of job overlap be-
tween classifications; are functionally integrated with
the Employer’s other employees; have frequent contact
with other employees; interchange with other employ-
ees; have distinct terms and conditions of employment;
and are separately supervised.

clearly allocated the burden of proof on this issue, we apply that rule in
affirming his decision, and do so without prejudicing any party.
10 The dissent argues that the petitioned-for unit does not track de-
partmental lines because the Employer has drawn no such lines “around
this collection of employees.” It states that the unit is “a list of four
departments gathered by the petitioner” (emphasis in dissent) and
therefore not an “identifiable group.”
This argument is flawed in two respects. First, the Act directs the
Board to make appropriate unit determinations that “assure to employ-
ees the fullest freedom in exercising the rights guaranteed by this Act.”
See Sec. 9(b). In doing so, “the petitioner’s desire[] as to the unit is
always a relevant consideration.’” Marks Oxygen Co. of Alabama, 147
NLRB 228, 230 (1964). And Sec. 9(a), “read in light of the policy of
the Act, implies that the initiative in selecting an appropriate unit re-
sides with the employees.” American Hospital Assn., supra at 610; see
also Overnite, supra at 724 (“In deciding the appropriate unit, the Board
first considers the union’s petition and whether that unit is appropri-
ate.”) As explained in fn. 7 above, these principles in no way contra-
vene the requirement in Sec. 9(c)(5) that the extent of organizing “shall
not be controlling.”
Second, the dissent misapprehends Specialty Healthcare’s prelimi-
nary requirement, that the petitioned-for employees be readily identifi-
able as a group. “Readily identifiable as a group” is not, as the dissent
seems to suggest, another version of the community-of-interest analy-
sis. It means simply that the description of the unit is sufficient to
specify the group of employees the petitioner seeks to include. Here,
the employees in the petitioned-for unit are readily identifiable as a
group based on the departments in which they work and their job func-
tions and classifications. Whether that group has enough in common
for bargaining collectively to make sense is addressed in the communi-
ty-of-interest analysis, and the Employer does not dispute that the peti-
tioned-for employees share a community of interest. Compare Berg-
dorf Goodman, supra, slip op. at 2 (“the petitioned-for employees are
readily identifiable as a group, [but] lack a community of interest”).
DPI SECUPRINT, INC. 5
Id., slip op. at 9 (quoting United Operations, Inc., 338
NLRB 123, 123 (2002)). Here, the petitioned-for unit “con-
form[s] to the departmental lines established by the
[E]mployer.” Bergdorf Goodman, supra, slip op. at 3. The
petitioned-for employees’ work is functionally integrated,
they have daily contact with each other and substitute for
one another on an ad hoc basis, and they have similar skill
levels, common supervision, and roughly equivalent wages,
hours, and working conditions.
The Employer does not deny that the petitioned-for
employees share a community of interest. Instead, as
stated above, it contends that the offset-press employees
share an overwhelming community of interest with the
petitioned-for employees, such that excluding the offset-
press employees would result in a “fractured” unit. See
Odwalla, supra, slip op. at 5. We do not agree.
As an initial matter, we acknowledge that the offset-
press employees share some community-of-interest fac-
tors with the petitioned-for employees, such as common
supervision and functional integration.11 The offset-press
employees also enjoy the same benefits as the petitioned-
for employees, and they have roughly similar pay rates.12
These commonalities, however, do not establish an
overwhelming community of interest. See DTG Opera-
tions, supra, slip op. at 7 (no overwhelming community
of interest despite common supervision, functional inte-
gration, and similar benefits and base wages).
Rather, other factors demonstrate that the offset-press
employees do not share an overwhelming community of
interest with other employees. First, there is no dispute
that the offset-press employees work in a separate de-
partment from the petitioned-for employees and that their
work requires greater skill and lengthier training. Alt-
hough there is considerable sharing of work at the facili-
ty, the offset-press employees are the only employees
who set up, operate, adjust, and maintain the offset print-
ing presses. The offset-press employees also work dif-
ferent hours from other employees. They work longer
shifts and are the only nonsupervisory employees who
work weekends. Unlike the other employees, the offset-
press operators are not sent home when work is slow.
In the face of these differences, the Employer’s argu-
ments that the offset-press employees must be included
in the unit are unpersuasive. First, the Employer sug-
gests that the unit should not include the digital press

11 As the Employer is primarily engaged in producing printed mate-
rials and utilizes a straight-line operation, we do not agree with the
Petitioner’s claim that the offset-press employees lack “meaningful
functional integration” with other employees.
12 The Petitioner claims that offset-press employees are paid more
than other employees, but the record shows that they are paid about $20
per hour and that prepress employees are paid $15 to $20 per hour.
operator while excluding feeder-tenders. The record,
however, does not establish that digital press operator
duties are similar to those of offset-press operators. In-
deed, Vice President Colline testified that the digital
printer is little more than a “glorified [photo]copier.”
Second, the Employer contends that the offset-press
employees must be included in the petitioned-for unit on
the basis of interchange and contact with the other em-
ployees. But concerning interchange, the record is clear
that any work the offset-press employees perform in oth-
er departments is infrequent and incidental to their pri-
mary duties.13 Moreover, as shown, any interchange runs
purely one way, as no employees in the petitioned-for
unit perform work in the offset-press department. Such
one-way “interchange” is not sufficient to establish an
overwhelming community of interest. See DTG Opera-
tions, supra, slip op. at 7.
As for contact, the Employer claims that offset-press
employees are in “constant” contact with bindery and
prepress employees, and argues that this “heavily” favors
including the offset-press employees in the unit. Alt-
hough we agree with the Employer that there is evidence
of regular contact, both production-related and informal,
between offset press and petitioned-for employees, such
contact in the absence of interchange does not establish
an overwhelming community of interest.14
The Employer’s remaining arguments are similarly
unconvincing. For example, the fact that bindery and
offset-press employees both set and adjust machines does
not establish that they share the same skill level. It is
undisputed that bindery machines and offset-press ma-
chines require distinct training; indeed, as Employer Vice
President Colline testified, no other employees are capa-
ble of operating the offset presses.
We also reject the Employer and dissent’s contention
that, without the offset-press employees, the petitioned-
for unit is “fractured,” as was the case in Odwalla, supra.
In Odwalla, the petitioned-for unit did not follow any

13 We nevertheless reject the Petitioner’s suggestion that to establish
interchange, the offset-press employees would occasionally have to
operate all of the production equipment or spend more than 30 percent
of their time outside their own department. The cases the Petitioner
cites do not impose such requirements; they merely show that such
facts demonstrate employee interchange weighing in favor of a com-
munity of interest. See Journal-Times Co., 209 NLRB 745, 747 (1974)
(all employees able to perform all jobs in particular department); Con-
tinental Can Co., 171 NLRB 798, 800 (1968) (employees spent about
30 percent of time performing other duties).
14 In each of the cases cited by the Employer, both interchange (or
shared functions) and contact were present. See Sears, Roebuck & Co.,
319 NLRB 607, 607–608 (1995) (shared job functions); Lifeline Mobile
Medics, 308 NLRB 1068, 1069–1070 (1992) (evidence of transfer
between classifications); Westin Hotel, 277 NLRB 1506, 1507 (1986)
(evidence of transfer policy).
6 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
lines drawn by the employer, such as classification, de-
partment, or function. The Board found “no rational
basis” for excluding certain employees—the merchandis-
ers—from the unit because no community-of-interest
factors suggested that the employees in the petitioned-for
unit shared a community of interest that the merchandis-
ers did not also share. Id., slip op. at 5. In fact, certain
classifications in the recommended unit had more in
common with the merchandisers than with another in-
cluded classification (route service representatives). See
id., slip op. at 5–6. The merchandisers worked in the
same department as other employees in the unit and
shared “very similar” functions with two of the included
classifications. Id., slip op. at 5. At the same time, the
petitioned-for unit included refurbishment center cooler
technicians, who worked in a functionally and structural-
ly separate part of the business, under separate supervi-
sion, from the rest of the included employees. Id., slip
op. at 2, 5.
As the foregoing discussion shows, the present case is
different: the offset-press employees work in a separate
department and share community-of-interest factors dis-
tinct from those of the petitioned-for employees. Noth-
ing in the record indicates that any of the employees in
the petitioned-for unit have more in common with the
offset-press employees than they do with other peti-
tioned-for employees.15

15 Our colleague notes the linear nature of the Employer’s operation
and contends that excluding the offset-press employee from the unit
“pluck[s] the heart from the production process.” He predicts a series
of dire consequences for the Employer’s ability to function. But the
nature of the Employer’s production process and the physical layout of
the plant do not make the petitioned-for unit inappropriate. This point
is demonstrated by the Board’s recent decision in AT Wall Co., 361
NLRB No. 62 (2014). In AT Wall, the workers in the employer’s new
Metalform department, a production department recently added to the
employer’s plant, did not share an overwhelming community of interest
with AT Wall’s unit employees, including two pre-existing production
departments (Tubing and Stamping) and four other departments (In-
spection, Maintenance, Materials Handling and Toolroom). The
Metalform department, like DPI’s offset-press department and the other
AT Wall production departments, occupied the middle of a linear oper-
ation: “Each manufacturing process begins with a material handler
moving the appropriate starting material to an inspection area where in
inspector verifies that it is ready to be placed in inventory for use. All
three processes end with a material handler moving the finished product
to the Employer’s shipping area.” Id., slip op. at 2. In addition, all
three production departments shared a common production floor, which
“consist[ed] mostly of open work areas marked off by yellow lines
painted on the floor.” Id.
While AT Wall was an accretion case, it used the same standard as
Specialty Healthcare, the overwhelming community-of-interest stand-
ard, to conclude that the existing unit remained appropriate notwith-
standing the exclusion of the Metalform employees. Thus, it illustrates
an important point: the exclusion of a department of employees from
the bargaining unit, even where they work in the same space and form
part of the same workflow as unit employees, will not necessarily ren-
Finally, we disagree with the Employer’s argument
that Board precedent concerning the “traditional litho-
graphic unit” requires the inclusion of offset-press em-
ployees in the unit. In cases involving the printing indus-
try, the Board has often stated that a unit comprising
press and pre-press employees—a “traditional litho-
graphic unit”—is an appropriate unit for bargaining. See
AGI Klearfold, 350 NLRB at 538. In most cases impli-
cating the traditional lithographic unit, a petitioner seeks
such a unit but another party contends that other employ-
ees must be added to the unit. See, e.g., Meyer Label
Co., 232 NLRB 933, 933–934 (1977); George Rice &
Sons, 212 NLRB 947, 947–948 (1974). Alternatively,
the traditional lithographic unit has been a consideration
in cases where a petitioner seeks a unit limited only to
press employees. See, e.g., AGI Klearfold, supra at 538;
Moore Business Forms, 216 NLRB at 833. Here, unlike
both of those scenarios, no party is seeking a press-only
unit, nor is any party seeking the traditional lithographic
unit.
The Employer nevertheless argues that if the peti-
tioned-for unit is found appropriate, it will have the ef-
fect of isolating the offset-press employees, which is
inconsistent with the traditional lithographic unit analy-
sis. We are aware of no case that stands for that proposi-
tion, but even if traditional lithographic unit principles
applied to this case, we would reach the same result. As
Specialty Healthcare makes clear, it does not displace
the industry-specific presumptions and rules that the
Board has developed over time. See Specialty
Healthcare, supra, slip op. at 13 fn. 29; Northrop Grum-
man, supra, slip op. at 4. But the traditional lithographic
unit is neither a presumption nor a rule. Instead, where
implicated, it is simply entitled to “appropriate weight.”
See AGI Klearfold, supra at 540.16 Thus, even if it ap-
plied here, the traditional lithographic unit would be but
one factor to consider. It does not establish the requisite
overwhelming community of interest between offset-
press employees and the petitioned-for employees.
The traditional lithographic unit cases cited by the
Employer do not warrant a contrary result. In both
Moore Business Forms and AGI Klearfold, the Board

der the remaining unit inappropriate, nor will it necessarily impede the
employer’s ability to manage its business effectively. Notably, in AT
Wall, it was the employer that argued against including the Metalform
employees in the unit, despite the fact that all of the production pro-
cesses took place on the same production floor, Metalform employees
and unit employees used the same common areas, and the unit employ-
ees’ work both preceded and followed that of the Metalform employees
in the linear production process.
16 The Employer is simply wrong when it states that AGI Klearfold
stands for the proposition that press employees cannot be separated
from other lithographic employees who share a community of interest.
DPI SECUPRINT, INC. 7
found that a press-only unit was inappropriate and that
the smallest appropriate unit was the traditional litho-
graphic unit. In AGI Klearfold, this finding was based
largely on the fact that press employees regularly entered
the prepress room to search for missing items, entered
the prepress room to solve problems when a prepress
supervisor was not on duty, and even made plates when
necessary. See AGI Klearfold, supra at 539–540. In
Moore Business Forms, the press employees similarly
entered the prepress area to correct and cut plates and
performed prepress work on the graveyard shift (when no
prepress employees were at work); the prepress employ-
ees also performed work in the press area on special col-
or or problem jobs and were assigned tasks in the press
area. See Moore Business Forms, supra at 833–834. In
the present case, by contrast, offset-press employees may
occasionally consult with prepress employees about
printing problems, but there is no evidence that the off-
set-press employees regularly enter the prepress area to
search for materials, solve problems when a prepress
supervisor is absent, or perform any prepress duties such
as cutting plates. Likewise, apart from consulting with
bindery employees about printing issues, there is no evi-
dence that the offset-press employees regularly enter the
digital press, digital bindery, offset bindery, or shipping
and receiving areas, nor is there evidence that offset-
press employees perform these departments’ functions
(apart from the limited assistance feeder-tenders lend to
the offset bindery). Finally, in contrast to Moore Busi-
ness Forms, there is no evidence that prepress or any
other employees have ever been assigned tasks in the
offset-press department.
Conclusion
We have little doubt that the offset-press employees
share a community of interest with the petitioned-for
employees. And if they do, a unit including the offset-
press employees would be an appropriate unit, or perhaps
even a more appropriate unit. But that is not, and has
never been, the relevant question. The Act requires only
that the unit be “appropriate,” and the petitioned-for unit
satisfies that standard. It is undisputed that the employ-
ees in the petitioned-for unit constitute an identifiable
group and share a community of interest, and the Em-
ployer has not carried its burden of proving that the off-
set-press employees share an overwhelming community
of interest with them. We therefore find that the peti-
tioned-for unit is an appropriate unit for bargaining.
ORDER
The Acting Regional Director’s finding that the peti-
tioned-for unit is appropriate is affirmed, and the case is
remanded to the Regional Director for further appropriate
action consistent with this Decision on Review and Or-
der.
Dated, Washington, D.C. August 20, 2015
______________________________________
Mark Gaston Pearce, Chairman
______________________________________
Kent Y. Hirozawa, Member
(SEAL) NATIONAL LABOR RELATIONS BOARD
8 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
APPENDIX
DPI SECUPRINT, INC. 9
MEMBER JOHNSON dissenting.
Eighty years ago, Francis Biddle, Chairman of the pre-
Wagner Act National Labor Relations Board as original-
ly established under Public Resolution 44, and one of the
architects of the 1935 Act, warned us of the precise dan-
ger presented by the Board’s decision today. Chairman
Biddle made clear at the Senate committee hearings prior
to passage of the Wagner Act that authority to determine
appropriate bargaining units must be vested with the
Board to avoid, on the one hand, manipulative gerry-
mandering were employers to make the decision, and, on
the other, destabilizing proliferation of units were the
decision left to employees. “If the employees themselves
make the decision without proper consideration of the
elements which should constitute the appropriate unit,
they could in any given instance defeat the practical sig-
nificance of the majority rule; and, by breaking off into
small groups, could make it impossible for the employer
to run his plant.” 1 He further recognized then that there
was always the risk “of your Board gerrymandering and
not carrying out the purposes of the Board,” but noted
that “any arbitrary act of the Board in selecting the unit is
subject to check on review by the court.”
As explained below, I dissent from today’s decision
both because it approves an inappropriate unit too narrow
in scope for bargaining, and because the manner in which
my colleagues conducted their analysis illustrates the
type of arbitrary gerrymandering that Chairman Biddle
was referring to in 1935. Additionally, Specialty
Healthcare & Rehabilitation Center of Mobile,2 as evi-
dent from the successive cases applying it, encourages
destabilizing proliferation of fractured units and mini-
mizes any meaningful analysis of the community-of-
interest factors in making unit determinations.
In making unit determinations, the Board over several
decades has applied our multifactor test to ensure, in
each case, that a petitioned-for unit of employees shares
a “community of interest” as distinct from other employ-
ees in the workplace, so that they comprise a unit appro-
priate for bargaining.3 Over the decades that the Board
has applied and refined the analysis in various workplac-

1 1 Leg. Hist. 1458–1459 (NLRA 1935) (statement of Francis Bid-
dle).
It was with the passage of the LMRA in 1947 that Congress added
Sec. 9(c)(5) to the Act, stating: “In determining whether a unit is ap-
propriate . . . the extent to which the employees have organized shall
not be controlling.” 29 U.S.C. § 159(c)(5) (emphasis added). See
Member Miscimarra’s discussion of the legislative history in Macy’s,
361 NLRB No. 4, slip op. at 25–26 (2014) (Member Miscimarra, dis-
senting).
2 357 NLRB No. 83 (2011), enfd. sub nom. Kindred Nursing Centers
East, LLC v. NLRB, 727 F.3d 552 (6th Cir. 2013).
3 E.g., Wheeling Island Gaming, 355 NLRB 637, 637 fn. 10 (2010).
es and industries, there has emerged a coherent body of
law that has generally guided this agency to results that
are intelligible and predictable based on the standard
itself, rather than the panel applying it.
It must be understood that our obligation to develop
and apply coherent legal standards requires ceding a cer-
tain amount of discretion in the interests of viable and
understood law, even where a panel would personally
prefer that law led to different results. That necessary
ceding of discretion is the tension at the heart of good
governance. And the agency has, at times, struggled
with the constraints imposed by clear legal standards
against the otherwise seductive freedom of purely result-
driven, and therefore arbitrary, action. Unfortunately,
today’s decision moves the ticker backward on the spec-
trum from clarity toward such arbitrary action, from un-
derstood standards to outcome-driven rationalizations.
To be sure, the Board is vested with the discretion to
interpret the statute and accordingly adjust and clarify
standards so that they can effectively evolve with the
changing American workplace. But shifts in the way we
construe and apply the Act can only be deemed clarifica-
tion if they actually provide clarity.4 What today’s deci-
sion illustrates is that Specialty Healthcare was more a
loosening of the constraints requiring the Board to act
with transparency and intelligibility than it was a clarifi-
cation of standards, and that it has introduced an ap-
proach to unit determination that permits easy rationali-
zation of any desired result.
Congress established the Board to resolve problems in
a rational manner so that the courts would not have to,
and so that the labor and business community would
have some certainty beforehand as to appropriate bar-
gaining units. But purportedly objective standards that
mask subjective ones whose application is only predicta-
ble by the composition of the agency at a given time look
like a mere screen for obfuscating result-driven jurispru-
dence. And, no agency can earn the trust of the regulated
community or deference of the courts if it does not trans-
parently weigh and apply its established standards. At
times, the courts have had to admonish the Board for
manipulating standards to achieve a desired outcome,
particularly in decisions involving the scope and compo-
sition of bargaining units. In Spentonbush/Red Star Co.
v. NLRB, 106 F.3d 484, 492 (2d Cir. 1997), denying. enf.
319 NLRB 988 (1995), for example, the court stated,
“Recognizing that the NLRB earns and forfeits deferen-
tial judicial review by its performance,” . . . [several cited
cases] “hold in substance that the Board's manipulation

4 Specialty Healthcare, id., slip op at 1 (claiming only to reiterate
and clarify the extant unit determination analysis).
10 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
of the definition of supervisor has reduced the deference
that otherwise would be accorded its holdings.” In an-
other context, writing for the United States Court of Ap-
peals for the District of Columbia Circuit, then-Judge
Roberts remarked that the need for the Board to explain
its analysis “is particularly acute when an agency is ap-
plying a multi-factor test through case-by-case adjudica-
tion.” LeMoyne-Owen College v. NLRB, 357 F.3d 55, 61
(D.C. Cir. 2004) (citations and quotations omitted),
denying enf. 338 NLRB No. 92 (2003). The application
of the multifactor analysis in the context there (whether
faculty were managerial personnel) “can lead to predict-
ability and intelligibility only to the extent the Board
explains, in applying the test to varied fact situations,
which factors are significant and which less so, and why.
. . . In the absence of an explanation, the totality of the
circumstances can become simply a cloak for agency
whim—or worse.” Id.
As for this workplace, we are looking at a very small,
20-employee print shop where all employees work in a
linear, functionally integrated production process. The
petitioned-for unit impermissibly excludes the offset-
press employees in the middle of the production line
from the unit—thereby plucking the heart from the pro-
duction process—despite an overwhelming community
of interest among all employees. As to the manner in
which the majority reaches its result, today’s decision
confirms that under Specialty Healthcare the determina-
tion of whether there is a readily identifiable group has
become an infinitely malleable standard that shows that
anything goes, regardless of whether the “group” tracks
any organizational or other lines drawn by the Employer.
As described below, the decision does not meaningfully
assess community-of-interest standards and provides no
explanation of the elevation of insignificant distinctions,
to the extent they exist, over critical factors such as func-
tional integration, contact, common supervision, similar
wages, and virtually all other factors. Today’s decision
also demonstrates that the shifting of the burden to the
Employer to show an overwhelming community of inter-
est imposes a nearly impossible requirement because the
majority has gone to extraordinary lengths to inflate the
most insignificant of distinctions to defeat the Employ-
er’s showing.
The decision here reads like a doctrinal obstacle course
where the overwhelmingly shared interests connecting
the petitioned-for and excluded employees are factors to
be explained away in a post-hoc justification of that re-
sult, a justification so strained that it is difficult to track
the actual rationale being applied here. Finally, today’s
decision marks a further retreat, beginning with Specialty
Healthcare, from the clear standards that we have suc-
cessively developed in our unit-determination decisions
to something more arbitrary that guarantees whatever
result the panel wants to achieve. As this decision shows
us, the more the Board strains to distort extant standards
into the circumstances of any given case, the more shape-
less those standards become, and the more our standards
regress from coherence to arbitrariness, from objectivity
to the appearance of bias.
I. SPECIALTY HEALTHCARE WILL DESTABILIZE LABOR
RELATIONS BECAUSE IT REMOVES CLARITY FROM OUR
UNIT DETERMINATION ANALYSIS AND ENCOURAGES
ROUTINE APPROVAL OF FRACTURED UNITS
In Specialty Healthcare, the Board majority did away
with the longstanding criteria for determining whether a
petitioned-for unit is appropriate for bargaining and re-
placed it with an open-ended standard that minimizes (or,
for all practical purposes, ignores) the importance of
shared interests between petitioned-for employees and
their excluded coworkers. Under the approach an-
nounced there, “when employees or a labor organization
petition for an election in a unit of employees who are
readily identifiable as a group (based on job classifica-
tions, departments, functions, work locations, skills, or
similar factors), and the Board finds that the employees
in the group share a community of interest after consider-
ing the traditional criteria, the Board will find the peti-
tioned-for unit to be an appropriate unit, despite a con-
tention that employees in the unit could be placed in a
larger unit . . . unless the party so contending demon-
strates that employees in the larger unit share an over-
whelming community of interest with those in the peti-
tioned-for unit.” Id., slip op. at 12–13 (fns. omitted).
Dissenting, former Member Hayes aptly explained that
the overwhelming community-of-interest test would
make the relationship between petitioned-for unit em-
ployees and excluded coworkers irrelevant “in all but the
most exceptional circumstances.” Id., slip op at 15
(Member Hayes, dissenting). He also predicted that the
majority approach would elevate the extent of organizing
to the paramount consideration in determining an appro-
priate unit, contravening the statutory requirement that
the extent of organizing not be given controlling weight.5

5 And as Member Miscimarra pointed out in dissenting in Macy’s,
above, slip op. at 25 (Member Miscimarra, dissenting), the Specialty
Healthcare standard is irreconcilable with the role that Congress in-
tended that the Board would play “in each case” regarding bargaining
unit questions, and Specialty Healthcare renders “controlling” the
“extent to which the employees have organized.” I agree with Member
Miscimarra that “Specialty Healthcare affords too much deference to
the petitioned-for unit in derogation of the mandatory role that Con-
gress requires the Board to play, contrary to Section 9(c)(5) and Sec-
tions 9(a) and 9(b) of the Act.” Id.
DPI SECUPRINT, INC. 11
The accuracy of these insights is clear with today’s deci-
sion.
Some of the changes Specialty Healthcare ushered in
are apparent only in successive decisions. It did not
change the threshold requirement that the petitioned-for
unit must be a readily identifiable group, nor change the
meaning of that phrase. But today’s decision jettisons
that requirement, either by sub silentio change in the def-
inition of the term or by factual error in its claim that the
petitioned-for unit here tracks department lines drawn by
the Employer, which it certainly does not.6 Further, Spe-
cialty Healthcare is somewhat cagey about its abandon-
ment of the historical requirement that the Board, in its
initial analysis, must determine whether petitioned-for
employees share a community of interest among them-
selves as distinct from other employees. Post-Specialty
Healthcare, the Board will merely look at shared inter-
ests of petitioned-for employees without regard for
whether others share identical interests, shifting the bur-
den to the Employer to show that any employees it seeks
to add to the unit have virtually identical interests to
those petitioned for.7 That is a mistake, a foolproof reci-

6 A main culprit here is the reliance in the post-Specialty Healthcare
cases upon job classification as the “lowest common denominator” to
drive the determination of what makes up Specialty Healthcare’s “read-
ily identifiable group.” In some limited circumstances, a job classifica-
tion may simultaneously define an employee’s function as so uniquely
separate and distinct from that of all other employees that the classifica-
tion serves as a proxy for a community of interest. But that tends to be
the exception and not the rule, especially in the modern era of ambigu-
ous (e.g., “Customer Service Specialist”) or multitiered (“Engineer II”)
job titles that entail overlapping work functions with other job titles.
Moreover, daisy chaining a number of distinct job classifications to-
gether, simply because they are distinct job classifications, cannot
logically create an organizational or departmental line in order to define
a legitimate “bargaining unit” any more than aggregating any group of
distinct cells will then result, biologically, in a functioning “organ.” If
the employer itself never recognized such classifications as a separate
department in its day-to-day operations, this should indicate that they
are merely a selective collection of functionally disparate workers and
not an appropriate unit for bargaining. See also fn. 7, infra, and ac-
companying text.
7 Despite this change, the Specialty Healthcare majority claimed to
be relying on extant precedent. In so doing, it relied heavily on deci-
sions requiring that, in order for a unit to be appropriate, the petitioned-
for employees must share community-of-interest factors as distinct
from other employees. See Wheeling Island Gaming, 355 NLRB at 637
fn. 2, which explained that the Board “‘never addresses, solely and in
isolation, the question whether the employees in the unit sought have
interests in common with one another. Numerous groups of employees
fairly can be said to possess employment conditions or interests ‘in
common.’ Our inquiry—though perhaps not articulated in every
case—necessarily proceeds to a further determination whether the
interests of the group sought are sufficiently distinct from those of other
employees to warrant the establishment of a separate unit.’ Newton-
Wellesley Hospital, 250 NLRB 409, 411–412 (1980) (emphasis add-
ed).” And see Specialty Healthcare, slip op. at 13 fn. 32 (affirming that
the decision in Wheeling Island Gaming is “an integral part of our
pe for an inappropriate unit, and, in the majority’s hands,
a nearly impossible burden for the Employer (unless a
union seeks to represent only some employees in a par-
ticular job classification or department). Specialty
Healthcare also purports to retain precedent that strongly
favors certain specific unit configurations in several in-
dustries, yet the approach established there makes no
room whatsoever for such considerations. That comes
into play here because today’s decision fails to give
meaningful weight to what the Board has recognized as
“the traditional lithographic unit” in multiple decisions
that weigh strongly in favor of keeping the two prepress
employees who work each shift in the same unit as the
excluded offset-press employees.8
In contrast with the analysis applied here, the pre-
Specialty Healthcare standards provided reasonably ef-
fective bulwarks against approval of units too narrow in
scope for bargaining. For most industries, we did not
need to ponder a maximum number of units a workplace
could reasonably bear because the standard itself had a
limiting function so long as it was objectively applied.
The requirement that the petitioned-for unit, in itself, be
readily identifiable as a group, coupled with an objective
analysis of the community-of-interest factors shared
among petitioned-for employees as distinct from those
excluded has, for the most part, preserved majority rule
and contributed to the establishment of appropriate bar-
gaining units as Congress intended.9 Also implicit in our
analysis and in the statutory language is a balancing of
the petitioned-for employees’ representation preferences
against the rights of other employees who should argua-
bly be included in a unit and the legitimate need of an
employer to manage a business.
Specialty Healthcare fairly well guarantees the prolif-
eration of fractured units that can only hobble a union-
ized employer’s ability to manage production and to re-

analysis here”). Specialty Healthcare also relied heavily on Blue Man
Vegas, LLC v. NLRB, 529 F.3d 417, 421 (2008), which also stated that
community-of-interest “factors include whether, in distinction from
other employees, the employees in the proposed unit have ‘different
methods of compensation, hours of work, benefits, supervision, training
and skills; if their contact with other employees is infrequent; if their
work functions are not integrated with those of other employees; and if
they have historically been part of a distinct bargaining unit’” (empha-
sis added).
8 E.g., AGI Klearfold, LLC, 350 NLRB 538 (2007).
9 Prior Boards spent much effort developing appropriate standards
for different industries, e.g., eight presumptively appropriate units in
acute health care facilities, Collective-Bargaining Units in the Health
Care Industry, 284 NLRB 1515 (1987–1988); AGI Klearfold, LLC,
above (traditional lithographic unit in printing industry); Charrette
Drafting Supplies Corp., 275 NLRB 1294 (1985) (presumption in
favor of the whole-store unit in the retail context); Colorado Interstate
Gas Co., 202 NLRB 847, 848–849 (1973) (systemwide units are “op-
timal” for public utilities).
12 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
tain a necessary flexibility to respond to industry change.
After Specialty Healthcare, even if all employees, peti-
tioned-for and not, share relevant interests, that won’t
play into the Board’s consideration except after the bur-
den shifts to the Employer to show an “overwhelming
community of interest.” That does not sound like much
at first: simply require the Employer to prove that a peti-
tioned-for unit must include additional employees. But it
is the application that counts, and the Board has set an
insurmountable bar that is apparent here in this case with
the majority’s seizing upon insignificant distinctions to
defeat the Employer’s showing in an arbitrary manner
without considering the relevance of the factors relied on
in relation to the work force as a whole.
The trend toward smaller units—or units comprised of
employees not significantly distinguishable from their
coworkers except by the extent of organizing—cannot
foster labor peace. The proliferation of such units can
only create instability from internal jurisdictional dis-
putes, from the costs and burdens of multiunit bargaining
and the administration of multiple separate contracts (in-
cluding, for example, separate benefit plans), from con-
flicting or irreconcilable demands from separate units,
and from the potential that one unit will disrupt produc-
tion with unique demands that burden all employees.
Moreover, multiple units in a functionally integrated
workplace with a linear production process like this one
erect artificial barriers separating employees and depart-
ments that can only impede an employer’s ability to re-
tain needed flexibility and respond quickly to industry
change. An employer in a small operation such as this,
where employees must be available to assist in other de-
partments, cannot function effectively if various interde-
pendent tasks become fixed in stone within discrete
units—fixed not because of anything inherent in the
work itself, but because a union has only organized some
subset of the employees who, together with nonorganized
employees, share in one linear production process.
Workflow management becomes driven not by efficien-
cies and the demands of the work but by artificial barri-
ers dividing functionally integrated production workers
into separate units so that the simplest of changes may
require negotiation with multiple and sometimes compet-
ing representatives and then the agreement of all of them.
And in an organization where complex decisionmaking
occurs both at the micro, or departmental level and on a
macro, or organization-wide level, a need to bargain effi-
ciencies and needs of each department with separate bar-
gaining representatives can grind an operation to a halt.
Department managers are, of course, part of the greater
organization and have the needs of the organization as a
whole in mind. They must cooperate to swiftly incorpo-
rate changes on a department level in concert with those
in other departments to facilitate change in the workplace
as a whole, in what is often a highly complex decisional
process involving shifting priorities among separate de-
partments. In contrast, multiple bargaining representa-
tives are obviously not part of one organization and do
not have to be responsive to the needs of all or most of
the employees within the organization. Nor do the repre-
sentatives of different departments have to act in concert.
Large-scale organizational changes will depend on
agreement of multiple separate entities who are not part
of the overall organization, may have interests that are at
odds with the interests of other departments and large-
scale flexibility and change, and may have no under-
standing of or interest in cooperating in complex, inte-
grated decisional processes necessary to move an organi-
zation forward. Finally, where, as here, we approve a
unit that excludes a subset of employees who are part of
the same production process as those in the unit and who
share virtually all employment terms subject to mandato-
ry bargaining, the excluded employees will be subject to
negotiated terms involving the production process and,
for all practical purposes, the same employment terms
for the unit employees, except in the unlikely event that
the employer sets up entirely separate benefits plans,
work rules, and other terms for unit and nonunit employ-
ees where none had existed in the past. That subverts the
fundamental requirement of majority rule in the represen-
tation decision.
Thus, in an objective assessment of our community-of-
interest factors—and I say our because it is a statutory
responsibility of the Board to make this assessment—we
should consider the scope of a petitioned-for unit with
respect to the potential number of units that could be
spawned by assumptions used by the initial unit determi-
nation, and be mindful of the number of units that any
given workplace could reasonably bear. Here, for exam-
ple, the rationale my colleagues apply in approving a unit
that excludes the offset-press employees also suggests
that appropriate units could be found in any configura-
tion of departments or a single department (except for the
lone digital-press employee—and who knows where that
individual would go, if not the offset-press depart-
ment?).10 The offset-press employees, by their exclusion
here, must be found appropriate; and the prepress, bind-

10 Of course my colleagues are not approving four separate units
now, but the salient points are their lack of awareness of the conse-
quences of approving this fractured unit and the possibility that if the
Union does not succeed this time, it can try and try again with increas-
ingly piecemeal, fractured configurations based on the groundwork laid
here. Using job classification as the building block for a unit often
guarantees this result. See fn. 6, above.
DPI SECUPRINT, INC. 13
ery, and shipping and receiving employees would also be
found appropriate under today’s rationale if petitioned
for separately. Each can be defined as a group because it
comprises employees in specific job classifications or in
the same department. Each has its own task within the
linear production and fulfillment process, and none is
any more distinguishable from the others than are the
excluded offset employees (although the shipping and
receiving employees are more removed from the produc-
tion employees than are the excluded press employees).
Each shares a community of interest among themselves
as well as almost identical interests with all workers in
the shop. And each can be distinguished from the others
by insignificant factors like the ones my colleagues rely
on here, as discussed below. Yet it would be absurd to
require a business with 20 employees working in one
production process to contend with four separate bar-
gaining representatives. This is a 20-employee print
shop, not a shipbuilding operation. And we would fail in
our obligation to the regulated community if we waited
to think about these impacts until after an employer is
locked in to an unsustainable obligation to bargain with
multiple, inappropriate units.
II. THE PETITIONED-FOR UNIT IS TOO NARROW IN SCOPE
FOR BARGAINING AND MUST INCLUDE THE OFFSET-PRESS
EMPLOYEES
1. The Employer’s linear production process
The 20 hourly employees work a linear, functionally inte-
grated production process where each task depends on the
completion of the preceding one. (See Figure 1, “The Em-
ployers’ Production Process,” with the process superim-
posed on the layout of the plant.) For both digital and offset
printing, the process begins with the two prepress employ-
ees (one per shift), where images and text supplied by the
customer are used to create proofs. Once the customer ap-
proves the proofs, prepress employees either create plates
and place them in a rack in the offset-press department,
comprising four press operators and three feeder-tenders
who assist them, or transmit a digital file to the one digital-
press employee. For offset jobs, the offset-press employees
use the plates developed by the prepress employees, print
text and images on blank paper using one of two types of
presses, and prepare printed sheets for finishing by offset-
bindery employees who work across the room from them.
There four offset-bindery employees cut, fold, or stitch the
printed paper, depending on the job. The two digital-
bindery employees perform much the same tasks. Once the
production process is complete, the finished product goes to
the shipping and receiving department, where there are three
employees who shrink wrap and palletize completed jobs,
fill out packing slips, and load finished jobs onto delivery
trucks.
The facility is arranged to accommodate workflow.
The prepress department is located at one end of the
building, with doors connecting it to the digital room and
the offset-press room. A door also connects the digital
room and the offset-press room. The offset press and
offset bindery occupy the same space. A traffic aisle
runs from the doors connecting the offset-press room to
the prepress and digital rooms, along the side of the off-
set-press area, through the offset-bindery area, and to the
opposite side of the offset-press and bindery room, ter-
minating at a door that connects the shipping and receiv-
ing room to the offset-bindery area.
2. The petitioned-for employees are not a readily
identifiable group
The threshold question in determining the appropriate-
ness of a bargaining unit is whether the petitioned-for
employees are readily identifiable as a group (based on
job classifications, departments, functions, work loca-
tions, skills, or similar factors).11 Today’s decision guts
the term of meaning. It declares that the petitioned-for
employees are “readily identifiable as a group” because
they are “all the hourly employees in the pre-press, digi-
tal press, bindery, and shipping and receiving depart-
ments—in short, all the hourly employees who do not
work on the offset presses.” It then declares that the peti-
tioned-for employees “are readily identifiable as a group
based on departments and functions,” and cites Bergdorf
Goodman, 361 NLRB No. 11, slip op. at 2 (2014), for the
proposition that the petitioned-for unit conforms “to the
departmental lines established by the employer.” Wrong.
First, as a group, the proposed unit conforms to nothing
except for the Petitioner’s organizing efforts. The Em-
ployer has drawn no departmental lines whatsoever
around this collection of employees. Rather, the pro-
posed unit is an amalgamation of all but one department.
It is the production employees minus one department
with the shipping and receiving employees lumped into
the mix. Yes, the unit can be described in a sentence, as
the majority has done (conceived by job classification,
the unit is a + c + d + e, but not b). That is not an identi-
fiable group. Not even close. It is a list of four depart-
ments gathered by the petitioner. Further, the majority’s
claim that we have a coherent group directly conflicts
with the Board’s post-Specialty Healthcare decision in
Odwalla, Inc., 357 NLRB No. 132 (2011).

11 Specialty Healthcare, above, slip op. at 12.
14 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
DPI SECUPRINT, INC. 15
There, like here, the petitioned-for unit included employees
in several job classifications and in different departments.
Recognizing an inappropriate unit when it saw one, the
Odwalla Board explained that the unit did not track any
lines drawn by the Employer: The unit was not drawn along
classification grounds because (like here) it “aggregates
various classifications,” nor along departmental lines be-
cause (like here) it comprises employees in various different
departments.12
Nor are the petitioned-for employees here readily iden-
tifiable based on any relatedness of job classification, as
the unit includes several disparate classifications (ship-
ping and receiving employees, offset-bindery employees,
digital bindery employees, digital-press employees, and
prepress employees). The departments that constitute the
petitioned-for unit also have the same supervision as the
excluded offset-press department. The proposed unit is
not readily identifiable based on functions or skills ei-
ther: The unit is most production employees (but pluck-
ing out those at the heart of production), all of whom
have varying functions and skills. Prepress employees
perform entirely different functions (creating digital
proofs, formatting and creating plates) than bindery em-
ployees (cutting, folding, and stitching printed sheets),
digital-press employees (running the digital press), and
the shipping and receiving employees (shrink wrapping
and palletizing finished jobs, filling out packing slips,
and loading trucks). The latter do not participate in the
actual production of the Employer’s products. There are
clearly different skills involved in these different func-

12 Compare what the Board has found to be a readily identifiable
group in other post-Specialty Healthcare cases. In Specialty
Healthcare itself the petitioned-for employees were such a group be-
cause they comprised a single, entire job classification: all the employ-
er’s CNAs. Id., slip op. at 12. In Bergdorf Goodman, above, slip op. at
2, the petitioned-for employees were at least readily identifiable as a
group along functional lines because they comprised all women’s shoes
salespeople. See also Northrop Grumman Shipbuilding, Inc., 357
NLRB No. 163, slip op. at 3 (2011) (all technical employees in radia-
tion control department, comprising the employees uniquely tasked
with protecting against radiation contamination at shipyard). In those
cases, there was an effort to conform to the requirement that the units
were identifiable groups. Contrary to my colleagues, neither I (nor
those cases) suggest that “readily identifiable as a group” is a version of
the community-of-interest analysis. And, in fact, my analysis flows
directly from, inter alia, Specialty Healthcare (see my fn. 11, above).
But the salient point is that there is no coherent or objectively recog-
nizable factor tying the petitioned-for employees together as an identi-
fiable group except that the Petitioner seeks to represent them. Nor is it
sufficient that the petitioned-for unit merely “specify” a group of em-
ployees the petitioner seeks to include. The ability to “specify” a pro-
posed unit by simply compiling various job classifications into a sen-
tence does not give us an actual, substantive group, as my colleagues
suggest. An agglomeration of disparate job classifications cannot logi-
cally constitute a “readily identifiable group,” just because job classifi-
cations are the basic unit of addition.
tions, so the petitioned-for employees are also not readily
identifiable based on skills.13 They are not readily identi-
fiable based on their work location, because all employ-
ees work in the same facility in close proximity. The
prepress employees work in a separate room, as do the
digital press and digital-bindery employees. Offset-
bindery employees work in the same room as the exclud-
ed offset-press employees, and shipping and receiving
employees also have a separate room.14 Although Spe-
cialty Healthcare indicates that “similar factors” can be
used to establish that a petitioned-for unit is readily iden-
tifiable as a group, the majority points to no unifying
characteristic that distinguishes them from the offset-
press employees, except for the fact that they are not
offset-press employees, and (the actual rationale here)
the fact that they comprise the employees that the Peti-
tioner seeks to represent.
3. The petitioned-for employees do not share a communi-
ty of interest distinct from the excluded press employees,
and the excluded employees have an overwhelming
community of interest with those in the
petitioned-for unit
The petitioned-for and excluded employees share al-
most identical community-of-interest factors. That is
evident even in my colleagues’ own recitation of the
facts. As my colleagues recognize, petitioned-for em-
ployees and the excluded offset-press operators and feed-
er-tenders are functionally integrated in the Employer’s
linear production process and work in the same space
with dividing walls laid out to facilitate workflow. The
offset-press employees are the heart of production and
without them there is no direct “functional integration”
connecting prepress and offset-bindery employees. The
excluded offset-print employees share space with peti-
tioned-for employees (the offset press department is in
the same room as the bindery department). All employ-
ees’ pay ranges from $10-$20 per hour. They have the
same health benefits, holiday pay, 401(k) plan, and are
subject to the same general policies and operating manu-
al. All use the same entrance, timeclock, and lunchroom.
They have common supervision, depending on the shift.

13 Nor on the level of skill or the amount of training required for em-
ployees in the various departments: It takes “months” to learn to oper-
ate bindery equipment, for example, similar to the offset-press opera-
tors.
14 Nor are the petitioned-for employees readily identifiable based on
their hours, because the offset bindery and shipping and receiving only
operate on one shift, whereas the other departments operate on two
shifts. The petitioned-for employees also are not readily identifiable
based on their wages, because the range of prepress wages ($15-
$20/hour) is notably higher than in shipping and receiving ($10-
$17/hour), while excluded press employees are paid $16/hour and
$20/hour, which is in line with petitioned-for employees.
16 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
All have considerable contact with each other both for-
mally and informally. Petitioned-for and excluded em-
ployees work varying hours and shifts. At all times
when petitioned-for employees are working, offset-press
employees are working too, although not all petitioned-
for employees have overlapping hours with one anoth-
er.15
While the offset-press employees work in a separate
department from the petitioned-for employees, the peti-
tioned-for employees are themselves divided among four
departments and multiple job classifications.
Contrary to my colleagues’ characterizations, other
factors are significantly in common as well. My col-
leagues contend that the offset-press employees require
“greater skill and lengthier training” than the petitioned-
for employees. That’s not accurate. The excluded off-
set-press employees are two classifications: Four press
operators, who may require up to 6 months on-the-job
training, and the three feeder-tenders who assist them by
feeding paper into the presses. The lengthy training ap-
plies to the four employees running the presses, but is not
necessary for the feeder-tenders, who require little train-
ing and are often hired with no prior experience.16
Moreover, while there was testimony that offset-press
operators may require 6 months training, there is no evi-
dence of how long it takes to train the prepress and ship-
ping employees, and the training of bindery employees
also takes “months.” There is no basis to extrapolate a
finding from that testimony that the four offset-press
employees require significantly more training than the
bindery employees except for the unconvincing distinc-
tion between “6 months” and “months.” None of the
employees, in any case, require special certification of
any kind to do their jobs or can be considered craft em-
ployees. That the majority ignores the feeder-tenders
from their analysis and then finds a dispositive line be-
tween “months” and “6 months” to differentiate press
operators from bindery employees illustrates my earlier
point that the majority rationale is so strained that it does
not appear to be the product of an objective standard at
all. It looks like the sort of post-hoc justification that is
antithetical to our obligation to develop and apply mean-
ingfully articulated standards that guide us to predictable
results. In sum, there is no meaningful distinction be-
tween petitioned-for and excluded employees on the ba-
sis of skills and training.

15 Despite such commonality, my colleagues grudgingly characterize
the offset-press employees as sharing only “some” community-of-
interest factors with the petitioned-for employees.
16 Even if some may learn how to use presses by virtue of working in
the department, they cannot be considered highly skilled employees as
compared to those in the petitioned-for unit.
The majority also finds that the excluded press em-
ployees work different hours from the petitioned-for em-
ployees and, implicitly, that the distinctions they find are
meaningful for collective-bargaining purposes. First, the
majority points to evidence that, unlike other employees,
offset-press operators have not been sent home when the
work is slow. But the record also shows that feeder-
tenders have been among those asked to go home, like
petitioned-for employees, so there’s no meaningful dis-
tinction between petitioned-for and excluded employees
there. (And none of this is to suggest that being asked to
go home on an ad hoc basis some unknown number of
times should have any role in this analysis). Second, the
majority contends that offset-press employees may be
scheduled on weekends, unlike petitioned-for employees,
and work 10-hour shifts. At the same time, however, the
petitioned-for employees themselves work multiple, var-
ied shifts. All have significantly overlapping hours with
offset-press employees even though they may have no
overlapping hours with each other. Thus, shipping and
receiving employees work 7 a.m. to 3:30 p.m. The first-
shift prepress employee begins work at 8:30 a.m. There
is evidence that this employee works until 5 p.m., but the
Acting Regional Director said that neither this nor the
second-shift hours were in the record. Digital bindery
employees work first and second shifts, but precise times
are not in the record. Offset-bindery employees work
from 7 a.m. to 3:30 p.m. One works from 5 a.m. until 1
p.m. Monday through Thursday, first-shift press em-
ployees work from 5 a.m. to 3 p.m.; second shift from 3
p.m. to 1 a.m. The Acting Regional Director stated that
it was not clear from the record exactly when the second-
shift prepress, digital-press, and digital-bindery employ-
ees start and finish, but that “it can reasonably be inferred
that their hours are similar.” The extent to which their
hours correspond to those of other employees (both in
and excluded from the unit) is not entirely clear. There is
always at least one petitioned-for employee working be-
tween 5 a.m. and midnight during the week, while, as
noted, offset-press employees work in two shifts span-
ning 5 a.m. to 1 a.m. One bindery employee begins his
day with the offset-press employees at 5 a.m. Thus,
some petitioned-for employees have no contact with oth-
ers during working time (there is a second shift for pre-
press and digital-bindery employees but not for offset-
bindery employees, for example), but all have signifi-
cantly overlapping shifts with offset-press employees.
At all times during all petitioned-for employees’ shifts,
offset-press employees are also scheduled to work. So a
rational justification for finding these distinctions signifi-
cant enough for a unit-determination decision escapes
me. The shift distinctions for the offset-press employees
DPI SECUPRINT, INC. 17
are overstated and far too insignificant to justify exclu-
sion from the unit given the varied shifts among all peti-
tioned-for employees. See Moore Business Forms, 216
NLRB 833, 834 (1975) (press and prepress employees
formed appropriate unit notwithstanding that, unlike
press employees, prepress employees did not work
graveyard shift).
Moreover, if the Board is going to rely on these minor
distinctions, it should explain how they are meaningful
for bargaining purposes. Assuming that one purpose for
looking at working hours is to determine whether em-
ployees have contact with each other at work, several
petitioned-for employees have no contact with each other
while all have contact with the offset-press employees
during their shifts, and my colleagues in any event rec-
ognize that there is significant contact among petitioned-
for and offset-press employees. If the presence of small
scheduling variations is relevant to looking at the extent
of functional integration, that is also not implicated here,
where extensive functional integration among all produc-
tion employees is beyond question. Nor are the distinc-
tions related to common supervision. If minor schedul-
ing variations are deemed relevant just because they are a
distinction that the majority can seize upon in the ab-
sence of actually relevant ones to find that the Employee
has not carried its overwhelming community-of-interest
burden, that underscores the arbitrariness of their as-
sessment of shared interests.
The majority also relies on a lack of interchange in
finding that the offset-press employees lack an over-
whelming community of interest with the petitioned-for
employees. There are particular problems in relying on
lack of interchange to defeat an employer’s overwhelm-
ing community-of-interest showing, which I describe
below. But first, it is certainly true that the employees
are in several job classifications and primarily perform
the duties of their particular classification. But a condi-
tion of employment common to all employees here is
that they are expected to—and do—cross into other de-
partments to assist their coworkers. The offset employ-
ees, like the petitioned-for employees, are expected to
help in other departments as needed, and do so on an ad
hoc basis, such as when an offset-press operator runs the
digital press, or—more commonly—a feeder-tender as-
sists offset-bindery employees, which was established by
both the Petitioner’s and Employer’s witnesses. Offset-
press employees have operated bindery equipment, par-
ticularly the jogger or cutter, when bindery employees
are not available to operate these machines. Moreover, it
is because employees generally perform their assigned
work based on their job description that my colleagues
dismiss evidence of interchange as incidental to their
regular work, and work the offset-press employees do in
other departments is dismissed as “one way.” As to the
latter point, the salient fact here is that even if employees
in other departments do not operate presses, the excluded
press employees may be called on to perform unit work.
That underscores the Employer’s need to maintain a flex-
ible workplace and cuts in favor of placing the offset
employees in the petitioned-for unit.
But more broadly, the manner in which my colleagues
use the interchange factor here confirms that an Employ-
er will rarely be able to establish an overwhelming com-
munity of interest when seeking to include additional job
classifications in a proposed unit. To be sure, a finding
of interchange among employees, where various tasks
can be assigned to any number of employees in a peti-
tioned-for unit, is very compelling evidence in finding a
shared community of interest (because employees are
doing the same work as each other). But while its pres-
ence is important, the lack of it is not. That is because
the lack of it tells us only that employees are in different
job classifications and usually do work assigned to their
particular jobs. To the extent that such employees per-
form work outside of their typical assignment or job
classification, that will of course be “incidental” to their
normal duties. Minimal interchange among employees
shows us that the Employer has an organized production
process where employees by and large keep to their job
responsibilities. If the lack of interchange is enough to
show the lack of an overwhelming community of inter-
est, that is tantamount to elevating the fact of separate
job classifications to a dispositive level, which is itself a
grave error, as I have addressed above. And most situa-
tions in which an employer does seek to include more
employees in the unit will involve an effort to include
additional job classifications. Which underscores the
point that an employer will almost never be able to show
an overwhelming community of interest when it seeks to
include additional job classifications in a petitioned-for
unit (unless job descriptions are so loose and permeable
that employees are regularly tasked with work outside of
their nominal classification).17
An objective view of the shared interests indicates that
approving the petitioned-for unit would not reflect ma-
jority rule. The production process is so tightly integrat-
ed that any changes to it requiring bargaining would im-

17 As former Member Hayes put it, so long as a union “does not
make the mistake of petitioning for a unit that consists of only part of a
group of employees in a particular classification, department, or func-
tion . . . it will be impossible for a party to prove that an overwhelming
community of interests exists with excluded employees.” DTG Opera-
tions, Inc., 357 NLRB No. 175, slip op. at 8–9 (2011) (Member Hayes,
dissenting).
18 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD
pact the offset-press employees who would be left out of
the bargaining process, and with the shared wages, bene-
fits, work rules, and virtually all other terms and condi-
tions of employment that are mandatory bargaining sub-
jects, the excluded press employees would be subject to
bargained-for changes as well unless the Employer estab-
lishes two regimes including entirely separate health
plans, 401(k) plans, etc. That is impractical and unrealis-
tic. In this way, the impact of an inappropriate unit is to
make the nonunit employees’ employment terms subject
to whatever is negotiated for the represented employees,
contravening our statutory mandate. This brings me
back to my earlier point that in the initial unit determina-
tion must rationally consider the overall impact on the
workplace—including the impact on excluded employees
who share almost all community-of-interest factors with
the unit.18
4. Today’s decision also departs from precedent general-
ly finding units of press and pre-press employees—the
“traditional lithographic unit”—appropriate
or bargaining
In Specialty Healthcare, the Board recognized that it
“has developed various presumptions and special indus-
try and occupation rules in the course of adjudication”
and made clear that the holding in Specialty Healthcare
was “not intended to disturb any rules applicable only in
specific industries.”
Here, such precedent establishes that press and pre-
press employees, comprising the “traditional lithographic

18 My colleagues reliance on AT Wall Co., 361 NLRB No. 62
(2014), is misplaced both with respect to facts and the applicable ana-
lytical framework. That was an accretion case and involved employees
working in separate production lines. The employer there acquired a
gun magazine manufacturing operation and moved it to an existing
facility where it already operated separate tubing and stamping manu-
facturing production lines. The question was whether the newly hired
gun magazine production employees (Metalform employees) should be
accreted into the existing and narrowly defined unit of employees
working on the other two production lines. Contrary to my colleagues’
suggestion, Metalform employees were engaged in an entirely different
production process from the employees in the existing unit. In the
current case, of course, the petitioned-for and excluded employees are
all engaged in the same linear production process. Further, in AT Wall,
the employees sought to be accreted had different wages, hours, super-
vision, vacation and holidays, and different medical insurance benefits
from unit employees. In sum, they produced “an entirely different
product using different processes under different working conditions.”
Id., slip op. at 3. The similarity, irrelevant in context, is that there too
all employees worked in the same open facility. Further, that was an
accretion case, governed by entirely different policy considerations
than the case at bar, as the Board follows a “restrictive policy in finding
accretions to existing units” to ensure that the right of employees to
determine their own bargaining representative is not foreclosed (Id.,
slip op. at 3). That the Board found that the Metalform employees
should not be accreted, and thus should retain the right to select their
own bargaining representative, is utterly irrelevant to today’s decision.
unit,” should ordinarily be included in the same unit. See
AGI Klearfold, LLC, above;19 Moore Business Forms,
above, decisions holding that a press-only unit was inap-
propriate because it excluded prepress employees.
The majority’s effort to distinguish these cases is un-
convincing. My colleagues are correct that in those cas-
es, press employees regularly entered the prepress room
for various reasons. See AGI Klearfold, above at 539;
Moore Business Forms, above at 834. But the same is
true here, as the offset-press employees seek out prepress
employees on a daily basis to discuss plate issues, and
there is no dispute that there is frequent contact between
all employees. The employees are engaged in a fully
integrated, linear production process beginning with the
one prepress employee on each shift, and the offset print-
ers’ work wholly depends on the one prepress employee
on that shift. The processes in place here, and the inter-
dependence of the prepress and press employees, are
substantially identical to those in AGI Klearfold and
Moore, and the prepress and offset-press employees have
the same roles and interdependence as in those cases. As
far as I can tell, my colleagues’ strained effort to distin-
guish those cases appear to come down to an asserted
difference in the amount of time that employees in the
one department come into the other department. Some-
how my colleagues have found a line demarking the suf-
ficient interaction in those cases from an assertedly insuf-
ficient interaction in this one. But contact and functional
integration have already been firmly established here.
And for all practical purposes these are the very factors
my colleagues rely on to try to distinguish Klearfold and
Moore. Thus, I do not see that the majority has given
any weight whatsoever to the “traditional lithographic
unit” nor that it can meaningfully distinguish AGI
Klearfold, and Moore Business Forms from the current
case. Longstanding precedent strongly favors including
the offset-press employees in a unit with the prepress
employees, which in this case favors including the offset-
press employees in the petitioned-for unit.
III. CONCLUSION
The petitioned-for unit is inappropriate under both our
traditional community-of-interest analysis and the stand-
ard announced in Specialty Healthcare because the off-

19 Noting that the Board has “ordinarily found a unit limited to press
and prepress employees appropriate, eventually referring to such a
grouping as the ‘traditional lithographic unit.’” Id. at 540 (citing Allen,
Lane & Scott, 137 NLRB 223, 226 (1962); Earl Litho Printing Co., 116
NLRB 1538, 1539 (1956); Shumate, Inc.,131 NLRB 98, 99 (1961);
A.B. Hirschfeld Press, Inc., 140 NLRB 212, 216 (1962); George Rice
& Sons, 212 NLRB 947, 947–948 (1974); Moore Business Forms, Inc.,
216 NLRB 833, 834 (1975); and Meyer Label Co., 232 NLRB 933, 934
(1977)).
DPI SECUPRINT, INC. 19
set-press employees share an overwhelming community
of interest with the employees the Union seeks to organ-
ize. My colleagues approve the unit based on insignifi-
cant distinctions that do not provide a rational basis for a
unit’s boundaries. As the Specialty Healthcare majority
itself said, “ . . . no two employees’ terms and conditions
of employment are identical, yet some distinctions are
too slight or too insignificant to provide a rational basis
for a unit’s boundaries.” Id. slip op. at 13. Today’s deci-
sion also omits an assessment of the relative importance
of the various community-of-interest factors relied upon,
particularly in light of those overwhelmingly shared. As
the Board explained in American Cyanamid Co., 131
NLRB 909, 911 (1961):
To be effective . . . each unit determination must have a
direct relevancy to the circumstances within which col-
lective bargaining is to take place. While many factors
may be common to most situations, in an evolving in-
dustrial complex the effect of any one factor, and there-
fore the weight to be given it in making the unit deter-
mination, will vary from industry to industry and from
plant to plant. We are therefore convinced that collec-
tive-bargaining units must be based upon all the rele-
vant evidence in each individual case.
And it bears repeating that the application of a multi-
factor test can “lead to predictability and intelligibility
only to the extent the Board explains, in applying the test
to varied fact situations, which factors are significant and
which less so, and why.” LeMoyne-Owen College v.
NLRB, 357 F.3d at 61. Here, my colleagues inflate the
significance of slight scheduling variations and minimal
interchange and transfer (i.e., the fact that employees
stick to their assigned jobs) to overcome the critical im-
portance of functional integration in a linear production
process, frequent contact, common supervision, and oth-
er accumulated and overwhelming shared interests be-
tween the petitioned-for and excluded employees. The
decision illustrates the unpredictability of the post-
Specialty Healthcare landscape, with the standard the
majority actually applies difficult to track despite their
claimed adherence to whatever remains of our communi-
ty-of-interest analysis.
Finally, today’s decision illustrates an agency’s re-
sistance to clear and intelligible standards that constrain
its ability to engage in a result-driven and thus arbitrary
decisional process when the standards lead to a result
contrary to the panel’s desired outcome. Our obligation
is to adhere to clarity, to apply the standard we claim to
be applying, and to avoid the impulse to manipulate a
desired result by distorting the standards that the regulat-
ed community is entitled to rely on. This decision fails
to fulfill that obligation, and I respectfully dissent.
Dated, Washington, D.C. August 20, 2015
_____________________________________
Harry I. Johnson, III, Member
NATIONAL LABOR RELATIONS BOARD