Labor Union Certification Procedures: Use of Secret Ballots and Card Checks







Prepared for Members and Committees of Congress



The National Labor Relations Act of 1935 (NLRA) gives private sector workers the right to join
or form a labor union and to bargain collectively over wages, hours, and other working
conditions. An issue before Congress is whether to change the procedures under which a union is
certified as the bargaining representative of a union chosen by a majority of workers.
Under current law, the National Labor Relations Board (NLRB) conducts a secret ballot election
when a petition is filed requesting one. A petition can be filed by a union, worker, or employer.
Workers or a union may request an election if at least 30% of workers have signed a petition or
authorization cards (i.e., cards authorizing a union to represent them). The NLRA does not require
secret ballot elections. An employer may voluntarily recognize a union if a majority of workers
have signed authorization cards.
Legislation was introduced in the 110th Congress that, if enacted, would have changed current
union certification procedures. The Employee Free Choice Act of 2007, H.R. 800 and S. 1041,
would require the NLRB to certify a union if a majority of employees sign authorization cards.
The Secret Ballot Protection Act, H.R. 866, would require secret ballot elections for union
certification.
Proponents of both measures sometimes use similar language to support their positions.
Employers argue that, under card check certification, workers may be pressured or coerced into
signing authorization cards and may only hear the union’s point of view. Unions argue that,
during an election campaign, employers may pressure or coerce workers into voting against a
union. Supporters of secret ballot elections argue that casting a secret ballot is private and
confidential. Unions argue that, during an election campaign, employers have greater access to
workers. Unions argue that card check certification is less costly than a secret ballot election.
Employers maintain that unionization may be more costly to workers, because union members
must pay dues and higher union wages may result in fewer union jobs.
Requiring card check certification may increase the level of unionization, while requiring secret
ballot elections may decrease it. Research suggests that, where card check recognition is required,
unions undertake more union drives and the union success rate is higher. The union success rate is
also greater where card check recognition is combined with a neutrality agreement (i.e., an
agreement where the employer agrees to remain neutral during a union organizing campaign).
To the extent that requiring secret ballot elections or requiring certification when a majority of
employees sign authorization cards would affect the level of unionization, the economic effects
may depend on how well labor markets fit the model of perfect competition. Requiring card
check certification may improve worker benefits and reduce earnings inequality—if more
workers are unionized. Requiring secret ballot elections may increase inequality in
compensation—if fewer workers are unionized. This report will be updated as issues warrant.






Legislation in the 110th Congress....................................................................................................1
The National Labor Relations Act...................................................................................................3
National Labor Relations Board (NLRB).................................................................................4
Organizing Campaign Rules.....................................................................................................4
Employers ...................................................................................................................... ..... 4
Employees ...................................................................................................................... ..... 5
Union Organizers................................................................................................................5
Unfair Labor Practices..............................................................................................................5
Employers ...................................................................................................................... ..... 6
Unions ................................................................................................................................. 6
Reme dies ....................................................................................................................... ...... 6
Union Certification and Recognition...............................................................................................8
Secret Ballot Elections..............................................................................................................8
Number of NLRB Elections..............................................................................................10
First Contract Agreements Following Certification.................................................................11
Voluntary Card Check Recognition.........................................................................................12
Neutrality Agreements......................................................................................................13
Corporate Campaigns........................................................................................................13
Number of Voluntary Recognitions..................................................................................14
Bargaining Orders...................................................................................................................14
Certification Versus Recognition............................................................................................15
Withdrawal of Recognition.....................................................................................................16
Joy Silk Doctrine.....................................................................................................................16
NLRB Review of Voluntary Recognition...............................................................................16
The United Auto Workers (UAW) and the Dana and Metaldyne Corporations................17
Shaw’s Supermarkets and the United Food and Commercial Workers (UFCW)
Union ............................................................................................................................. 17
NLRB Review of Withdrawal of Recognition........................................................................18
Potential Effects of Changes in Union Certification Procedures...................................................19
Research Findings...................................................................................................................19
Is There an Economic Rationale for Protecting the Rights of Workers to Organize and
Bargain Collectively?.................................................................................................................23
Government Intervention in Labor Markets............................................................................23
Distribution of Earnings..........................................................................................................24
Collective Voice......................................................................................................................24
Conclusion .............................................................................................................................. 25
Figure 1. Unfair Labor Practice Charges, Fiscal Years 1970-2007................................................8





Table 1. Number of Representation Elections Conducted by the NLRB, FY1994-FY2007........10
Table 2. First Contract Agreements Following Certification..........................................................11
Table 3. Number of Voluntary Recognitions in Which the Federal Mediation and
Conciliation Service (FMCS) Provided Assistance for Initial Contracts, FY1996-
FY2004 ....................................................................................................................................... 14
Table 4. Common Arguments Made by Proponents of Requiring Card Check Certification
and Requiring Secret Ballots......................................................................................................20
Author Contact Information..........................................................................................................25





he National Labor Relations Act of 1935 (NLRA), as amended, gives private sector
workers the right to join or form a labor union and to bargain collectively over wages, 1
hours, and other conditions of employment. The act also requires employers to bargain in T


good faith with a union chosen by a majority of employees. An issue before Congress is whether
to change the procedures under which a union is certified as the bargaining representative of a
union chosen by a majority of workers.
Legislation was introduced in the 110th Congress that, if enacted, would have changed existing th
union certification procedures. Similar legislation is expected to be introduced in the 111 th
Congress. This report summarizes legislation introduced in the 110 Congress. The report then
reviews the rights and responsibilities of workers and employers under the NLRA and the
different ways that workers may form or join a union. Next, the report analyzes the potential
effects of changes in union certification procedures. Finally, the report considers whether there is
an economic rationale for protecting the rights of workers to organize and bargain collectively.

H.R. 800, the Employee Free Choice Act of 2007, would require the National Labor Relations
Board (NLRB) to certify a union if a majority of employees in a bargaining unit sign 2
authorization cards designating the union as their bargaining representative. If less than a
majority, but at least 30%, of employees sign authorization cards, a secret ballot election may be
held. (See the section on “Secret Ballot Elections” later in this report.)
H.R. 800 would also establish procedures for reaching a first contract agreement. Within 10 days
(or a longer period if agreed to by both the union and employer) after receiving a request to begin
bargaining, the employer and union must meet to begin negotiations over an initial contract
agreement. If the employer and union cannot reach an agreement within 90 days after bargaining
has begun (or a longer period if agreed to by both parties), either party may request mediation by
the Federal Mediation and Conciliation Service (FMCS). If an agreement cannot be reached
within 30 days through mediation (or a longer period if agreed to by both parties), the dispute
would be subject to binding arbitration. The legislation would increase the penalties for employer
violations of certain unfair labor practices committed during a union organizing campaign or
during negotiation of a first contract. H.R. 800 was introduced by Representative George Miller 3
on February 5, 2007, and referred to the House Committee on Education and Labor. A hearing on

1 The NLRA is also known as the Wagner Act, after Senator Robert Wagner of New York who sponsored the bill in the
U.S. Senate. Representative William Connery of Massachusetts sponsored the bill in the U.S. House of
Representatives.
2 This section uses terms and refers to agencies—e.g., certification, recognition, bargaining unit, unfair labor practices,
National Labor Relations Board (NLRB), and Federal Mediation and Conciliation Service (FMCS)—that are described
later in this report.
An authorization card may serve more than one purpose. A single-purpose card authorizes the union to represent the
employee signing the card. A dual-purpose card requests an election and designates the union as the employees’
bargaining representative. The NLRB may not issue a bargaining order (see “Bargaining Orders” later in this report) if
a union uses dual-purpose cards. Dual-purpose cards could indicate that a majority of employees want an election, but
not that a majority of employees want to be represented by a union. Commerce Clearing House, Labor Law Reporter: rd
Labor Relations (Chicago: CCH Inc., 2007), ¶ 3042. Bruce S. Feldacker, Labor Guide to Labor Law, 3 ed.
(Englewood Cliffs, N.J.: Prentice Hall, 1990), p. 72. (Hereinafter cited as Feldacker, Labor Guide to Labor Law.)
3 Representative George Miller, “Introduction of H.R. 800, The Employee Free Choice Act,” Congressional Record,
vol. 153, February 5, 2007, p. E260.



the legislation was held on February 8, 2007, by the House Subcommittee on Health, 4
Employment, Labor, and Pensions. The Committee on Education and Labor approved the bill on 5
February 14, 2007, by a vote of 26 to 19. The full House approved the measure on March 1,

2007, by a vote of 241 to 185.


S. 1041, the Employee Free Choice Act of 2007, was introduced in the Senate on March 29, 2007
by Senator Edward Kennedy. The bill was referred to the Committee on Health, Education,
Labor, and Pensions (HELP). Hearings on the legislation were held in the Senate on March 27, 6
2007. On June 26, 2007, the Senate voted to consider H.R. 800. By a vote of 51 in favor and 48
opposed, the Senate fell 9 votes short of the 60 needed to end debate and proceed to a vote on the
legislation.
H.R. 866, the Secret Ballot Protection Act, would require a secret ballot election for union
certification. The bill would make it an unfair labor practice for an employer to recognize or
bargain with a union that has not been selected by a majority of employees in a secret ballot
election conducted by the NLRB. It would also be an unfair labor practice for a union to cause or
attempt to cause an employer to recognize or bargain with a union that has not been chosen by a
majority of employees in a secret ballot election. H.R. 866 was introduced by the late
Representative Charlie Norwood on February 7, 2007 and was referred to the House Committee 7
on Education and Labor. Senator Jim DeMint introduced S. 1312, the Secret Ballot Protection 8
Act of 2007 in the Senate on May 7, 2007.
On March 1, 2007, during floor debate on H.R. 800, Representative Howard “Buck” McKeon
offered an amendment that would have substituted H.R. 866 for H.R. 800. The amendment failed
by a vote of 173 to 256.

4 U.S. Congress, House Committee on Education and Labor, Subcommittee on Health, Employment, Labor and
Pensions, Strengthening Americas Middle Class Through the Employee Free Choice Act, February 8, 2007, available
at http://edlabor.house.gov/hearings/help020807.shtml.
5 For more information on the “Employee Free Choice Act,” see CRS Report RS21887, The Employee Free Choice
Act, by Jon O. Shimabukuro. Legislation that would require card check recognition was introduced in Congress as early thth
as the 95 Congress (1977-1978). Early in the 95 Congress, Representative Frank Thompson Jr. introduced H.R. 77,
the Labor Reform Act of 1977. H.R. 77 would have required the NLRB to certify a union as the exclusive bargaining
representative of employees in a bargaining unit if 55% of the employees signed authorization cards. No hearings were th
held and no action was taken on the bill. Later in the 95 Congress, President Jimmy Carter sent to Congress proposals
to amend the NLRA. H.R. 8410/S. 1883, which was also called the Labor Reform Act of 1977, was introduced in the
House by Representative Frank Thompson Jr. and in the Senate by Senators Harrison Williams Jr. and Jacob Javits.
H.R. 8410/S. 1883 would have created timetables for holding representation elections. The bill passed the House. In the
Senate, the Senate Human Resources Committee reported a bill (renumbered as S. 2467). The measure was filibustered
on the Senate floor. After six cloture votes, the bill was returned to committee for changes. The committee did not
report another bill.
6 U.S. Congress, Senate Committee on Health, Education, Labor, and Pensions, The Employee Free Choice Act:
Restoring Economic Opportunity for Working Families, hearing on the Employee Free Choice Act, 110th Cong., 1st
sess., March 27, 2007, available at http://help.senate.gov.
7 Representative Charlie Norwood died on February 13, 2007.
8 S. 1312 includes a provision not in H.R. 866. Section 4(b) of S. 1312 states thatNothing in this Act (or the
amendments made by this Act) shall be construed to limit or otherwise diminish the remedial authority of the National
Labor Relations Board.” With this provision, the NLRB should, for example, be able to issue a bargaining order as a
remedy to employer unfair labor practices. See the section on “Bargaining Orders” later in this report.






The NLRA, as amended, provides the basic framework governing labor-management relations in 9
the private sector. The act begins by stating that the purpose of the law is to improve the
bargaining power of workers:
The inequality of bargaining power between employees ... and employers ... substantially
burdens and affects the flow of commerce, and tends to aggravate recurrent business
depressions by depressing wage rates and the purchasing power of wage earners ... and by
preventing the stabilization of competitive wage rates and working conditions within and
between industries....
It is hereby declared to be the policy of the United States to eliminate the causes of certain
substantial obstructions to the free flow of commerce and to mitigate and eliminate these
obstructions when they have occurred by encouraging the practice and procedure of 10
collective bargaining....
The NLRA gives workers the right to join or form a labor union and to bargain collectively over
wages, hours, and other conditions of employment through a representative of their choosing.
Under the act, workers also have the right not to join a union. The act requires an employer to
bargain in good faith with a union chosen by a majority of employees. To protect the rights of 11
employers and workers, the act defines certain activities as unfair labor practices.
The NLRA does not apply to railroads or airlines, federal, state, or local governments, agricultural
workers, family domestic workers, supervisors, independent contractors, and others. The
definition of “employee” in the NLRA does not exclude unauthorized workers. Thus, 12
unauthorized workers can engage in union activities.

9More specifically, the NLRA applies to employers engaged in interstate commerce. 29 U.S.C. § 152(6).
10 29 U.S.C. § 151. Many economists argue that there is not an inequality of bargaining power between employers and
employees. For example, see Morgan O. Reynolds, Power and Privilege: Labor Unions in America, New York:
Universe Books, 1984, pp. 59-62; and Morgan O. Reynolds, The Myth of Labor’s Inequality of Bargaining Power,”
Journal of Labor Research, vol. 12, spring 1991, pp. 168-183. The argument that workers and employers have equal
bargaining power is generally based on the premise that labor markets fit the economic model of perfect competition.
See the section later in this report on whether there is an economic rationale for protecting the rights of workers to
organize and bargain collectively.
11 National Labor Relations Board, Basic Guide to the National Labor Relations Act (Washington: U.S. Govt. Print.
Off., 1997), pp. 1, 17, available at http://www.nlrb.gov. (Hereinafter cited as NLRB, Basic Guide to the NLRA.)
The Labor Management Relations Act of 1947 (P.L. 80-101, commonly called the Taft-Hartley Act) amended the
NLRA to add language that employees have the right to refrain from joining a union, unless a collective bargaining
agreement with a union security agreement is in effect. A union security agreement may require bargaining unit
employees to join the union after being hired (i.e., a union shop) or, if the employee is not required to join the union, to
pay a representation fee to the union (i.e., an agency shop). Under Section 14(b) of the Taft-Hartley Act, states may
enact right-to-work laws, which do not allow contracts to include a union security agreement. Michael Ballot, Laurie
Lichter-Heath, Thomas Kail, and Ruth Wang, Labor-Management Relations in a Changing Environment, New York:
John Wiley and Sons, Inc., 1992, pp. 265-268.
12 NLRB, Basic Guide to the NLRA, p. 37.
In a 1984 decision (Sure-Tan, Inc. v. NLRB, 467 U.S. 883), the U.S. Supreme Court ruled that the definition of
employee in the NLRA does not exclude unauthorized workers. Thus, unauthorized workers can engage in union
organizing and collective bargaining, can vote in NLRB elections, and are protected from unfair labor practices. But,
the Supreme Court has also ruled that unauthorized workers cannot be awarded back pay as the result of violations of
unfair labor practices. See CRS Report RS21186, Hoffman Plastic Compounds v. NLRB and Backpay Awards to
(continued...)





The NLRA is administered and enforced by the NLRB, which is an independent federal agency
that consists of a five-member Board and a General Counsel. The Board hears appeals of unfair
labor practices and resolves questions about the composition of bargaining units. It also resolves 13
objections and challenges to secret ballot elections. The General Counsel’s office conducts
secret ballot elections, investigates complaints of unfair labor practices, and supervises the 14
NLRB’s regional and other field offices.
Campaign rules differ for employers, employees, and union organizers. Rules also differ for
soliciting union support (e.g., expressing support for a union or distributing authorization cards)
and for distributing union literature. Because of exceptions to the basic rules, the rules that apply 15
to a specific union organizing campaign may differ from the general rules described here.
Employers may campaign against unionization.16 Employers may require employees to attend
meetings during work hours where management can give its position on unionization. These
meetings are called “captive audience” meetings. Employers cannot hold a captive audience
meeting during the 24-hour period before an election. Supervisors can give employees written
information (including memos and letters) and hold individual meetings with employees.

(...continued)
Undocumented Aliens, by Jon O. Shimabukuro.
13 A bargaining unit is a group of employees represented, or seeking representation, by a union. A bargaining unit is
generally determined on the basis of acommunity of interest of the employees involved. Employees who have the
same or similar interests with respect to wages, hours, and other working conditions may be grouped together into a
bargaining unit. A bargaining unit may include the employees of one employer, one establishment, or one occupation
or craft. A bargaining unit may include both professional and nonprofessional employees, provided a majority of
professional employees vote to be members of the unit. Guards cannot be included in the same bargaining unit as other
employees. A union and employer may agree on the appropriate bargaining unit. If not, the issue is settled by the
NLRB. NLRB, Basic Guide to the National Labor Relations Act, p. 7. Feldacker, Labor Guide to Labor Law, pp. 39-
44.
14 NLRB, Basic Guide to the NLRA, p. 33. William N. Cooke, Union Organizing and Public Policy: Failure to Secure
First Contracts (Kalamazoo, MI: W.E. Upjohn Institute, 1985), p. 85.
15 Unless noted otherwise, this section is based on Stephen I. Schlossberg and Judith A. Scott, Organizing and the Law,
4th ed., Washington: Bureau of National Affairs, 1991, pp. 45-55. (Hereinafter cited as Schlossberg and Scott,
Organizing and the Law.) James J. Brudney, Neutrality Agreements and Card Check Recognition: Prospects for
Changing Paradigms, Public Law and Legal Theory Working Paper Series No. 28, November 2004, p. 8, available at
http://www.law.bepress.com/osulwps. (Hereinafter cited as Brudney, Neutrality Agreements and Card Check
Recognition.) Feldacker, Labor Guide to Labor Law, pp. 74-79.
16 The Taft-Hartley Act of 1947 amended the NLRA to add Section 8(c), which gives employers and unions the right to
express their views on unionization, provided suchexpression contains no threat of reprisal or force or promise of
benefit.” For a legal history of this provision, see Kate E. Andrias, A Robust Public Debate: Realizing Free Speech in
Workplace Representation Elections, Yale Law Journal, vol. 112, June 2003, pp. 2419-2432.





During work hours, employees can campaign for union support from their coworkers in both
work and nonwork areas (e.g., a coffee room or the company parking lot). But employees can
only campaign on their own time (e.g., at lunchtime or during breaks). If an employer does not
allow the distribution of literature in work areas, employees may only distribute union literature
in nonwork areas. If an employer allows the distribution of other kinds of literature in work areas,
employees may also distribute union literature in those areas.
An employer may prevent employees from using the employer’s e-mail for union activities (e.g.,
organizing and bargaining), provided the employer does not allow employees to use their work e-17
mail to solicit support for other causes or organizations. Conversely, if an employer allows
employees to use their work e-mail to solicit support for other causes or organizations, employees 18
may also use their work e-mail for union activities.
In general, union organizers cannot conduct an organizing campaign on company property. A
union cannot reply to an employer’s captive audience speech if the union has other means of 19
reaching employees. Organizers may be allowed in the workplace if the site is inaccessible (e.g.,
a logging camp or remote hotel) or if the employer allows nonemployees to solicit on company
property. Organizers may meet with employees on union property. They may also meet with
workers and distribute literature in public areas on employer property (e.g., a cafeteria or parking
lot) or in public areas (e.g., sidewalks or parking areas). Organizers may also contact employees 20
at home by phone or mail or may visit employees at home. Under a neutrality agreement
(described later in this report), an employer may allow organizers onto company property.
To protect the rights of both employees and employers, the NLRA defines certain activities as
unfair labor practices.

17 In a December 2007 decision, the Board ruled that an employers e-mail system is the employers property and that
employees do not have a statutory right to use their work e-mail for union activities. National Labor Relations Board,
The Guard Publishing Company d/b/a The Register-Guard and Eugene Newspaper Guild, CWA Local 37194, vol. 351,
no. 70, December 16, 2007, available at http://www.nlrb.gov/shared_files/Board Decisions/351/V35170.pdf, pp. 1, 5, 7.
18 National Labor Relations Board, Office of the General Counsel, Report on Case Developments, May 15, 2008,
available at http://www.nlrb.gov/shared_files/GCMemo/2008/GC 08-07 Report on Case Development.pdf.
19 Comment, “Labor Law Reform: The Regulation of Free Speech and Equal Access in NLRB Representation
Elections, University of Pennsylvania Law Review, vol. 127, January 1979, p. 768.
20 Under what is known as the “Excelsior” rule, within seven days after the NLRB has directed that a representation
election be held or after a union and employer have agreed to hold an election, an employer must provide the regional
director of the NLRB a list of the names and addresses of employees eligible to vote in the election. This list is made
available to all parties. National Labor Relations Board, Office of the General Counsel, An Outline of Law and
Procedures in Representation Cases, Washington: U.S. Govt. Print. Off., April 2002, p. 251. U.S. Departments of
Labor and Commerce, Fact Finding Report: Commission on the Future of Worker-Management Relations, May 1994,
available at http://digitalcommons.ilr.cornell.edu/cgi/viewcontent.cgi?article=1279&context=key_workplace, p. 68.
The latter report is popularly called theDunlop report,” after former Secretary of Labor John T. Dunlop, who chaired
the commission. (It is hereinafter cited as John T. Dunlop, Fact Finding Report: Commission on the Future of Worker-
Management Relations.)





Although employers have the right to campaign against unionization, they cannot interfere with,
restrain, or coerce employees in their right to form or join a union. An employer cannot threaten
employees with the loss of their jobs or benefits if they vote for a union or join a union. An
employer cannot threaten to close a plant should employees choose to be represented by a union.
An employer cannot raise wages to discourage workers from joining or forming a union. An
employer cannot discriminate against employees with respect to the conditions of employment
(e.g, fire, demote, or give unfavorable work assignments) because of union activities. An
employer must bargain in good faith with respect to wages, hours, and other working 21
conditions.
Employees have the right to organize and bargain collectively. But a union cannot restrain or
coerce employees to join or not join a union. A union cannot threaten employees with the loss of
their jobs if they do not support unionization. A union cannot cause an employer to discriminate
against employees with respect to the conditions of employment. A union must bargain in good
faith with respect to wages, hours, and other working conditions. A union cannot boycott or strike 22
an employer that is a customer of or supplier to an employer that the union is trying to organize.
An unfair labor practice may be filed by an employee, employer, labor union, or any other person.
After an unfair labor practice charge is filed, regional staff of the NLRB investigate to determine
whether there is reason to believe that the act has been violated. If no violation is found, the
charge is dismissed or withdrawn. If a charge has merit, the regional director first seeks a
voluntary settlement. If this effort fails, the case is heard by an NLRB administrative law judge. 23
Decisions by administrative law judges can be appealed to the five-member Board.
The NLRA attempts to prevent and remedy unfair labor practices. The purpose of the act is not to
punish employers, unions, or individuals who commit unfair labor practices. The act allows the
NLRB to issue cease-and-desist orders to stop unfair labor practices and to order remedies for
violations of unfair labor practices. If an employer improperly fires an employee for engaging in
union activities, the employer may be required to reinstate the employee (to their prior or
equivalent job) with back pay. If a union causes a worker to be fired, the union may be 24
responsible for the worker’s back pay.

21 NLRB, Basic Guide to the NLRA, pp. 14-22.
22 Ibid., pp. 23-32.
23 National Labor Relations Board, Annual Report of the National Labor Relations Board, for the Fiscal Year Ended
September 30, 2007 (Washington: U.S. Govt. Print. Off., October 16, 2008), p. 4, available at http://www.nlrb.gov.
(Hereinafter cited as NLRB, Annual Report, Fiscal Year 2007.) NLRB, Basic Guide to the NLRA, p. 36.
24 29 U.S.C. § 160(c). NLRB, Basic Guide to the NLRA, p. 38.
The amount of back pay awarded isnet back pay plus interest. Net back pay is the amount of compensation (i.e.,
wages plus benefits) that a worker would have received if he or she had not been unlawfully fired less the amount of
compensation received (less the expenses from looking for work) from other work during the back pay period. If a
discharged employee is able to work but does not look for work, compensation that he or she could have received from
work may be deducted from gross back pay. Interest on net back pay is simple interest (i.e., not compounded). National
(continued...)





In FY2007, 29,821 employees were awarded $117.3 million in back pay. Employers paid $116.8 25
million to 29,559 employees, while unions paid $0.5 million to 262 employees. Back pay can be
awarded to workers who were fired, demoted, denied work, or were otherwise discriminated
against for union activities. Estimates of the number of workers who are illegally fired for union 26
activities range from 1,000 to 3,000 a year, with more firings in the 1980s than in later years. In
a study of 400 NLRB election campaigns conducted in 1998 and 1999, Kate Bronfenbrenner 27
concluded that workers are fired for union activities in 25% of union campaigns.
Figure 1 shows the trend in the number of unfair labor practice charges filed from FY1970 to
FY2007. During this period, the number of charges filed peaked at 44,063 in FY1980. The 28
number stood at 22,331 in FY2007. In FY2007, 38.7% of the charges filed were found to have 29
merit. In FY2007, 73.2% of the charges were filed against employers (by unions or individuals) 30
and 26.8% were filed against unions (by employers or individuals).

(...continued)
Labor Relations Board, NLRB Casehandling Manual, Part 3, Compliance Proceedings, available at
http://www.nlrb.gov/Publications/Manuals, §§ 10536 and 10566.
In a September 2007 decision, the Board ruled that if a worker who is fired for union activities does not start to look for
work within two weeks of being fired, back pay does not begin to accrue until the worker starts to look for work.
National Labor Relations Board, Grosvenor Resort, vol. 350, no. 86, September 11, 2007, available at
http://www.nlrb.gov/shared_files/Board%20Decisions/350/v35086.pdf, p. 3.
25 NLRB, Annual Report, Fiscal Year 2007, Table 4.
26 John Schmitt and Ben Zipperer, Dropping the Ax: Illegal Firings During Union Election Campaigns, Center for
Economic and Policy Research, January 2007, available at http://www.cepr.net/documents/publications/
unions_2007_01.pdf, p. 9. John T. Dunlop, Fact Finding Report: Commission on the Future of Worker-Management
Relations, pp. 69-70, 84.
27 Kate Bronfenbrenner, Uneasy Terrain: The Impact of Capital Mobility on Workers, Wages and Union Organizing,
U.S. Trade Deficit Review Commission, 2000, available at http://digitalcommons.ilr.cornell.edu/cgi/
viewcontent.cgi?article=1002&context=reports, p.743.
28 NLRB, Annual Report, Fiscal Year 2007, Chart 2.
29 NLRB, Annual Report, Fiscal Year 2007, Chart 5. From FY1970 to FY2007, the percent of unfair labor practice
charges found to have merit ranged from about 30% to 40%. NLRB, Annual Report, various years.
30 The percentage calculations do not include charges alleging a “hot cargo” agreement or charges that a union did not
give at least a 10-day notice before picketing or striking a health care institution. (Ahot cargo agreement is where an
employer and union agree that the employer will not do business with another employer.) NLRB, Annual Report,
Fiscal Year 2007, Table 2. NLRB, Basic Guide to the NLRA, pp. 21, 32.





Figure 1. Unfair Labor Practice Charges,
Fiscal Years 1970-2007
Source: NLRB, Annual Reports, various years.

Section 9(a) of NLRA states that a union may be “designated or selected for the purposes of
collective bargaining by the majority of the employees” (emphasis added). Currently, there are
three ways for employees to join or form a union. First, a union that is selected by a majority of
employees in an election conducted by the NLRB is certified as the employees’ bargaining
representative. Second, an employer may voluntarily recognize a union if a majority of employees
in a bargaining unit have signed authorization cards. Finally, the NLRB may order an employer to
recognize and bargain with a union if a majority of employees have signed authorization cards
and the employer has engaged in unfair labor practices that make a fair election unlikely.
A union must be certified through a secret ballot election or recognized by an employer before
collective bargaining can begin. As discussed below under “Certification,” a union that is
certified as the result of a secret ballot election has certain advantages over a union that is
recognized by an employer without an election.
The NLRB conducts a secret ballot election when a petition is filed requesting one. A petition can
be filed by a union, worker, or employer. Employees or a union may petition the NLRB for an
election if at least 30% of employees have signed a petition or authorization cards. An employer
may request an election if a union has claimed to represent a majority of its employees and has
asked to bargain with the employer (and the union itself has not requested an election). An





employer is not required to give a reason for requesting an election.31 If a majority of employees
voting (i.e., not a majority of employees in the bargaining unit) in an NLRB-conducted election
choose to be represented by a union, the union is certified by the NLRB as the employees’ 32
bargaining representative. The NLRA does not provide a timetable for holding an election.
Certification of a union by the NLRB does not require that a union and employer reach an initial 33
contract agreement.
After a petition is filed requesting an election, the employer and union may agree on the time and
place for the election and on the composition of the bargaining unit. If an agreement is not
reached between the employer and union, a hearing may be held in the regional office of the
NLRB. The regional director may then direct that an election be held. The regional director’s 34
decision may be appealed to the Board.
In a secret ballot election, employees choose whether to be represented by a labor union. If an
election has more than one union on the ballot and no choice receives a majority of the vote, the 35
two unions with the most votes face each other in a runoff election.
The right of an individual to vote in an NLRB election may be challenged by either the employer
or union. If the number of challenged ballots could affect the outcome of an election, the regional
director determines whether the ballots should be counted. Either the employer or union may file
objections to an election, claiming that the election or the conduct of one of the parties did not
meet NLRB standards. A regional director’s decision on challenges or objections may be 36
appealed to the Board.
A union and employer may also agree to a secret ballot election conducted by a third party, such 37
as an arbitrator, clergyman, or mediation board.
The NLRB also conducts secret ballot elections to decertify a union that has previously been
certified or recognized. A decertification petition may be filed by employees or a union acting on
behalf of employees. A decertification petition must be signed by at least 30% of the employees

31 U.S. Supreme Court, “National Labor Relations Board v. Gissel Packing Co., Inc.,” United States Reports, vol. 395
(Washington: U.S. Govt. Print. Off., 1969), pp. 593-594, 609. (Hereinafter cited as U.S. Supreme Court, NLRB v.
Gissel Packing.) In NLRB v. Gissel Packing, the U.S. Supreme Court consolidated four NLRB cases. In each case, a
majority of employees signed authorization cards. The employer refused to bargain, arguing that authorization cards are
inherently unreliable. The NLRB concluded that the employers committed unfair labor practices that made a fair
election unlikely and ordered the employers to bargain with the respective unions. U.S. Supreme Court, NLRB v. Gissel
Packing, pp. 575-595.
32 29 U.S.C. § 159(c). NLRB, Annual Report, Fiscal Year 2007, p. 27. National Labor Relations Board, The NLRB:
What it is, What it Does, National Labor Relations Board, p. 3, available at http://www.nlrb.gov. U.S. Supreme Court,
NLRB v. Gissel Packing, pp. 593-594, 609. NLRB, Basic Guide to the NLRA, p. 8.
33 Some evidence indicates that within three years of winning an election, approximately one-fourth of unions have not
reached a first contract with the employer. Thomas F. Reed, “Union Attainment of First Contracts: Do Service Unions
Possess a Competitive Advantage? Journal of Labor Research, vol. 11, fall 1990, pp., 426, 430. William N. Cooke,
The Failure to Negotiate First Contacts: Determinants and Policy Implications,” Industrial and Labor Relations
Review, vol. 38, January 1985, p. 170.
34 NLRB, Basic Guide to the NLRA, pp. 8-9. National Labor Relations Board, NLRB Rules and Regulations, available
at http://www.nlrb.gov/nlrb/, Sec. 102.67.
35 NLRB, Basic Guide to the NLRA, p. 36.
36 NLRB, Annual Report, Fiscal Year 2007, pp. 5, 110, 113.
37 Schlossberg and Scott, Organizing and the Law, p. 176.





in the bargaining unit represented by the union. Under what is called a “certification bar,” a union
that is certified after winning a secret ballot election is protected for a year from a decertification
petition and from an election petition filed by another union. A secret ballot election is required 38
for decertification.
Table 1 shows the number of secret ballot elections conducted by the NLRB from FY1994 to
FY2007. In FY2007, the NLRB conducted 1,905 elections. Unions won 54.9% of these elections, 39
which was up from 44.4% in FY1994.
In most elections conducted by the NLRB, the employer and union agree on the composition of
the bargaining unit and on the time and place for an election. In FY2007, 85.6% of elections were 40
based on agreements between the two parties.
Table 1. Number of Representation Elections Conducted by the NLRB,
FY1994-FY2007
Fiscal Year Number of Elections Conducted Number of Elections Won by Unions Percentage of Elections Won by Unions
2007 1,905 1,045 54.9%
2006 2,181 1,195 54.8%
2005 2,745 1,504 54.8%
2004 2,826 1,447 51.2%
2003 3,077 1,579 51.3%
2002 3,151 1,606 51.0%
2001 3,975 1,591 40.0%
2000 3,467 1,685 48.6%
1999 3,743 1,811 48.4%
1998 4,001 1,856 46.4%
1997 3,687 1,677 45.5%
1996 3,470 1,469 42.3%
1995 3,632 1,611 44.4%
1994 3,752 1,665 44.4%
Sources: National Labor Relations Board, Annual Report of the National Labor Relations Board, for the Fiscal Year
Ended September 30, 2007 (Washington: U.S. Govt. Print. Off., October 16, 2008), Chart 12, available at

38 NLRB, Annual Report, Fiscal Year 2007, p. 27. National Labor Relations Board, The National Labor Relations
Board and YOU: Representation Cases, p. 2, available at http://www.nlrb.gov. Feldacker, Labor Guide to Labor Law,
p. 57. House, Committee on Education and the Workforce, Subcommittee on Employer-Employee Relations, H.R. th
4343, Secret Ballot Protection Act of 2004, hearings, 108 Congress, second session, Serial No. 108-70, September
2004, Washington: U.S. Govt. Print. Off., p. 11. (Hereinafter cited as House Education and the Workforce, H.R. 4343,
Secret Ballot Protection Act of 2004.)
39 NLRB, Annual Report, Fiscal Year 2007, Chart 12.
40 NLRB, Annual Report, Fiscal Year 2007, Table 11A.





http://www.nlrb.gov. National Labor Relations Board, Annual Report of the National Labor Relations Board, for the
Fiscal Year Ended September 30, 2003 (Washington: U.S. Govt. Print. Off., April 20, 2004), Chart 12, available at
http://www.nlrb.gov.
Note: The number of elections conducted includes elections that resulted in a runoff or rerun.
The NLRB does not collect data on how long it takes for a union and employer to reach a first
contract agreement after a union wins an NLRB election. Nor does the NLRB collect data on
whether the parties reach a first contract agreement. However, a recent study estimated that,
within two years of winning an NLRB election, a contract had not been reached in over two-fifths
of cases. This is a higher percentage than found in estimates published in previous studies.
Estimates from several studies are shown in Table 2.
Table 2. First Contract Agreements Following Certification
Period A First Contract A First Contract Agreement
Studied Sample Agreement Was Reached Was Not Reached
October 1, First contract agreement after In 56% of certifications, a In 44% of certifications, a contract
1999 to June a union won an NLRB contract was agreed to within was not agreed to within two
1, 2004a election. two years of the election. years of the election.
FY1986 to First contract agreement after At least 56% of certifications At most, 44% of certifications did
FY1993b a union won an NLRB resulted in a first contract. not result in a first contract. The
election. The actual percentage may be actual percentage may be closer
closer to two-thirds. to one-third.
1982 to cFirst contract agreement after 85% of service unions achieved 15% of service unions did not
1986 a union won an NLRB a first contract agreement; reach a first contract agreement;
election. 64% of manufacturing unions 36% of manufacturing unions did
achieved a first contract. not reach a first contract
agreement.
April 1979 First contract agreement after
to March da union won an NLRB
1981 election. Sample included 63% 37%
bargaining units of 100 or
more employees.
1979 to First contract agreement after
1980e a union won an NLRB 77% 33%
election.
1970f First contract agreement after
a union won an NLRB 78% 22%
election.
July 1, 1957 First contract agreement after
to June 30, a union won an NLRB 86% 14%
1962g election.
a. The estimate is based on 8,155 NLRB elections won by unions in cases closed between October 1, 1999
and June 1, 2004. Employers must bargain in good faith for one year after an NLRB election is won by a
union. Therefore, the study used information from the FMCS for the period from October 1, 1999 to June
1, 2005. In recent years, the FMCS has attempted to contact the parties involved in first contract
negotiations. John-Paul Ferguson, “The Eyes of the Needles: A Sequential Model of Union Organizing
Drives, 1999-2004,” Industrial and Labor Relations Review, vol. 62, October 2008, pp. 3-6.
b. The 56% estimate is based on 10,783 union elections certified by the NLRB and contract agreements in
which the FMCS was involved. Other certifications may have resulted in a first contract agreement (e.g.,





where the FMCS was not contacted for help). Therefore, the actual percentage of certifications that
resulted in a first contract may be closer to two-thirds. The estimates were calculated by the FMCS for the
Commission on the Future of Worker-Management Relations (the “Dunlop Commission”). U.S.
Departments of Labor and Commerce, Fact Finding Report: Commission on the Future of Worker-Management
Relations, May 1994, available at
[digitalcommons.ilr.cornell.edu/cgi/viewcontent.cgi?article=1279&context=key_workplace], pp. 73, 87.
c. The estimates are based on a survey of union organizers involved in 128 elections won by unions. Thomas
F. Reed, “Union Attainment of First Contracts: Do Service Unions Possess a Competitive Advantage?,”
Journal of Labor Research, vol. 11, Fall 1990, pp. 428-430.
d. The estimates are based on a survey by the AFL-CIO of 271 elections won by unions. William N. Cooke,
“The Failure to Negotiate First Contracts: Determinants and Policy Implications,” Industrial and Labor
Relations Review, vol. 38, January 1985, p. 164.
e. The estimates are based on a survey of unions that won 118 elections in Indiana during 1979 and 1980. The
survey was conducted between June 1, 1982 and October 1, 1982. William N. Cooke, “The Failure to
Negotiate First Contracts: Determinants and Policy Implications,” Industrial and Labor Relations Review, vol.
38, January 1985, pp. 169-170.
f. The estimates are based on a 1975 survey by the AFL-CIO of 2,656 elections won by unions in 1970. The
estimates are based on the number of responses, which was not reported. William N. Cooke, “The Failure
to Negotiate First Contracts: Determinants and Policy Implications,” Industrial and Labor Relations Review,
vol. 38, January 1985, p. 164.
g. Paul Weiler, “Striking a New Balance: Freedom of Contract and the Prospects for Union Representation,”
Harvard Law Review, vol. 98, December 1984, pp. 353-355.
Although the NLRA does not provide a specific timetable for holding an election, most elections
are held within two months of the filing of a petition. In FY2007, 93.9% of initial representation
elections were conducted within 56 days of filing a petition. The median time to proceed to an 41
election from the filing of a petition was 39 days.
In FY2007, objections were filed in 163 of the elections conducted. Most (66.9%) of the
objections were filed by unions. The remainder were filed by employers (33.1%) or by both 42
parties. For decisions reached in FY2007, it took a median of 102 days between a regional 43
hearing on a contested election and a decision from the Board. As of September 30, 2007,
representation cases awaiting a Board decision had been pending for a median of 318 days from 44
the date that an election petition was filed.
The NLRA does not require secret ballot elections. An employer may voluntarily recognize a
union when presented with authorization cards signed by a majority of employees in a bargaining
unit. An employer may also enter into a card check agreement with a union before union

41 National Labor Relations Board, General Counsel, Summary of Operations: Fiscal Year 2007, Memorandum GC 08-
01, p. 6, available at http://www.nlrb.gov.
42 Ibid., Table 11C.
43 Ibid., Table 23.
An analysis by the General Accounting Office (GAO) of cases appealed to the Board found that among cases closed
between 1984 and 1989 the median time from the date of regional action on an appeal to a decision by the Board was
between 190 and 256 days. U.S. General Accounting Office, National Labor Relations Board: Action Needed to
Improve Case-Processing Time at Headquarters, Report HRD-91-29, January 1991, pp. 21-22. The General
Accounting Office is now called the Government Accountability Office.
44 NLRB, Annual Report, Fiscal Year 2007, Table 23.





organizers begin to collect signatures. A card check agreement between a union and employer
may require the union to collect signatures from more than a majority (sometimes called a 45
supermajority) of bargaining unit employees. A neutral third party often checks, or validates,
signatures on authorization cards. A collective bargaining contract may include a card check
arrangement for unorganized (including new) branches, stores, or divisions of a company.
Under voluntary recognition, employees have 45 days to file a decertification petition or an
election petition requesting representation by another union. After 45 days, an election petition
cannot be filed for “a reasonable period of time.” (See the section on “NLRB Review of
Voluntary Recognition” later in this report.)
A card check arrangement may be combined with a neutrality agreement. Not all neutrality
agreements are the same. However, in general, under a neutrality agreement an employer agrees
to remain neutral during a union organizing campaign. The employer may agree not to attack or
criticize the union, while the union may agree not to attack or criticize the employer. The
agreement may allow managers to answer questions or provide factual information to employees.
A neutrality agreement may give a union access to company property to meet with employees and
distribute literature. An employer may also agree to give the union a list of employee names and 46
addresses. A neutrality agreement may cover organizing drives at new branches of a company.
To gain an agreement from an employer for a card check campaign—possibly combined with a
neutrality agreement—unions sometimes engage in “corporate campaigns.” A corporate campaign
may include a call for consumers to boycott the employer; rallies and picketing; a public relations
campaign (e.g., press releases, Internet postings, news conferences, or newspaper and television
ads); charges that the employer has violated labor or other laws; public support from political, 47
civic, and religious leaders; and other strategies.

45 One study of card check agreements found that, under some agreements, a union needed signatures from at least 65%
of bargaining unit employees. Adrienne E. Eaton and Jill Kriesky, Union Organizing Under Neutrality and Card
Check Agreements, Industrial and Labor Relations Review, vol. 55, October 2001, p. 48. (Hereinafter cited as Eaton
and Kriesky, Union Organizing Under Neutrality and Card Check Agreements.)
46 Eaton and Kriesky, Union Organizing Under Neutrality and Card Check Agreements, pp. 47-48. Charles I. Cohen,
“Neutrality Agreements: Will the NRLB Sanction Its Own Obsolescence?” The Labor Lawyer, vol. 16, fall 2000, pp.
203-204. Brudney, Neutrality Agreements and Card Check Recognition, pp. 5-6.
It has been argued that, under the NLRA, neutrality and card check agreements, may be unlawful. See Arch Stokes,
Robert L. Murphy, Paul E. Wagner, and David S. Sherwyn, “Neutrality Agreements: How Unions Organize New
Hotels Without an Employee Ballot,” Cornell Hotel and Restaurant Administration Quarterly, vol. 42, October-
November 2001, pp. 91-94. A counter argument can be found in Brudney, Neutrality Agreements and Card Check
Recognition, pp. 28-53.
47 A union may engage in a corporate campaign to achieve other objectives, e.g., a contract agreement. Charles R.
Perry, Union Corporate Campaigns (Philadelphia: Industrial Research Unit, Wharton School, University of
Pennsylvania, 1987), pp. 1-8, 37-53.
For different views on corporate campaigns, see U.S. Congress, House Committee on Education and the Workforce, th
Subcommittee on Workforce Protections, Compulsory Union Dues and Corporate Campaigns, hearings, 107 Cong., nd
2 sess., July 23, 2002, Serial No. 107-74 (Washington: GPO, 2002). For a discussion of corporate campaigns
published by the U.S. Chamber of Commerce, see Jarol B. Manheim, Trends in Union Corporate Campaigns: A
Briefing Book (Washington: U.S. Chamber of Commerce, 2005), available at http://www.uschamber.com/publications/
(continued...)





Historically, the NLRB has not collected data on voluntary recognitions.48 The FMCS, however,
is involved in voluntary recognitions. The FMCS was created by the Labor Management
Relations Act of 1947 (the Taft-Hartley Act). The main purpose of the FMCS is to mediate
collective bargaining agreements. FMCS mediators act as a neutral third-party to help settle 49
issues during the bargaining process. Some of the requests received by the FMCS are for
mediation where an employer has voluntarily agreed to negotiate with a union. Table 3 shows the
number of voluntary recognitions, for FY1996 to FY2004 (the most recent year for which
published data are available), where the FMCS helped mediate a first contract. Cases where an
employer voluntarily recognized a union and reached a first contract without FMCS assistance
are not included in these numbers. Therefore, the actual number of voluntary recognitions is
probably greater than the numbers shown in Table 3.
Table 3. Number of Voluntary Recognitions in Which the Federal Mediation and
Conciliation Service (FMCS) Provided Assistance for Initial Contracts,
FY1996-FY2004
Fiscal Year Number of Voluntary Recognitions
2004 258
2003 240
2002 273
2001 420
2000 381
1999 260
1998 227
1997 249
1996 173
Source: Federal Mediation and Conciliation Service, Annual Report, Fiscal Year 2004, p. 18, available at
http://www.fmcs.gov. Federal Mediation and Conciliation Service, Annual Report, Fiscal Year 2000, p. 39, available
at http://www.fmcs.gov.
The final way that a union may be recognized by an employer is through a bargaining order. The
NLRB may order an employer to recognize and bargain with a union if a majority of employees

(...continued)
reports/06union_campaigns.htm.
48 In a September 2007 decision, the Board said that the Regional Office of the NLRB must be notified in writing (by
the employer, union, or both) of a voluntary recognition. National Labor Relations Board, Dana Corporation, vol. 351,
no. 28, September 29, 2007, available at http://www.nlrb.gov/shared_files/Board%20Decisions/351/v35128.pdf, p. 10.
(Hereinafter cited as NLRB, Dana Corporation, September 29, 2007.) Also see the section later in this report onThe
United Auto Workers (UAW) and the Dana and Metaldyne Corporations.”
49 Federal Mediation and Conciliation Service, Annual Report, Fiscal Year 2004, p. 29, available at
http://www.fmcs.gov.





have signed authorization cards and the employer has committed unfair labor practices that make
it unlikely that a fair election can be held.
According to Feldacker, “[h]ard and fast rules are not possible in determining the situations in
which the Board will issue a bargaining order. Each case is based on the specific facts of the 50
employer’s violations. Bargaining orders may be appealed to the U.S. Court of Appeals and to 51
the U.S. Supreme Court.
A union that wins a secret ballot election is certified by the NLRB as the bargaining
representative of employees in that bargaining unit. Voluntary recognition or a bargaining order
do not result in certification by the NLRB. The Taft-Hartley Act of 1947 (P.L. 80-101) eliminated 52
certification through any method other than an election conducted by the NLRB.
Certification gives a union certain advantages. For instance, a union that is certified after winning
a secret ballot election is protected for a year from a decertification petition and from an election
petition requesting representation by another union (the “certification bar”). Under voluntary card
check recognition, employees have 45 days to file a decertification petition or an election petition
requesting to be represented by a different union (the “recognition bar”).
The duration of an employer’s duty to bargain also depends on whether a union has been certified
by the Board or has been recognized voluntarily by the employer. If a union wins an NLRB
election (or under a bargaining order), the employer is required to bargain in good faith for a year.
Under voluntary card check recognition, the employer is required to bargain with the union for “a 53
reasonable period of time.”

50 Feldacker, Labor Guide to Labor Law, pp. 90-91.
51 Daniel Quinn Mills, Labor-Management Relations, 5th ed., New York: McGraw-Hill, Inc., 1994, pp. 213-217.
52 When enacted in 1935, Section 9(c) of the NLRA (P.L. 74-198) stated that whenever a question of employee
representation arises the NLRB “may take a secret ballot of employees, or utilize any other suitable method to ascertain
such representatives.” Alternative methods of selection could include authorization cards, petitions, employee
testimony, affidavits of union membership, participation in a strike, or acceptance of strike benefits. Whichever method
was used, if a majority of employees chose to be represented by a union, the union would be certified by the NLRB.
During the five years after the NLRA was enacted, the NLRB issued 897 certifications after an election and 272
certifications (or 23.3% of the total) without an election. (These numbers do not include cases where the union and
employer agreed to hold an election.) By 1939, the NLRB only certified unions that had been chosen by a secret ballot
election. This approach was written into law by the Taft-Hartley Act. The act amended Section 9(c) to say that the
Board “shall direct an election by secret ballot and shall certify the results thereof.” The words “or utilize any other
suitable method to ascertain such representatives were removed. National Labor Relations Board, Legislative History
of the National Labor Relations Act of 1935 (Washington: U.S. Govt. Print. Off., 1949), p. 3274. National Labor
Relations Board, Legislative History of the Labor Management Relations Act, 1947 (Washington: U.S. Govt. Print.
Off., 1985), p. 1670. Craig Becker, “Democracy in the Workplace: Union Representation Elections and Federal Labor
Law, Minnesota Law Review, vol. 77, 1992, pp. 507-510. Alan Roberts McFarland and Wayne S. Bishop, Union
Authorization Cards and the NLRB: A Study of Congressional Intent, Administrative Policy, and Judicial Review
(Philadelphia: University of Pennsylvania, 1969), pp. 12-14, 50. William B. Gould IV, A Primer on American Labor th
Law, 4 ed. (Cambridge, MA: MIT Press, 2004), p. 89.
53 Feldacker, Labor Guide to Labor Law, pp. 57, 139-140.





Under certain circumstances, an employer may withdraw recognition of a union before a contract
agreement has been reached. After one year, if an employer and a certified union have not
reached a contract agreement, the employer may withdraw recognition of the union if both parties
have engaged in good faith bargaining and the employer doubts, on the basis of objective
information (e.g., a petition signed by a majority of employees and given to the employer), that a
majority of employees no longer support the union. Under a voluntary recognition, if no contract
agreement has been reached after a reasonable period of time, an employer may withdraw
recognition if the employer has reasonable doubt on the basis of objective information that a 54
majority of employees support the union.
Until 1947, only a union or employees could request a secret ballot election. The Taft-Hartley Act 55
gave employers the right to request an election. Soon thereafter, a U.S. Appeals court ruled that,
if a majority of employees signed authorization cards, an employer could only insist on an 56
election if he had “good faith doubt” that the union represented a majority of employees. This 57
approach was known as the “Joy Silk doctrine.” An employer had good faith doubt if he
believed that signatures on the authorization cards were obtained through misrepresentation or 58
coercion. The Board could find a lack of good faith doubt and order an employer to bargain if
the employer committed unfair labor practices that were evidence of bad faith or if the employer
rejected a demand by a majority of workers to bargain but did not have reasons for good faith
doubt. By 1969, the Board said that it had largely abandoned the Joy Silk doctrine. Thereafter, if a
majority of employees signed authorization cards, an employer could voluntarily recognize the
union or could insist on an election, either by requesting the union to file an election petition or 59
by filing a petition himself. In a 1974 ruling, the U.S. Supreme Court said that, if an employer 60
insists on an election, the union must take the next step and file an election petition.
In recent years, the NLRB has considered cases involving voluntary recognition.

54 Feldacker, Labor Guide to Labor Law, p. 140.
55 Marie C. Grossman, “Labor Law—Employer’s Duty to BargainAuthorization Cards,Case Western Reserve Law
Review, vol. 21, 1970, p. 308.
56 Michael Eugene Earwood and Herbert C. Ehrhardt, “Labor LawEmployer’s Duty to Bargain on the Basis of
Authorization Cards—Union Has the Burden of Seeking an NLRB Election,Mississippi Law Journal, vol. 46, 1975,
p. 522.
57 Joy Silk Mills, Inc., 85 NLRB 1263 (1949)
58 McFarland and Bishop, Union Authorization Cards and the NLRB, p. 55.
59 U.S. Supreme Court, NLRB v. Gissel Packing, pp. 591-592, 594.
60 U.S. Supreme Court, “Linden Lumber Division, Summer & Co. v. National Labor Relations Board,” United States
Reports, vol. 419 (Washington: U.S. Govt. Print. Off., 1974), p. 310.





In June 2004, the Board voted 3-2 to review two cases where bargaining unit employees filed a
decertification petition within weeks after the employer recognized a union under a card check
agreement. In the first case, the United Auto Workers (UAW) and Metaldyne Corporation entered
into a card check and neutrality agreement in September 2002. Metaldyne recognized the UAW
as the bargaining representative of production and maintenance workers at its St. Marys,
Pennsylvania plant in December 2003. In the second case, the UAW and Dana Corporation
entered into a card check and neutrality agreement in August 2003. The company recognized the
union at its Upper Sandusky, Ohio plant in December 2003.
In both the Dana and Metaldyne cases, the UAW and the employers entered into card check and
neutrality agreements before signatures on authorization cards were collected. The signatures
were validated by a neutral third party. In both cases, employees filed decertification petitions
after the UAW was recognized by the employer, but before an agreement was reached on a
contract. Regional NLRB directors dismissed both decertification petitions, saying that they were
inconsistent with the Board’s “recognition bar” doctrine. Under this doctrine, following an
employer’s voluntary recognition of a union, employees or another union cannot file a petition for
an election for a “reasonable period of time.”
Employees at both Dana and Metaldyne Corporations petitioned the NLRB to review the
dismissals. The employees were represented by the National Right to Work Legal Defense
Foundation. The NLRB granted the request, saying that the issue was whether voluntary
recognition should prevent employees from filing a decertification petition within a reasonable 61
time in cases where an employer and union enter into a card check agreement.
In September 2007, the Board issued a decision in both cases. The Board said that, following a
voluntary recognition, employees have 45 days to file a petition to decertify the union. Similarly,
a rival union has 45 days to file an election petition. The petitions must be signed by at least 30%
of bargaining unit employees. Employees must also receive notice of the voluntary recognition
and their right to petition for a decertification or representation election. If a petition is not filed
within 45 days of notice of the voluntary recognition, an election petition cannot be filed during 62
the recognition bar period (i.e., for “a reasonable period of time”).
In another case involving voluntary card check recognition, the NLRB agreed to review a case
where a union claimed that an employer had agreed to voluntary card check recognition at newly
acquired stores. In December 2004, the Board by a vote of 2-1 agreed to review a case involving 63
Shaw’s Supermarkets and the United Food and Commercial Workers Union (UFCW). In August
2003, Shaw’s opened a new store in Mansfield, Massachusetts. A majority of workers at the new
store signed authorization cards. The UFCW claimed that, under a clause in an existing

61 National Labor Relations Board, Order Granting Review, Cases 8-RD-1976, 6-RD-1518, and 6-RD-1519, vol. 341,
no. 150, June 7, 2004, available at http://www.nlrb.gov/shared_files/Board Decisions/341/341-150.pdf.
62 NLRB, Dana Corporation, September 29, 2007, pp. 1, 9-10.
63 The NLRA (Section 3(b)) allows the Board to delegate decisions to a group of three or more members.





bargaining contract, Shaw’s had agreed to voluntary card check recognition at newly acquired 64
stores. Shaw’s filed a petition requesting a secret ballot election. In May 2004, an acting
regional director of the NLRB dismissed Shaw’s petition without a hearing. The Board agreed to
review the case and returned it to the regional office for a hearing. In its decision, the Board said
The issues in this case include (1) Whether the Employer clearly and unmistakably waived
the right to a Board election; (2) if so, whether public policy reasons outweigh the
Employer’s private agreement not to have an election.
The Board went on to say: “We do not resolve these issues at this stage. We merely hold that they 65
are worthy of review.” After the hearing ordered by the Board, a regional director, in March 66

2005, again dismissed Shaw’s petition for an election. In March 2006, the Board again agreed to 67


review the case. As of November 24, 2008, the Board has not issued a final decision.
Once a union and employer enter into a collective bargaining agreement, election petitions are
subject to a “contract bar.” A contract of up to three years bars an election petition for the duration 68
of the contract. The election petition may be for a decertification election or for representation
by another union.
In August 2007, the Board issued a decision allowing an employer to withdraw recognition of a
union after the third year of a longer-term contract. In January 1999, Shaw’s Supermarkets
entered into a five-year contract with the UFCW. After three years, a majority of employees
signed a petition requesting a decertification election. Instead of going forward with a
decertification election, Shaw’s withdrew recognition of the union. The action was appealed to
the NLRB.
The General Counsel of the NLRB argued that Shaw’s should not be allowed to withdraw
recognition of the union during the term of the five-year contract. The General Counsel said that,
during the term of a contract, neither the employer nor the union can file an election petition. But,
after three years of a contract of more than three years, employees or another union (but not the
employer or existing union) can file an election petition requesting decertification or
representation by another union. By a vote of 2-1, the Board disagreed with the General Counsel.
The majority members of the Board concluded that Shaw’s had acted properly when it withdrew
recognition of the union. The majority said that the employer relied on evidence of a loss of

64 These clauses have been called “after-acquired stores” clauses, “additional stores” clauses, and “Kroger” clauses.
The latter term is the result of NLRB and court decisions involving the Kroger Company.
65 National Labor Relations Board, Shaws Supermarkets and United Food and Commercial Workers Union Local 791,
AFL-CIO, Case 1-RM-1267, December 8, 2004, pp. 1-3, available at http://www.nlrb.gov/nlrb/shared_files/decisions/
343/343-105.pdf.
66 National Labor Relations Board, Shaws Supermarkets and Local 791, United Food and Commercial Workers
Union, AFL-CIO, Case 1-RM-1267, http://www.nlrb.gov/nlrb/shared_files/decisions/dde/2005/1-RM-1267(3-22-
05).pdf.
67 Bureau of National Affairs,NLRB 2-1 Grants Shaws Request for Review in Case Involving After-Acquired Store
Clause,” Daily Labor Report, no. 55, March 22, 2006, p. A-1.
68 NLRB, Basic Guide to the NLRA, p. 10.





majority support for the union (i.e., signatures of a majority of employees). The dissenting 69
member said that the NLRB should have gone forward with a decertification election.


Changes in union certification procedures may affect the level of unionization in the United 70
States. This section summarizes the most common arguments made in favor of requiring secret
ballot elections and the most common arguments made in support of card check certification. The
section also reviews research on the effects of different union certification procedures on union
success rates.
The most common arguments made by the proponents of requiring card check certification and 71
the proponents of requiring secret ballot elections are summarized in Table 4.
Proponents of each view sometimes use similar language in support of their positions. Employers
argue that, under card check certification, employees may be pressured or coerced into signing
authorization cards and that employees may only hear the union’s point of view. On the other
hand, unions argue that, during an election campaign, employers may pressure or coerce
employees into voting against a union. Proponents of secret ballot elections argue that, unlike
signing an authorization card, casting a secret ballot is private and confidential. Unions argue
that, during an election campaign, employers have greater access to employees (e.g., captive
audience meetings and access to employees on company property). Unions argue that card check
certification is less costly than a secret ballot election. But employers maintain that unionization
may be more costly to employees, because union members must pay dues and higher union wages
may result in fewer union jobs.
Little research has been done comparing the effects of requiring card check certification versus
the effects of requiring secret ballot elections. The research that exists, however, suggests that
changes in union recognition procedures could affect the level of unionization in the United
States. Research suggests that the union success rate is greater with card check certification than
with secret ballots. Unions also undertake more unionization drives under card check
certification. The union success rate under card check certification is greater when a card check
campaign is combined with a neutrality agreement.

69 National Labor Relations Board, Shaws Supermarkets, vol. 350, no. 55, August 10, 2007, available at
http://www.nlrb.gov/shared_files/Board%20Decisions/350/v35055.pdf, pp. 1-3, 7.
70 For a discussion of union membership trends in the United States, see CRS Report RL32553, Union Membership
Trends in the United States, by Gerald Mayer.
71 The arguments for and against requiring card check certification and secret ballot elections are considered in House,
Committee on Education and the Workforce, Subcommittee on Employer-Employee Relations, H.R. 4343, Secret
Ballot Protection Act of 2004.





Table 4. Common Arguments Made by Proponents of Requiring Card Check
Certification and Requiring Secret Ballots
Proponents of Requiring Proponents of Requiring
Card Check Certification Secret Ballot Elections
Card check certification requires signatures from Casting a secret ballot is private and confidential. A secret
more than 50% of bargaining unit employees. A ballot election is conducted by the NLRB. Under card check
secret ballot election is decided by a majority of certification, authorization cards are controlled by the union.
workers voting.
During a secret ballot campaign, the employer has Under card check certification, employees may only hear the
greater access to employees. union’s point of view.
Because of potential employer pressure or Because of potential union pressure or intimidation, some
intimidation during a secret ballot election, some workers may feel coerced into signing authorization cards.
workers may feel coerced into voting against a
union.
Employer objections can delay a secret ballot Most secret ballot elections are held soon after a petition is
election. filed.
Allegations against a union for unfair labor practices Allegations against an employer for unfair labor practices can
can be addressed under existing law. Existing be addressed under existing law. Existing remedies do not
remedies do not deter employer violations of unfair deter union violations of unfair labor practices.
labor practices.
Card check certification is less costly for both the Union members must pay union dues. Unionization may result
union and employer. If secret ballot elections were in fewer union jobs.
required, the NLRB would have to devote more
resources to conducting elections.
Neutrality agreements and card check certification An employer may be pressured by a corporate campaign into
may lead to more cooperative labor-management accepting a neutrality agreements and card check certification.
relations. If an employer accepts a neutrality agreement, employees who
do not want a union may hesitate to speak out.
Source: Table compiled by CRS.
Evidence from Canada suggests that the union success rate is higher under automatic card check
recognition than under secret ballots. In Canada, each of the 10 provinces has laws governing 72
union recognition. In 1976, all 10 provinces allowed card check recognition. Beginning with 73
Nova Scotia in 1977, five provinces currently require secret ballot elections. British Columbia
changed from card check recognition to requiring secret ballot elections in 1984, repealed 74
mandatory voting in 1993, and restored mandatory voting in 2001. Under mandatory voting a

72 Gary N. Chaison and Joseph B. Rose, “The Canadian Perspective on Workers Rights to Form a Union and Bargain
Collectively,” Edited by Sheldon Friedman, Richard W. Hurd, Rudolph A. Oswald, and Ronald L. Seeber, in Restoring
the Promise of American Labor Law (Ithaca, NY: ILR Press, 1994), p. 244.
73 The five Canadian provinces that currently require secret ballot elections are Alberta, British Columbia,
Newfoundland, Nova Scotia, and Ontario. Keith Godin, Milagros Palacios, Jason Clemens, Niels Veldhuis, and Amela
Karabegovic, “An Empirical Comparison of Labour Relations Laws in Canada and the United States,” Centre for
Labour Market States, No. 2, May 2006, available at http://www.fraserinstitute.ca/admin/books/files/
EmpCompLRL.pdf, p. 10. (Hereinafter cited as Godin et al., An Empirical Comparison of Labour Relations Laws in
Canada and the United States.) Susan Johnson, “The Impact of Mandatory Votes on the Canada-U.S. Union Density
Gap: A Note, Industrial Relations, vol. 43, April 2004, p. 357.
74 Beginning in 1993, British Columbia eliminated the requirement for secret ballot elections. Union certification
occurred when at least 55% of employees signed authorization cards. Elections were held if 45% to 55% of employees
(continued...)





union must receive a majority of votes in a secret ballot election to be recognized as the
bargaining agent. Under card check recognition, a union is automatically recognized if the
number of employees who sign authorization cards meets a minimum threshold. In general, a
union is automatically recognized if more than 50% to 65% of employees, depending on the 75
province, sign authorization cards.
A study of the union success rate under mandatory voting and automatic card check recognition
concluded that the union success rate in Canada is nine percentage points higher under card check
recognition than under secret ballots. The study examined 171 union organizing campaigns 76
between 1978 and 1996 in nine provinces.
In the province of British Columbia, union recognition based on card checks was allowed until
1984. From 1984 through 1992, union certification required a secret ballot election. Card checks
were again allowed beginning in 1993. (As noted above, mandatory voting resumed in 2001.) The
union success rate fell almost 19 percentage points (from 93.1% to 74.5%) after mandatory voting
was adopted in 1984 and increased by about the same amount when card check recognition was
reinstated in 1993. In addition, during the period when mandatory voting was in effect, there were
about 50% fewer attempts to organize workers. After 1993, the number of union organizing 77
drives did not return to their pre-1984 levels.
Evidence also suggests that card check recognition may be more successful under a neutrality
agreement. A study of union organizing drives in the United States concluded that union success
rates are higher when a card check agreement is combined with a neutrality agreement. The study
examined 57 card check agreements involving 294 organizing drives. Unions had a success rate
of 78.2% in drives where card check recognition was combined with a neutrality agreement and a 78

62.5% success rate in cases where there was only a card check agreement.



(...continued)
signed authorization cards. Elections were held within ten days, or a longer period if the election was conducted by
mail. Canada, Human Resources and Social Development, Highlights of Major Developments in Labour Legislation
(1992-1993), available at http://www.hrsdc.gc.ca/en/lp/spila/clli/dllc/11_1992_1993.shtml. Beginning in 2001, secret
ballot elections were requiredwhen at least 45% of employees in a bargaining unit signed authorization cards. An
election must be held within 10 days, or longer if the vote is conducted by mail. Canada, Human Resources and Social
Development, Highlights of Major Developments in Labour Legislation (2000-2001), available at
http://www.hrsdc.gc.ca/en/lp/spila/clli/dllc/03_2000_2001.shtml.
75 Godin et al., An Empirical Comparison of Labour Relations Laws in Canada and the United States, p. 11.
76 Susan Johnson,Card Check or Mandatory Representation Vote? How the Type of Union Recognition Procedure
Affects Union Certification Success,Economic Journal, vol. 112, pp. 355-359.
77 The data are based on 6,550 private sector union drives from 1978 to 1998. The calculations of the union success rate
are for the six years before 1984when card check recognition was in effect, the nine years from 1984 to 1992 when
mandatory voting was in effect, and the six years from 1993 to 1998 after card check recognition was restored. Chris
Riddell,Union Certification Success Under Voting Versus Card-Check Procedures: Evidence from British Columbia,
1978-1998,” Industrial and Labor Relations Review, vol. 57, July 2004, pp. 493-494, 506-507, 510.
78 The success rate was measured as the percentage of organizing campaigns that resulted in union recognition. The
results include some agreements in the public sector. Some of the agreements were with employers where a union
represented other workers. Some of the agreements were with employers with whom the union had no existing
bargaining relationship. Eaton and Kriesky, Union Organizing Under Neutrality and Card Check Agreements, pp. 45-
48, 51-52.
Because the Employee Free Choice Act may increase the number of unionized workers, a report by the Institute for
Americas future argues that, if the legislation were enacted, 3.5 million more persons would have health insurance and
2.8 million more people would have pensions. Alex Carter and Eric Lotke, The Employee Free Choice Act: Impact on
Health Care and Pension Benefits, Institute for America’s Future, April 2007, available at [cdncon.vo.llnwd.net/o2/
(continued...)





The union success rate may be higher under card check recognition because, in part, employers
have less of an opportunity to campaign against unionization. Unions may initiate more
organizing drives under card check recognition because a card check campaign costs less than a
secret ballot election. A secret ballot election may take longer than a card check campaign and 79
employer opposition may be greater (requiring a union to expend more resources). Unions may
have a higher success rate when card check recognition is combined with a neutrality agreement
because there may be less employer opposition to unionization under a neutrality agreement.
(Some research has concluded that management opposition is a key factor affecting union success 80
rates in NLRB conducted elections.)
Finally, two surveys provide added information about secret ballot elections and card check 81
recognition. According to a March 2006 survey conducted for the Center for Union Facts (a
business group), 75% of 1,000 persons surveyed said that they believe that a secret ballot election
is the most fair and democratic way for employees to decide whether or not to join a union. By
contrast, 12% of respondents said that card check recognition is the most fair and democratic way 82
to form a union. According to a 2005 survey conducted by American Rights at Work (a labor
group), 22% of 430 workers who had gone through a union organizing campaign said that they
experienced a “great deal” of pressure from management. By contrast, 6% of workers said that
they experienced a great deal of union pressure. Among workers who signed authorization cards
in the presence of a union organizer, 5% said that the presence of the organizer made them feel 83
pressure to sign the cards.
Requiring card check certification if a majority of employees sign authorization cards may
increase the union success rate. Whether or not requiring card check certification would reverse
the decline in private sector unionization in the United States is not certain. Shrinking
employment in unionized firms and decertifications may offset any increase in union membership
due to requiring card check recognition. In addition, requiring card check recognition may
increase employer opposition during the collection of authorization cards.

(...continued)
fotf/EFCA/UnitedStatesofAmerica.pdf].
79 Robert J. Flanagan, “Has Management Strangled U.S. Unions?, Journal of Labor Research, vol. 26, winter 2005, p.
51.
80 Richard B. Freeman and Morris M. Kleiner, “Employer Behavior in the Face of Union Organizing Drives,
Industrial and Labor Relations Review, vol. 43, April 1990, p. 351.
81 For information on the two surveys, see Bureau of National Affairs, Two Surveys Reach Different Conclusions on
Benefits of Card Checks, NLRB Elections, no. 55, March 22, 2006, p. A-5.
82 The survey was conducted by the Opinion Research Corporation (a social and marketing research firm). Center for
Union Facts, Nationwide Poll: Americans Overwhelmingly Prefer Secret Ballot Elections to Card Checks, March 21,
2006, available at http://www.unionfacts.com/news.cfm?id=13.
83 The survey was prepared by two university professors and conducted by the Eagleton Research Center at Rutgers
University. American Rights at Work, Fact Over Fiction: Opposition to Card Check Doesnt Add Up, March 2006,
available at http://www.americanrightsatwork.org.








The NLRA gives private sector workers the right to organize and bargain collectively over wages,
hours, and other working conditions. It also requires employers to bargain in good faith with a
union chosen by a majority of employees. The act says that the purpose of the law is to improve
the bargaining power of workers. This section considers whether there is an economic rationale
for protecting the rights of workers to organize and bargain collectively.
Governments may intervene in labor markets for a number of reasons. One of these reasons is to 84
improve competition. According to standard economic theory, competitive markets generally
result in the most efficient allocation of resources, where resources consist of individuals with
different skills, capital goods (i.e., buildings and equipment and associated technology), and
natural resources. In turn, an efficient allocation of resources generally results in greater total
output and consumer satisfaction.
In competitive labor markets workers are paid according to the value of their contribution to
output. Under perfect competition, wages include compensation for unfavorable working
conditions. The latter theory, called the “theory of compensating wage differentials,” recognizes
that individuals differ in their preferences or tolerance for different working conditions—such as 85
health and safety conditions, hours worked, holidays and annual leave, and job security.
If labor markets do not fit the model of perfect competition, increasing the bargaining power of
workers may raise wages, improve benefits (e.g., for health care and retirement), and improve
working conditions to levels that would exist under competitive conditions. In labor markets
where a firm is the only employer (called a monopsony) unionization could, within limits, 86
increase both wages and employment.

84 The following conditions are the general characteristics of a competitive labor market: (1) There are many employers
and many workers. Each employer is small relative to the size of the market. (2) Employers and workers are free to
enter or leave a labor market and can move freely from one market to another. (3) Employers do not organize to lower
wages and workers do not organize to raise wages. Governments do not intervene in labor markets to regulate wages.
(4) Employers and workers have equal access to labor market information. (5) Employers do not prefer one worker
over another equally qualified worker. Workers do not prefer one employer over another employer who pays the same
wage for the same kind of work. (6) Employers seek to maximize profits; workers seek to maximize satisfaction. Lloyd th
G. Reynolds, Stanley H. Masters, and Colletta H. Moser, Labor Economics and Labor Relations, 11 ed. (Englewood
Cliffs, N.J.: Prentice-Hall, 1998), pp. 16-21.
85 Randall K. Filer, Daniel S. Hamermesh, and Albert E. Rees, The Economics of Work and Pay, 6th ed., New York:
Harper Collins, 1996, pp. 376-390. Ronald G. Ehrenberg and Robert S. Smith, Modern Labor Economics: Theory and th
Public Policy, 7 ed. (Reading, MA: Addison-Wesley, 2000), pp. 251-259. (Hereinafter cited as Ehrenberg and Smith,
Modern Labor Economics.)
86 Bruce E. Kaufman, The Economics of Labor Markets, 4th ed. (Fort Worth: Dryden Press, 1994), pp. 277-280.
(Hereinafter cited as Kaufman, The Economics of Labor Markets.)





On the other hand, increasing the bargaining power of employees in competitive labor markets
may result in a misallocation of resources—and reduce total economic output and consumer
satisfaction. In competitive labor markets, higher union wages may reduce employment for union 87
workers below the levels that would exist in the absence of unionization. If unions lower
employment in the unionized sector, they may increase the supply of workers to employers in the 88
nonunion sector, lowering the relative wages of nonunion workers.
It is difficult, however, to determine the competitiveness of labor markets. First, identifying the
appropriate labor market may be difficult. Labor markets can be local (e.g., for unskilled labor),
regional, national, or international (e.g., for managerial and professional workers). Second,
measuring the competitiveness of labor markets is difficult. Finally, labor markets may change 89
over time because of demographic, economic, technological, or other changes.
A second reason governments may intervene in labor markets is to reduce earnings inequality.90
Competitive labor markets may allocate resources efficiently, but they may result in a distribution
of earnings that some policymakers find unacceptable. Unionization may be a means of reducing
earnings inequality. Some economists argue that, during a recession, greater earnings equality
may increase aggregate demand and, therefore, reduce unemployment.
Finally, some economists maintain that unions give workers a “voice” in the workplace.
According to this argument, unions provide workers an additional way to communicate with
management. For instance, instead of expressing their dissatisfaction with an employer by
quitting, workers can use dispute resolution or formal grievance procedures to resolve issues 91
relating to pay, working conditions, or other matters.

87 In competitive labor markets, unions can offset the employment effect of higher wages by trying to persuade
consumers to buy union-made goods (e.g., campaigns to “look for the union label”), limiting competition from foreign
made goods (e.g., though tariffs or import quotas), or negotiating contracts that require more workers than would
otherwise be needed. Kaufman, The Economics of Labor Markets, pp. 276-277. Ehrenberg and Smith, Modern Labor
Economics, p. 493. Toke Aidt and Zafiris Tzannatos, Unions and Collective Bargaining: Economic Effects in a Global
Environment (Washington: The World Bank, 2002), p. 27.
88 If unions raise the wages of union workers and lower employment in the union sector, the supply of workers
available to nonunion employers may increase, resulting in greater competition for jobs and lower wages for nonunion
workers (thespillover” effect). On the other hand, nonunion employers, in order to discourage workers from
unionizing, may pay higher wages (thethreat effect). Ehrenberg and Smith, Modern Labor Economics, pp. 504-508.
89 Kaufman argues that labor markets in the United States have become more competitive since World War II. Bruce E.
Kaufman, “Labors Inequality of Bargaining Power: Changes over Time and Implications for Public Policy, Journal
of Labor Research, vol. 10, summer 1989, pp. 292-293.
90 Governments may also intervene in private markets to produce “public” goods (e.g., national defense) or correct
instances where the market price of a good does not fully reflect its social costs or benefitscalled, respectively,
negative and positiveexternalities.” Air and water pollution are frequently cited as examples of negative externalities;
home maintenance and improvements are often cited as examples of positive externalities.
91 Richard B. Freeman and James L. Medoff, “The Two Faces of Unionism,” Public Interest, no. 57, fall 1979, pp. 70-
73. Richard B. Freeman, “The Exit-Voice Tradeoff in the Labor Market: Unionism, Job Tenure, Quits, and
Separations,” Quarterly Journal of Economics, vol. 94, June 1980, pp. 644-645.





The economic impact of requiring card check certification or secret ballot elections may rest on
the desired objectives of policymakers.
By bargaining collectively, unionized workers may obtain higher wages, improved benefits, and 92
better working conditions than if each worker bargained individually. But, depending on how
well labor markets fit the model of perfect competition, collective bargaining may improve or
harm the allocation of resources (i.e., economic efficiency). If labor markets are competitive,
increasing the bargaining power of workers may reduce economic output and consumer
satisfaction, but may increase equality. On the other hand, if labor markets are not competitive,
increasing the bargaining power of workers may improve the allocation of resources as well as 93
increase equality.
By requiring card check certification, the number of organizing campaigns and the union success
rate may increase. Conversely, by requiring secret ballot elections, the number of organizing
drives and the union success rate may decline. Thus, compared with existing recognition
procedures, requiring secret ballot elections may lower the level of unionization, whereas
requiring card check certification may raise it. Accordingly, depending on the competitiveness of
labor markets, requiring card check certification may either improve or harm economic
efficiency. Similarly, requiring secret ballot elections may either improve or harm efficiency. If
either change were enacted, it may be difficult, however, to predict or measure the size of the
effects.
Regardless of the competitiveness of labor markets, requiring secret ballot elections may increase
earnings inequality—if fewer workers are unionized. Requiring card check certification may
reduce inequality—if more workers are unionized. Again, the size of the effects may be difficult
to predict or measure.
Gerald Mayer
Analyst in Labor Policy
gmayer@crs.loc.gov, 7-7815




92 Bargaining between employers and workers includes the right of workers to strike (in the private sector) and the right
of employers to lock out employees.
93 The results of research on the wage differential between union and nonunion workers vary. But, in general, most
studies find that, after controlling for individual, job, and labor market characteristics, the wages of union workers are
in the range of 10% to 30% higher than the wages of nonunion workers. Although the evidence is not conclusive, some
studies have concluded that unions reduce earnings inequality in the overall economy. CRS Report RL32553, Union
Membership Trends in the United States, by Gerald Mayer.