Is the NLRB Still Relevant? A Performance Review M I C H A E L H. LEROY

University of Illinois, Champaign, IL 61820 I. Introduction The National Labor Relations Board (NLRB or Board) is often perceived as irrelevant. As unions continue a prolonged decline, so does the Board. Its decisions, once prominently reported in newspapers and business periodicals, are mostly unnoticed. When the agency makes news, it is often chastised for living in the past. This occurred after the Board ruled that a nonunion employee participation group violated Section 8(a)(2) of the National Labor Relations Act. The Board's decline is reflected in two academies that are its intellectual consumers. Beginning in the 1940s, industrial relations focused heavily on the study of labor-management relations and, by extension, the Board's labor law policies. This emphasis changed in the 1980s and 1990s: "In universities ranging from M.I.T. to Chicago to UCLA, a basic course in 'labor' or 'industrial relations'.., has become an HRM course" (Lewin, p. 486). Law schools devote less attention to collective bargaining because law firms with traditional union-management practices have diversified to survive. A recent American Bar Association publication noted that the "real action in the workplace in the past two decades has been in employment litigation, not union negotiation" (Carter, p. 58). But recent events suggest that reports of the NLRB's demise are premature. New rulings reflect the Board's more activist bent. Few employers (including those who are nonunion) or unions would consider the following decisions irrelevant: 9 A recent decision [M.B. Sturgis (2000)] reversed thirty years of precedent by holding that temporary employees are eligible for inclusion in bargaining units. This has profound implications for the ubiquitous HR practice of minimizing core employment by strategically subcontracting with temporary worker agencies. 9 Another recent decision [Epilepsy Foundation of Northeast Ohio (2000)] extended the right to have a co-worker or representative present in a disciplinary meeting to nonunion employers and employees. This unprecedented ruling has potential to undercut a cornerstone legal doctrine, employment-at-will, that underpins highly mobile and flexible labor markets in the U.S. 9 A 1995 ruling (O.E. Butterfield) put unsuspecting employers in a new kind of trap when a union goes on strike. Under well-settled law, an employer has a right to JOURNAL OF LABOR R E S E A R C H Volume XXII, Number 4

Fall 2001

782

JOURNAL OF LABOR RESEARCH

hire permanent striker replacements. A 1983 Supreme Court ruling (Belknap v. Hale, 1983) appeared to mitigate that right by holding that when an employer settles a strike with a union by agreeing to discharge replacements and immediately reinstate strikers, the replacement workers may sue for breach of an implied or actual contract of permanent employment. Employers responded to this ruling by making their promises of permanent employment to replacements more vague. In O.E. Butterfield, the Board held that unless an employer explicitly hires striker replacements on a permanent basis, the employer is assumed to have hired only temporary replacements. This has great potential for minimizing strike costs for unions. The trap for employers is to avoid making enforceable promises that result in contracts of permanent employment but by eluding this problem, employers are at the mercy of a union that wants its striking members reinstated at anytime. II. Research Questions and Methods As the twenty-first century begins, there is some irony in the fact that the Board has reached its 65th anniversary. This is, of course, the traditional retirement age. Although government agencies do not stop working, NLRB critics have described the Board in geriatric terms. In their minds, the Board is ill-suited for the New Age economy. The Board's official celebration of its fiftieth anniversary prompted an outpouring of academic reflection on the relevance of the Board and the law it administers. This literature provides a helpful set of research questions for a current assessment. William Gould IV (1985), who later became Chairman of the NLRB, wondered aloud if the Board and the NLRA should be retired. His query grew out of an analysis of Supreme Court decisions that showed declining support for the NLRB and unions. His survey of NLRB cases by the Supreme Court from 1935 to September 1953 showed that of 79 decisions favoring either labor or management, pro-union positions prevailed in 60 opinions (70 percent). In the Warren Court's 55 decisions that favored either labor or management, 44 favored labor (80 percent). But in the Burger Court's 53 opinions that favored either labor or management, unions prevailed in only 31 (58 percent). Gould thought it was possible that the burdens of labor law outweighed the benefits to unions, but he ultimately concluded that "the Board needs to be changed much more than the law - - although the law ought to be strengthened" (p. 941). Gross (1985) offered this indispensable perspective: "The NLRB has never known peace in its time" (p. 8). It has been subjected to unrelenting political pressure from Congress, an intractable problem rooted in the text and structure of the NLRA. The Board cannot possibly reconcile the conflicting dictates of collective and individual employment rights. The best the Board can do is to maintain its judicial impartiality while responding to all these pressures. His overriding point was that the Board must serve a lawmaking role because of the ambiguities and conflicts built into the NLRA. Gould's and Gross's views represent this most recent wave of retrospective NLRB assessments. One additional view from this period also provides useful metrics of Board

MICHAEL H. LEROY

783

performance. Weiler (1983) attributed the persistent decline in union membership to coercive resistance offered by employers. He suggested that the NLRA's representation process fostered this behavior. Thus, he advocated reform of the NLRB's regulatory framework - - specifically, speedier and stiffer sanctions and the kind of "instant election" that Canadian labor law provides. These studies organized their evaluation of the Board's performance around a common question: How well does the NLRB balance its lawmaking role with its duty to administer the NLRA? To answer this question, I examine the Board's recent interaction with Congress, the federal courts, and the presidency. As discussed below, the reaction of those institutions to the Board reflects the agency's performance. This method also provides useful information about the Board's ability to exercise interpretive control over the NLRA. Often, that exercise is constrained when a Board decision offends one of these supervening institutions. In short, I define performance in terms of the Board's ability to implement its view of collective bargaining rights and duties without being vetoed. In addition, I examine a more practical aspect of Board performance: measures of its use by unions and employers, speed in handling unfair labor practice (ULP) complaints, imposition of remedies for violations of the law, and the like. III.

Congress

Congressional behavior is one barometer of NLRB performance. For the most part, Congress devotes little attention to the Board. However, at various times since 1935, congressional interest has been piqued by NLRB rulings or enforcement practices. Congress exercises three main checks on the NLRB: (1) Board members, including the Chairman, and also the General Counsel, are subject to Senate confirmation; (2) Congress appropriates funding for the agency; and (3) Congress has legislative power to override or change a Board decision or action. Congress has rarely used any of these powers, except for routinely approving annual appropriations. When the Board has been challenged, Congress and the Presidency have been controlled by different political parties. The most pronounced response has occurred when Republicans controlled Congress. This conflict first surfaced after Republicans gained control of the Congress in the 1946 elections. Distressed by what they perceived as the Board's pro-labor bias, they legislated a structural reform of the Board by separating the agency's enforcement and adjudication activities, a reform that occurred over President Truman's veto. More recently, conflict between the president and Congress spilled over to the Board during the Reagan administration. Although Reagan was elected with a conservative mandate, he dealt with a Democratic House of Representatives throughout his eight-year tenure. Congressional Democrats had two general complaints about the Reagan Board. They condemned its reversal of pro-union policies and precedents. Some accusations went further, suggesting that the Board was rewriting the NLRA.

784

JOURNAL OF LABOR RESEARCH

As with some congressional complaints of the Board, this grievance reflected a conveniently short memory. The Board's policy on employer propaganda during representation elections went back and forth, especially in the late 1970s. After the Board settled on a policy of restricting employer speech that was arguably coercive, the Reagan Board reversed course in its much-maligned Midland National Life Insurance (1982). Democratic complaints ignored the fact that the Board usually tilts toward management or labor depending on the president's political party. In fact, Midland National Life Insurance did not make any new law, but merely reinstated a previous policy. Democrats also overlooked the fact that the Midland National Life Insurance decision was based on a highly regarded empirical study that showed that few workers were swayed by election propaganda (Getman et al., 1976). The other Democratic complaint was valid. The backlog of ULP cases grew to embarrassing levels in the mid-1980s. This resulted from the Reagan administration's torpor in filling Board-member vacancies. Working short-handed, the Board took two years or longer just to rule on a case, and as a consequence, infuriated Democrats ("Congressman Reviews Administration," 1986). A few cases languished for ridiculous lengths, prompting one federal court to blast the NLRB as the Rip Van Winkle of administrative agencies (NLRB v. Thill, 1993). The NLRB under President Bush was remarkably effective; that is, it received no meaningful criticism from Congress. The Board reduced the huge backlog left from the Reagan administration from 1,647 cases in 1984 to 472 cases in 1990 ("Selected Statements on Delays," 1990), significantly shortening the average time for case disposition. The Board found itself enmeshed in new controversies, however, after the election of President Clinton in 1992, and a Republican Congress in 1994. Unsurprisingly, the NLRB was entangled in this larger power struggle. This conflict was carried out on three fronts.

Congress Attacked Electromation. Congress expressed unusual displeasure with the NLRB by proposing the TEAM Act, a bill aiming to loosen the severe restrictions imposed on employer control of employee participation groups. This was an unusual rebuke. In contrast to other legislative efforts, for example, a bill to ban employer hiring of permanent striker replacements, the TEAM Act was a response to a single NLRB decision. It was also atypical because Republicans lawmakers challenged a decision rendered by a Republican-dominated NLRB. Electromation (1992) prompted this quarrel. Although controversial, Electromation's facts were ordinary. A president of a company with about 200 employees set up five "action committees" after business turned bad. His purpose was to get employees and managers talking about some workplace issues. He limited these to absenteeism, smoking restrictions, communication network, premium pay, and an attendance bonus. He asked employees to volunteer for these committees. When there were too few takers, he appointed some employees. Management representatives sat on all of these corn-

MICHAEL H. LEROY

785

mittees. These groups were ineffective: Only one recommended a policy, and this was rejected by the comptroller as too expensive. Two months later the company learned that the Teamsters were attempting to organize its employees. Electromation then voluntarily disbanded these groups. Notably, the company did not intimidate or harass employees. If it had, the NLRB would have found that the company violated a different part of the NLRA, Section 8(a)(1). The Teamsters lost its representation election; but not having the more conventional mechanism for appealing a loss (i.e., charging that the employer coerced or interfered with the right of some or all employees to form a union), the union's only ground for appeal was Section 8(a)(2). In so many words, the union argued that the company's creation of five action committees amounted to formation of a company union. Thus, according to the Teamsters, Electromation denied its employees a right to bargain collectively. The Board ruled in favor of the union, and as a consequence, the union was granted a second election (which it won). Republican lawmakers, led by Rep. Newt Gingrich, made Electromation a political issue. In a highly unusual move, they filed an amicus brief with the NLRB. These politicians appeared to be unsuccessful, because the NLRB ruled that Electromation's committees were unlawful. A careful reading of the Board's decision shows, however, that their brief probably had some influence. The Board explained that if an employee group confines itself to issues of product quality or production efficiency, Section 8(a)(2) is not violated, even if an employer controls such a group (Electromation, p. 998). In a concurring opinion, Member Oviatt suggested that employee involvement groups, particularly those "involving worker self-fulfillment and self-enhancement" be given wide latitude under the NLRA (Electromation, p. 1004). Although the TEAM Act failed to become law, it openly challenged the Board's decisions in Electromation and a handful of other Section 8(a)(2) cases: Unfortunately, decisions by the National Labor Relations Board have cast doubt on the legality of many forms of workplace cooperation. In particular, the Board held in the Electromation case that certain employer-employee committees violated the National Labor Relations Act's prohibition against "employer assisted" labor organizations. This ruling has had a chilling effect on the thousands of companies that have employee involvement programs (Teamwork for Employees and Managers Act of 1997, S.295, statement of Sen. James Jeffords, Chairman of the Senate Committee on Labor and Human Resources).

Congress Mounted Strong Opposition to Board Chairmen and Members Who Were Perceived As Presidential Ideologues. Since 1980, the Board has been led by two types of chairmen. In fact, it is quite possible that all four chairmen (Dotson, Stephens, Gould, and Truesdale) had or have similar voting records. But Congress has evaluated Board leadership in symbolic terms. On this dimension, chairmen have been considered either as ideologues or administrators.

786

JOURNAL OF LABOR RESEARCH

Congress treated Donald Dotson and William Gould as ideologues. When this occurred, the Board appeared to lose a degree of independence by finding its decisions and activities more closely monitored by Congress. In both cases these chairman suffered unfairly from guilt by association. For example, congressional Democrats were already angered by President Reagan's handling of the PATCO strike. Ironically, no Republican other than Reagan was identified with the administration's decision to hire 12,000 permanent replacements for PATCO strikers. Nevertheless, Donald Dotson - - who had no apparent input in that decision - - was frequently demonized by congressional Democrats for Board decisions that upheld an employer's right to hire striker replacements. This was unfair in two respects. As Board chairman, Dotson had no authority to deprive employers of their Mackay Radio right to hire these substitutes for strikers. In addition, since his chairmanship coincided with a period of unusual labor-management strife, he participated in an exceptional number of controversial striker replacement cases. The tables were turned in the 1992 national elections. Some congressional Republicans could not forget the image of candidate Bill Clinton marching arm-in-arm with striking Caterpillar workers in Peoria just before the Illinois primary. Clinton's choice for chairman was probably doomed at that point to face a contentious confirmation. But for Republicans, there was even more provocation: William Gould was the first chairman to have been on a union's payroll. Moreover, as a Stanford professor, Clinton's nominee had just published an ambitious labor policy reform book. Consequently, Gould was subjected to nearly a year of Republican objections and stalling tactics before he was narrowly confirmed by the Senate. As with the Dotson Board, the Gould Board followed a predictable and nonpolitical path in adjudicating most ULP cases. To illustrate, presented with the opportunity to reverse a lead precedent from the Dotson Board on offensive lockouts, the Board under Gould's leadership applied that decision in a high-profile labor dispute. In this case, the Board ruled that an Illinois power utility did not violate the NLRA when it locked-out hundreds of union members who were engaged in a work slowdown. Chairman Gould sounded like Chairman Dotson when he rebuked a dissenting Democrat for ignoring the well-settled law that governed this dispute: "It is well established and frequently stated, however, that the Board may not act as arbiter of the sort of weapons the parties can use in seeking to gain acceptance of their bargaining positions. Our dissenting colleague pays lip service to this principle, but her subsequent analysis betrays her allegiance" (Central Illinois Public Service Co., 1998, slip op. at 8). The strained relationship between Clinton and the 104th and 105th Congresses left another unhealthy mark on the NLRB. Clinton undermined the nomination process by using recess appointments to fill vacancies. By this convention, a president appoints a Board member while Congress is officially adjourned. While he was not the first president to engage in this practice, his reliance on this procedure left the Board without a full complement of members from 1995 through 1999 ("Clinton Names Fox," 1999). As a result, Senate input - - and the possibility of opposition - - was greatly muted.

MICHAEL H. LEROY

787

Congress Attacked the Board by Passing Legislation to Cut the Agency's Budget by 30 Percent. In the course of trying to legislate his "Contract with America," Speaker of the House Newt Gingrich often used appropriations bills to make policy (Hager, 1995). He combined funding cuts with smothering legislative language to hobble federal agencies, including the NLRB. The Board was one of Gingrich's priority targets. For months, an appropriations bill proposed a 30 percent funding cut and language to restrict the Board's power to seek injunctions in unfair labor practices cases. IV. The Judiciary Like most federal agencies, the Board's policies and decisions are reviewable by federal courts. This constrains the Board to apply congressional mandates instead of legislating new labor policies. The breadth of judicial review is much greater, however, for the Board as compared to other federal agencies because none of its orders is selfexecuting. An employer or union found by the Board to be in violation of the NLRA incurs no liability for disobeying a Board's remedial order. The only mechanism to enforce a Board order is for an interested party - - the Board, a union, or an employer to petition a federal court of appeals.

-

-

Empirical Assessments of the Relationship between the NLRB and Federal Courts. Due to the Board's lack of enforcement powers, federal courts play a greater than usual role in controlling the interpretive doctrines of this agency. But the extent to which the Board and judiciary separately define the meaning of vague standards in the NLRA is poorly understood. Current empirical research has given some focus to this picture. In analyzing 1,224 federal court decisions that reviewed NLRB orders from 1986-1993, Brudney (1996) found that the judiciary appeared to "update" the NLRA in light of fundamental changes in the employment relationship since the 1930s and 1940s. One of his main empirical findings was that courts tend to give more weight than the NLRB to individual employment rights. For example, appellate courts tended to reverse Board bargaining orders when employers withdraw recognition from unions and attempt to sever the bargaining relationship. Brudney was unsurprised by this result. These were controversies in which there was evidence of employee discontent with a union, but the Board's bargaining orders overrode employee choice of representation. In trying to deter employer misconduct while protecting employee free choice, the Board typically ordered restoration of the bargaining relationship. But courts showed remarkable indifference, in Brudney's view, to the NLRA's concern for stability of bargaining relationships. He concluded that courts have less patience than before for Board assertions grounded in precedent or legislative history. Flynn (1995) reached a contrary conclusion. Courts play too small a role in supervising the NLRB's de facto policymaking. She found that the Board's adjudicatory method hinders the ability of courts to supervise the formation of national labor policy. She noted that most Board doctrines resulted from adjudicated cases. Thus, the

788

JOURNAL OF LABOR RESEARCH

Board engaged in de facto rulemaking. While the Board's methodology succeeded in securing greater policymaking autonomy for itself, Flynn concluded that this has come at the expense of effective administration of the NLRA. A different empirical study showed the Board's de facto policymaking in a more favorable light. A survey of 565 respondents asked union members whether, and under what circumstances, they would cross their union's picket line (LeRoy, 1991). Younger, less senior, and poorer union members were most likely to cross. The study also found that a majority of lower income members would honor his or her union's picket line, but would begin to cross at an earlier time than older and better-paid members. The study concluded that employee attitudes played only a part in a striker's decision to honor his or her union's picket line. Even members who support their union might cross for purely economic reasons. This study provided empirical justification for an important change in the Board's policy concerning strike crossovers. In the 1950s and 1970s, the Board had a policy that presumed that strike crossovers did not support their union when they crossed their union's picket line (Stoner Rubber, 1959; Peoples Gas Systems, 1974). In effect, the Board permitted employers seeking to sever a bargaining relationship to count these crossovers as votes against union representation. However, in the 1940s and early 1980s, the Board used the pro-union presumption, meaning that strike crossovers were presumed to support their union (National Plastic Products Co., 1948; Pennco Inc., 1980). Eventually, the Board concluded that both presumptions were simplistic and lacked empirical justification. They replaced these presumptions with a policy that permitted employers to come forward with objective evidence that replacement workers or strike-crossovers no longer supported the incumbent union. Without this evidence, an employer could not count crossovers as employees who oppose union representation (Buckley Broadcasting, 1987). The judiciary approved this new approach. In reviewing the "no presumption" policy (NLRB v. Curtin Matheson, 1990), the Supreme Court stated that the Board's new position was entitled to "considerable deference" (494 U.S. at 786). The Court was unconcerned by the NLRB's lack of policy consistency, noting that "a Board rule is entitled to deference even if it represents a departure from the Board's prior policy" (494 U.S. at 787). Expounding on its support for the Board, the Court said: "The use by an administrative agency of the evolutional approach is particularly fitting. To hold that the Board's earlier decisions froze the development of this important aspect of the national labor law would misconceive the nature of administrative decisionmaking" (id).

Supreme Court Treatment of NLRB Decisions and Policies. Besides occasional empirical studies, there is another way to assess the relationship between the Board and federal courts. Supreme Court decisions that review NLRB doctrines provide another useful indicator of the Board's performance. Decisions in Table 1 show Supreme

MICHAEL H. LEROY

789

Table 1 U.S. S u p r e m e C o u r t A f f i r m a n c e a n d R e v e r s a l o f N L R B R u l i n g s ( 1 9 8 6 - 2 0 0 0 )

Decisions Affirming NLRB Ruling

Prevailing Party

Auciello v. NLRB, 517 U.S. 781 (1996)

Union

Holly Farms Corp. v. NLRB, 517 U.S. 392 (1996)

Union

NLRB v. Town & Country Electric, Inc., 516 U.S. 85 (1996)

Union

NLRB v. Health Care & Retirement Corp,, 511 U.S. 571 (1996)

Employer

ABF Freight Co. v. NLRB, 510 U.S. 317 (1994)

Union

American Hosp. Ass'n v. NLRB, 499 U.S. 606 (1991)

Union

NLRB v. Curtin Matheson, Inc., 494 U.S. 775 (1990)

Union

Fall River Dyeing & Finishing Corp. v. NLRB, 482 U.S. 27 (1987)

Union

Decisions Reversing NLRB Ruling

Prevailing Party

Allentown Mack Sales & Service v. NLRB, 522 U.S. 359 (1998)

Employer

Lechmere v. NLRB, 502 U.S. 527 (1992)

Employer

Litton Financial Printing v. NLRB, 501 U.S. 190 (1991)

Employer

Edward J. De Bartolo Corp. v. Fla. Gulf Coast Bldg. & Const. Trades Council, 485 U.S. 568 (1988)

Union

NLRB v. IBEW Local 340, 481 U.S. 573 (1987)

Union

NLRB v. Financial Institution Employees of America, Local 1182. 475 U.S. 192 (1986)

Union

Court affirmances and reversals of N L R B rulings. They extend G o u l d ' s 1985 survey of Supreme Court decisions that validated or overturned N L R B rulings. In the current period (1986-2000) the Supreme Court decided far fewer cases. Gould's survey showed that the Court decided about four cases per year from 1935 to 1953 (79 cases over 18 years) and about three cases per year while Earl Warren (55 decisions from 1953-1969) and Warren Burger (53 decisions from 1969-1985) were Chief Justices. In marked contrast, the Rehnquist Court has decided approximately one N L R B case per year (14 decisions from 1986-2000). The other difference is that the union win rate increased in the current period to 71 percent (compared to 58 percent during the Burger Court). W h a t do the current statistics mean? The most c o m p e l l i n g inference is that the Board and N L R A are less relevant. There is a g o o d contrary argument. One w o u l d expect the Court to review fewer NLRB rulings as the Board acts on the full range o f ULPs that arise under the law. Since fewer novel issues occur now, Board doctrines

790

JOURNAL OF LABOR RESEARCH

should be more established. Policies that conflict with the NLRA should be weeded out by federal courts and finally the Supreme Court, Ultimately, this should reduce the number of Board cases requiring the Court's attention. This sounds persuasive, but is flawed. First, the very pronounced and abrupt decline in recent Supreme Court activity is suggestive. In the preceding periods, there was only a slight decline in annual Court rulings. Even more important, the Rehnquist Court decided a large number of labor dispute cases that were outside the NLRB's unfair labor practice complaint jurisdiction. This is the real story behind the Supreme Court's reduced volume of NLRB rulings. The following cases highlight this trend: 9 Communications Workers o f America v. Beck (1988) involved a typical union security agreement, which required the tendering of union dues as a condition of employment. Section 8(a)(3) of the NLRA allows an employer and a union to enter into this kind of agreement, regardless of whether an employee actually becomes a union member. Here, the agreement required all represented employees who did not become union members to pay union agency fees in amounts equal to the dues paid by union members.

Beck sued in federal court to challenge the union's use of his dues for purposes other than collective bargaining, such as lobbying for labor legislation and funding political candidates. His lawsuit alleged a violation of the CWA's duty of fair representation and the First Amendment, in addition to a ULP under Section 8(a)(3) of the NLRA. The Court's ruling appeared to have two major impacts. First, by ruling that federal courts rather than the Board had jurisdiction of this lawsuit, the Court created a significant precedent to bypass the Board's ULP processes. This implied that the Board is not competent to adjudicate a bread-and-butter NLRA issue. Second, this ruling gave more weight to individual than collective employment rights. As a highly visible precedent, Beck opened the door to adjudicate NLRA claims that also have an individual-rights component. Thus, not only was the Board's importance diminished by its more restricted jurisdiction; so, too, was its more protective view of collective employment rights. 9 Marquez v. Screen Actors Guild (1998) validated the jurisdictional concern implicated in Beck. The Screen Actors Guild and a movie producer had a labor agreement that contained a union security clause which reproduced the main text of Section 8(a)(3). Marquez argued that the clause also should have explicitly stated that an employee can satisfy the membership condition of Section 8(a)(3) merely by paying to the union an amount equal to its initiation fees and core expenses.

Before Beck, one might have expected the NLRB to rule on this. After Beck, federal courts became the main tribunal for these membership issues. In Marquez the Court ruled that a union does not breach its duty of fair representation by negotiating a union security clause that uses Section 8(a)(3)'s language without also incorporating Beck's ruling about reducing fees or dues. 9 Fort Halifax Packing Co. v. Coyne (1988) represented a different threat to the Board's jurisdiction. Maine enacted a severance pay law, and severance pay is also a

MICHAEL H. LEROY

791

mandatory subject of bargaining under the NLRA. When the state sued on behalf of 80 laid-off workers, the company argued that the NLRA preempted this form of state regulation. By the company's reasoning, this was a benefit they could have negotiated, but failed to do so. To the contrary, the Court ruled that Maine's severance pay statute is not preempted by the NLRA. In a similar case, Building and Const. Trades Council of Metropolitan Dist. (1993) [also called the Boston Harbor case], Massachusetts passed a law that required successful contract-bidders on the $6 billion Boston harbor clean-up project to sign a pre-hire agreement with building trades unions. A group of nonunion contractors sued in federal court, claiming that the state law was preempted by the NLRA. The Court rejected this argument because the state law did not expressly violate any provision in the NLRA, nor did it attempt to regulate the free play of economic forces between employers and unions. Here again, federal courts - - and not the NLRB - - made this important determination. The common threat posed by Fort Halifax and the Boston Harbor case is that states are more free to regulate in place of substantive terms of a collective bargaining agreement. This development is ironic because declining unionization was the general impetus to these state laws. Certainly, a complex of factors have caused the decline in union representation, but among these, one can point to Board rulings and doctrines that enabled employers to defeat unions by using highly effective election propaganda. The Board also decertifies hundreds of unions every year. Some unions have countered these reversals with protective state legislation that treats the NLRB as part of their problem.

V. The NLRB As a federal agency, the NLRB compiles and publishes data on its own performance. These data are used to evaluate its request for appropriations. They also used by academics, employers, unions, and public policy groups to make one point or another about the state of private sector collective bargaining. Rarely, however, are these data ever used to assess the performance and relevance of the Board. One source, An Administrative Appraisal of the NLRB (Miller, 2000), serves this purpose. Since its discussion is organized around a wealth of primary data that measure virtually all aspects of Board performance, the book is like an encyclopedia. It is all the more valuable because its author is a former Board Chairman. Thus, it reflects his expert sensitivity for issues that even knowledgeable Board observers might overlook - - for example, how many attorneys are employed in the General Counsel's office (see Table 5, p. 139). Although an arcane statistic, it would interest a keen observer of the Board because this office has special power over the injunction provisions of the NLRA. Thus, this employment statistic is one measure of the Board's capacity to enforce the NLRA by using this extraordinary power.

792

JOURNAL OF LABOR RESEARCH

The Board's cornucopia of data is hard to summarize, but overall, suggests that the Board continues to be relevant to a significant number o f employers, unions, and employees. For example: 9 Employee interest in unions appears to be growing - - or, more unions are seeking representation elections. Either interpretation is plausible in light of data showing that the number of votes cast in representation elections is growing (167,246 in 1991; 161,221 in 1992; 177,527 in 1993; 164,044 in 1994; 169,214 in 1995; 166,873 in 1996; 187,290 in 1998; and 197,344 in 1999 [Table 2, pp. 130-31]). But that is a short-term view. This activity is sharply down from the late 1940s through late 1970s, when the annual number of votes cast in NLRB certification elections often exceeded 400,000. 9 The Board has greatly increased its use of remedial powers under the Clinton administration (backpay amounts were $34,641,876 in 1988; $57,582,311 in 1989; $44,444,004 in 1990; $53,880,884 in 1991; $50,681,827 in 1992; $53,441,668 in 1993; $79,608,425 in 1994; $60,272,725 in 1995; $73,500,784 in 1996; $79,601,595 in 1997; and $89,854,833 in 1998). On the one hand, this is clear evidence of an activist Board. Its enforcement posture has changed dramatically during the Clinton administration. On the other hand, even with two to three times the amount of backpay, these aggregate awards are much smaller than blockbuster damages in a single race- or sex-discrimination lawsuit. 9 Although there is a strong perception that the Board has improved on its ability to process appeals of administrative law judge decisions in ULP cases, the data compiled by Miller paint a more mixed picture. If the maxim "justice delayed is justice denied" is true, this delay has negative implications for the Board (314 days elapsed from ALJ decision to NLRB decision in 1990; 266 days in 1991 ; 216 days in 1992; 227 days in 1993; 241 days in 1994; 213 days in 1995; 217 days in 1996; 193 days in 1997; and 304 days in 1998). VI. Conclusions Comparisons of the NLRB to dinosaurs and other extinct species, fallen empires, and other relics are overdone. Nevertheless, the Board must confront some worrisome signs of institutional decay. On very separate fronts - - federal courts and state legislatures - - the Board's ULP jurisdiction is eroding. Even more troubling for the Board, these incursions are coming from different sources. Unions seem to be winning the battle to legislate protective labor standards when the collective bargaining process that is administered by the N L R B fails them. Individual e m p l o y e e s , and by extension certain employers, are circumventing the Board's ability to define fundamental issues of union membership. On a separate front, the Board continues to face a generally hostile Congress. W h i l e this is flattering in a perverse w a y - - if the Board were irrelevant, Congress would not bother with it - - this must exact some price in diminished resources, prestige, influence, and within the Board, morale and esprit de corps. It is fair to conclude that these threats to the Board's vitality are real. With this in mind, it is also true that the Board has lived a precarious existence for all of its 65 years.

MICHAEL H. LEROY

793

Given the fact that it administers a law that is explicitly redistributive, and does so by an adjudicatory process that intensifies conflict, this will be the case if the Board sees its 100th anniversary. This study nevertheless highlights the main issues that will revivify the Board in the coming decade. Its seminal decisions in Electromation, M.B. Sturgis, and Epilepsy Foundation of Northeast Ohio have a common and arresting theme: the NLRB's increasing regulation of the nonunion work force. By ruling that most nonunion work teams violate the NLRA, that most nonunion workers employed by temporary agencies can join a union, and that all nonunion employees are entitled to representation during a disciplinary hearing, the Board has not only avoided irrelevance. It has once again raised the hopes of unions, and presented itself as a serious concern to employers.

REFERENCES Belknap v. Hale, 463 U.S. 491 (1983). Brudney, James J. "Reflections on Group Action and the Law of the Workplace,' Texas Law Review 74 (June 1996): 1563-99. 9"A Famous Victory: Collective Bargaining Protections and the Statutory Aging Process." North Carolina Law Review 74 (April 1996): 939-1036. Buckley Broadcasting, 284 N.L.R.B. 1339 (1987). Building & Construction Trades Council of Metropolitan Dist. v. Associated Builders and Contractors of Massachusetts/Rhode Island, 507 U.S. 218 (1993). Carter, Terry. "Unions Appear Stronger Than Ever, After Having Helped Win a Multitude of Protective Statutes for Workers." ABA Journal 86 (April 2000): 55-619 Central Illinois Public Service Co., 326 N.L.R.B. No. 80 (1998). "Clinton Names Fox to Recess Appointment One Day after Her Term As Member Expired." Daily Labor Report (Dec. 20, 1999), AA-I. Communications Workers of America v. Beck, 487 U.S. 735 (1988). "Congressional Group Urges Preservation of Participation Programs in Brief to NLRB." Daily Labor Report (February 3, 1992), A-9. "Congressman Reviews Administration Handling of Labor Relations Issues." Daily Labor Report (April 14, 1986), A-I. Electromation, Inc., 309 N.L.R.B. 990 (1992)9 Epilepsy Foundation of Northeast Ohio, 331 N.L.R.B. No. 92 (2000). Flynn, Joan. "The Costs and Benefits of 'Hiding the Ball': NLRB Policymaking and the Failure of Judicial Review." Boston University Law Review 75 (March 1995): 387-446, Fort Halifax Packing Co. v. Coyne, 482 U.S, 1 (1988). Getman, Julius G., Stephen Goldberg, and Jeanne B. Herman 9 Union Representation Elections: Law and Reality. New York: Russell Sage Foundation, 1976. Gould IV, William B. "Some Reflections on Fifty Years of the National Labor Relations Act: The Need for Labor Board and Labor Law Reform." Stanford Law Review 38 (April 1986): 938-44. Gross, James. "Conflicting Statutory Purposes: Another Look at Fifty Years of NLRB Law Making" Industrial and Labor Relations Review 39 (October 1985): 7-18.

794

JOURNAL OF LABOR RESEARCH

Hager, George. "As They Cut, Appropriators Add a Stiff Dose of Policy." Congressional Quarterly Weekly Report (August 1, 1995): 2245-48. LeRoy, Michael H. "Strike Crossovers and Striker Replacements: An Empirical Test of the NLRB's NoPresumption Policy." Arizona Law Review 33 (1991): 291-335. Lewin, David. "Research Expectations of U.S. IR-HR Units in the 1990s: Challenges and Choices." Proceedings of the Forty-Third Annual Meeting, Industrial Relations Research Association, John F. Burton, ed., 1990, pp. 484-91.

Marquez v. Screen Actors Guild, 525 U.S. 33 (1998). M.B. Sturgis, 331 N.L.R.B. No. 173 (2000). Midland National Life Insurance Co., 263 N.L.R.B. 127 (1982). Miller, Edward B. An Administrative Appraisal of the NLRB. Fairfax, Va.: George Mason University, 2000.

National Plastic Products Co., 78 N.L.R.B. 699 (1948). NLRB v. Currin Matheson, 494 U.S. 775 (1990). NLRB v. Thill, 980 F.2d 1137 (7 th Cir. 1993). O.E. Butterfield, Inc., 319 N.L.R.B. 1004 (1995). Pennco, Inc., 250 N.L.R.B. 716 (1980). Peoples Gas Systems, 214 N.L.R.B. 944 (1974). "Selected Statements on Delays in NLRB Casehandling." Daily Labor Report (October 4, 1990), D-1.

Stoner Rubber, 123 N.L.R.B. 1440 (1959). "Teamwork for Employees and Managers Act of 1997, S.295." Congressional Testimony (February 12, 1997), available in WESTLAW, 1997 WL 60487 (statement of Sen. James Jeffbrds, Chairman of the Senate Committee on Labor and Human Resources). Weiler, Paul. "Promises to Keep: Securing Workers' Rights to Self-Organization under the NLRA." Harvard Law Review 96 (June 1983): 1769-827.