The Decline of Unionization in the United States

Abstract

This article examines the decline in unionization that has occurred in the United States over the past half century, focusing on the role that employer opposition to unions has played, together with relatively weak labor law. It compares the U.S. experience and labor law regime to those of Canada. It finds that, compared to their Canadian counterparts, U.S. workers have much more difficulty in exercising their right to freely join and form unions and participate in collective bargaining, in large part due to ill-restrained employer opposition.

Keywords

unionization United States Canada first contract arbitration card-check certification

Introduction

The unionization rate in the United States rose considerably after the passage of the National Labor Relations Act (NLRA) of 1935 (and was greatly aided by the industrial boom caused by World War II), but it has been on a steady downward trajectory since about 1960.

Over the years, various explanations have been put forward for this, with many focusing on structural changes to the economy or workforce, usually related to globalization or technological progress.¹ Much emphasis has been placed on the fact that the United States has moved from a manufacturing economy to a “post-industrial” one, with fewer male, blue-collar, and less-educated workers and more female, white-collar, and more-educated workers. The complementary contention is that this changing workforce has less need for unions and thus less desire to be unionized.²

Though the changes in the composition of the workforce and the economy—including the detrimental effect upon manufacturing union workers—are undoubtedly real, what is less certain is the explanatory power of these arguments, particularly with respect to the desire for unionization.

As Paul Weiler stated 30 years ago, “There is no reason to believe that white-collar and service workers are inherently unreceptive to collective bargaining” (1983, 1773, fn. 6). After pointing out the increasing unionization amongst these groups in Western Europe and in the public sector in the U.S., he went on to cite survey data from Freeman and Medoff (1984), which showed that “union organization in these sectors of the work force lags behind worker interest in unionization.” The ensuing years have further proved the point: as the composition of the workforce has changed, the composition of the unionized workforce has changed to reflect it.³

Additionally, workers today have a larger interest in unionizing than they have had in decades. After briefly bottoming out in the 1980s,⁴ desire for unionization increased slightly in the 1990s and significantly in the mid-2000s, when a majority of non-union workers said they would vote for union representation if given the opportunity (Freeman 2007, 2). Throughout this period, however, the unionization rate continued its steady decline.

Furthermore, if structural issues were a major reason for the decline in the unionization rate in the United States, we would expect to see a similar trend in Canada, the country that is arguably more like the U.S. than any other and whose economy and workforce have largely gone through similar changes.⁵ Unionization trends in the two countries, however, have differed sharply since the mid-1960s.

For nearly five decades starting from 1920, the unionization rates in the United States and Canada were fairly similar and moved more or less in step with one another. Then, in the mid-1960s, the rates began to diverge. The U.S. rate steadily decreased until it stood at just 11.8 percent in 2011, while the Canadian rate increased, held steady for a period of time, and then began a slight decline in the early 1990s. As of 2011, the Canadian rate was 29.7 percent (see Figure 1).⁶

Figure 1.

Union membership in the United States and Canada, 1920-2009.

These are overall rates, however, and can be somewhat misleading, as they tend to mask the different experiences of the public and private sectors. In both the United States and Canada, the public-sector unionization rate has held steady since at least about 1980. Over the same period, the private-sector unionization rate has declined, though more so in the United States than in Canada.

While the loss of heavily unionized manufacturing jobs has been a contributing factor in the declining private-sector unionization rate in both countries, it would be a mistake to place too much emphasis on this in and of itself because it cannot explain the inability of workers to realize their desires for unionizing the new jobs they have moved into. Instead, in both countries, the increased ability of employers to effectively oppose unionization offers a more compelling explanation. In the United States this has been a long-standing issue, while in Canada it is a relatively newer phenomenon, related in large part to a change in the way unions can be formed.

After first briefly comparing the current state of organized labor in the United States and Canada, this article then discusses how unions are formed in each country and analyzes how different methods interact with employer opposition to unionization and affect union campaign outcomes. Following this is a discussion of first contract arbitration (FCA), a policy existing across most Canadian jurisdictions that allows for a way through a bargaining impasse. Finally, some concluding remarks are offered on the current state of U.S. labor policy in light of the comparison to Canada.

A Brief Overview of Labor in the U.S. and Canada

The United States and Canada have a broadly similar industrial relations framework through which workers are able to form unions and collectively bargain with their employers. In both countries, collective bargaining typically occurs at the “bargaining unit” level—that is, between a single employer and a group of that employer’s workers who are deemed to have a similar “community of interest.” Although broader collective bargaining can occur—as demonstrated by the agreements that spanned much of the U.S. steel industry in the middle of the 20th century—this is not the norm and rather has to do with the ability of unions to force employers to negotiate in this manner.

In contrast, collective bargaining in many European countries usually occurs at a much broader level—along multiemployer, industrial, or sectoral lines.⁷ Additionally, the collective bargaining agreements negotiated by unions and employers in many of these countries will often extend to non-union employees and workplaces, creating large differences between union membership rates and union coverage rates. This is not the case in either the United States or Canada, as can be seen by the virtually identical membership and coverage rates (see Figure 2).^8,9

Figure 2.

Union membership and union coverage rates in 21 wealthy countries, 2007-2010.

Labor policy is more decentralized in Canada than in the United States and is largely a provincial matter. While Canadian federal law covers some 10 percent of workers (those employed by the federal government and such industries as banking, telecommunications, and radio and television broadcasting), the remaining 90 percent are covered by provincial or territorial law.¹⁰ In the U.S., on the other hand, the federal government has a larger role in labor policy, especially for private-sector workers, who fall under the jurisdiction of the NLRA of 1935.¹¹ However, while Canadian provincial law generally covers workers in both the public and private sectors (Block 1994, 250, fn. 1), this is not true in the United States, where public-sector labor policy falls under state jurisdiction.

While some commentators have pointed to the very high unionization rate in the public sector in Canada—and its slightly larger size relative to the United States—as a major factor explaining Canada’s higher unionization rate (Clemens, Veldhuis, and Karabegovic 2005), as shown in Table 1, the unionization rate is higher across all categories of the workforce and economy. While much lower than its public-sector unionization rate, Canada’s overall private-sector unionization rate is more than twice as high as the rate in the United States. In a similar vein, Figure 3 shows the unionization rate by U.S. state and Canadian province.

Table 1.

Union Membership Rates in the United States and Canada, by Selected Characteristics, 2011.

	United States (percent)	Canada (percent)	Difference (percentage points)
All	11.8	29.7	17.9
Gender
Male	12.4	28.2	15.8
Female	11.2	31.1	19.9
Age
16 to 24*	4.4	14.5	10.1
25 to 54	12.5	32.1	19.6
55 and over	14.5	34.3	19.8
Education
Less than High School	6.0	20.2	14.2
High School	11.1	24.8	13.7
Some College	11.8	21.7	9.9
College and Advanced	13.8	33.8	20.0
Sector
Public	37.0	71.1	34.1
Private	6.9	16.0	9.1
Selected Industries
Mining	7.2	19.3	12.1
Construction	14.0	30.3	16.3
Manufacturing	10.5	24.9	14.4
Utilities	25.0	63.8	38.8
Services	5.8	30.4	24.6
Healthcare and Social Assistance	7.5	52.7	45.2
Employment Status
Full-time	13.1	31.1	18.0
Part-time	6.4	23.6	17.2

Notes: *15 to 24 in Canada. Canadian rates are for the first half of 2011 only.

Sources: U.S.: Analysis of Current Population Survey data by the Center for Economic and Policy Research; Canada: Uppal (2011, Table 1).

Figure 3.

Union membership rates in U.S. states and Canadian provinces, 2011.

Unionization

In a market economy, it is common for businesses to be created and for existing businesses to close. Insofar as those that close are unionized and those that are created are not, there will be a downward effect on the unionization rate. Under the particular arrangement of the industrial relations systems of the United States and Canada, where all new businesses start off non-union,¹² this effect is more pronounced than in many European countries, where many new businesses will often be union from the start, due to industrial or sectoral agreements.¹³

Given this arrangement, the method by which employees form unions at their workplaces, in particular how easy or how difficult it is to form a union, has important implications for organized labor and the unionization rate of a country.

Methods of Forming Unions

In both the United States and Canada, there are three methods by which a union can be formed in a workplace: recognition, imposition, and certification.¹⁴ In recognition, employers can voluntarily recognize unions—because they accept that a majority of their employees want a union, to block more militant unions from organizing, or from submission to employee or community pressure.¹⁵ Though there are no official data on the number of employees organized in this way, evidence suggests this method is used by a growing portion of unionization campaigns.¹⁶ Imposition, a rare occurrence in both the United States and Canada, occurs when the labor board determines that employer actions have created such an atmosphere as to deny employees a free and fair choice, certifies the union without a vote, and orders the employer to the bargaining table.¹⁷

Finally, there is certification, which as Freeman (1987) and Ferguson (2008) note, has been the predominant method of forming unions and is the way workers can attempt to unionize in the face of employer opposition. Thus the following discussion will center on the certification processes in the two countries. Generally speaking, certification involves a labor board verifying that a majority of workers at a workplace have chosen to form a union for purposes of collective bargaining. This is typically done against the wishes of employers, who would prefer to operate in a non-union environment.

Certification in the United States

The National Labor Relations Board (NLRB) administers the provisions in the NLRA—which covers most private-sector workers—including those for union certification. The NLRB conducts elections to do this.¹⁸ In order for the NLRB to hold an election, workers or their representatives must file a petition with the NLRB showing support, by signed authorization cards, of at least 30 percent of the employees in a proposed bargaining unit, although usually petitions won’t be filed unless there is upwards of 65 percent support (Bronfenbrenner 1994, 78). Typically an existing union, which the employees have worked with during the organizing drive and whose ranks they will join if their drive is successful, files the petition to the NLRB. The labor board will then verify the signed cards against the employer’s payroll. At this point, the employer may voluntarily recognize the union if a majority has signed cards. If not, the NLRB will schedule and hold an election. If a majority of employees who participate in the election vote to unionize, the NLRB certifies their union and contract negotiations may begin. If the majority votes against unionization, no union is formed, and another petition cannot be filed for one year.¹⁹

Public-sector employees make up approximately 15 percent of employees in the United States. Just as there is no nationwide law that governs their collective bargaining rights, there is no nationwide method of certification. Some states, such as Illinois, New Jersey, Oregon, New Hampshire, California, and Massachusetts, have card-check authorization (see below) (Malin 2012, 152), while others have elections similar to those conducted by the NLRB.

Certification in Canada

Like almost all labor law in Canada, the union certification process varies by jurisdiction and generally applies to both private- and public-sector workers.²⁰ Six provinces currently have mandatory elections, covering about 75 percent of the workforce (Riddell 2004, 494), while the remaining four provinces and the federal jurisdiction have card-check authorization.

The mandatory election process in Canada is very similar to that in the United States: a labor board must be presented with proof of employee support for unionization, ranging from 35 to 45 percent. Like in the U.S., though, petitions for elections aren’t typically filed without strong majority support (Slinn 2004, 291). The board will then schedule and hold elections and will certify a new union if the majority of voters supports unionization.

Card-check authorization (also called “automatic certification”), on the other hand, is more streamlined. Under card check, once a majority of employees sign up in support of unionization (ranging from more than 50 percent to 65 percent, depending upon the jurisdiction), a petition is made to the labor board for certification. The labor board verifies (“checks”) the signed cards against the employer’s payroll. While the employer can challenge the appropriateness of certain positions being in the bargaining unit—just as in the election process in both the United States and Canada at this stage—generally speaking, the organizing drive is over. Presuming the employees favoring unionization do in fact constitute the majority, the bargaining unit is certified, and the employees are officially unionized.

These two different certification regimes did not always co-exist in Canada. Prior to 1977, all jurisdictions had card-check authorization for organizing drives with support above a certain threshold (between 50 to 65 percent), and elections for those with support below that level (Riddell 2004, 493). This is still the way certification happens today for those jurisdictions that have card check.

In 1977, Nova Scotia became the first province to break with this tradition and begin to require elections for all union certifications where the employer doesn’t voluntarily recognize the union. In 1984, British Columbia followed suit (and subsequently reversed this decision in 1993 before again reverting to mandatory elections in 2001), then Alberta (in 1988), Newfoundland and Labrador (1994), Ontario (1995), Manitoba (1997), and Saskatchewan (2008) (Slinn 2008, 693, fn. 10). Along with Manitoba, which repealed its mandatory election legislation in 2000, the federal jurisdiction, New Brunswick, Prince Edward Island, and Quebec currently have card-check authorization. As will be discussed later, this shift in certification method has reduced union success rates in organizing campaigns.

Employer Opposition to Unionization and the Role of Delay

By adding a step to the process of unionization, mandatory-election certification has a built-in delay that card check does not. These delays can be significant in the United States. According to the NLRB, in 2011, there was a median of 38 days between a petition and an election (NLRB 2011, 35).²¹ In those provinces that have mandatory elections in Canada, it is not nearly as long: elections must generally take place within 5 to 10 days after the petition for an election is filed (so-called quick votes) (Slinn and Hurd 2009, 109).²²

This delay is important because employers don’t typically stand by idly while their workers attempt to form unions. The longer they know about a union campaign, the longer they have to counter with an anti-union campaign. Many of the tactics that employers use to oppose unionization are illegal; however, in the United States there has long been abundant opportunity and little disincentive to engage in them. Illegal tactics to oppose unionization are, therefore, a standard business practice:

Decades of experience in the United States have given rise to volumes of research and evidence showing that, under the NLRA as presently constructed, there is widespread, serious, and effective illegal conduct by employers that deprives workers of their opportunity to make a free choice about workplace representation. (Slinn and Hurd 2009, 104)²³

Employers usually don’t act alone in their opposition to unionizing efforts. In the U.S., 75 percent of the time private-sector employers hire outside experts to help them stave off organizing attempts (Bronfenbrenner 2009, 10). These experts of the “union avoidance industry” fall into four groups—consultants, law firms, industry psychologists, and strike management firms (Logan 2006, 651). While not new, “[t]he tremendous growth in the size, scope, and sophistication of the union avoidance industry since the 1970s … is a recent development—and one that has contributed significantly to the current crisis of organized labor in the United States” (Logan 2006, 652). It has grown up from a cottage industry in the 1950s to being worth several hundred million dollars today (Logan 2006, 651); its legitimacy within the industrial relations framework in the United States is unique in the developed world (Logan 2006, 652).

Overall, the employer’s expert-assisted anti-union campaign will consist of such things as one-on-one meetings with supervisors, captive audience meetings, anti-union leafleting, media campaigns, surveillance of employees, promises of improvement, bribes, interrogation of employees about union activity, threats of cuts to benefits or wages, threats to close the workplace, and threats to file for bankruptcy. Furthermore, “[e]mployers penalize workers [for attempting to unionize] … by transferring them to more onerous work assignments, cutting wages or benefits, layoffs, contracting out, and, most egregiously, discharging workers or shutting down, contracting out, or outsourcing all or part of the facility” (Bronfenbrenner and Warren 2011, 4).

While opposition to unionization takes many forms, they all have the same aim—to influence employees to not sign cards in favor of unionization or to vote against unionization:

Typically, the firm will mount a vigorous campaign to fend off the threat of collective bargaining. It will emphasize to its workers how risky and troubled life might be in the uncharted world of collective bargaining: the firm might have to tighten up its supervisory and personnel practices and reconsider existing, expensive special benefits; the union would likely demand hefty dues, fines, and assessments, and might take the employees out on a long and costly strike with no guarantee that there would be jobs at the end if replacements had been hired in the meantime; if labor costs and labor unrest became too great, the employer might have to relocate. (Weiler 1983, 1777-8)

Employer opposition to unionization in the United States has increased not only in frequency but also in intensity.²⁴ Noting the changes in employer anti-union campaigns from 1986 to 2003, Bronfenbrenner (2009, 14) commented that “It seems that most employers feel less need to bother with the carrot and instead are going straight for the stick.” Employers are less and less often offering bribes, improvements to working conditions, and the like and are focusing instead on more aggressive tactics such as threatening to close workplaces (29 percent did so in 1986-1987 compared to 57 percent in 1999-2003). Additionally, in the earlier period, no employers used more than 10 different anti-union tactics in their campaigns, while 49 percent did in the later one (Bronfenbrenner 2009, 13).

Firing workers for union activity is perhaps the biggest stick that employers have. Schmitt and Zipperer (2009, 3) estimated that workers were illegally fired in approximately 30 percent of certification elections in the United States in 2007, above the average of 26 percent from 2001 to 2007, and up from an average of 16 percent from 1996 to 2000. Over the whole period from 1951 to 2007, the only time this figure was higher than in 2007 was from 1981 to 1985, when illegal firings occurred in 31 percent of unionization campaigns (Schmitt and Zipperer, 2009, 3). Such illegal firings can be a substantial factor in blocking unionization drives because they often remove those employees most involved in organizing efforts. However, the impact goes beyond this because cases filed with the NLRB will typically not be resolved before an election happens (if it occurs at all), creating a large chilling effect on any other workers considering unionization (Weiler 1983, 1793). This is demonstrated by the U.S. Dunlop Commission’s finding that 41 percent of non-union workers thought they might be fired if they tried to organize a union at their workplace (1994, 40).

Without the delay associated with mandatory elections, there is much less opportunity for employers to engage in the pervasive use of illegal tactics to oppose unionization. Under card check, organizing campaigns are normally completed before employers become aware of them, or they are discovered so late that opposition efforts are limited (Riddell 2004, 497). In Canada, the change from card check to mandatory elections affecting the bulk of the workforce has allowed employers more opportunity to oppose unionization, even if they still have less space to do so than their counterparts in the United States. In recent years, there has been an increased focus on this fairly new phenomenon.²⁵

In studies on British Columbia (Riddell 2004) and Ontario (Slinn 2004), which together account for about half of the Canadian labor force (Campolieti, Riddell, and Slinn 2007, 33), the evidence suggests that the delay inherent in mandatory elections had a significant, negative effect on union wins in certification campaigns, despite the short duration.²⁶ In the first period that British Columbia had mandatory elections (from 1984 to 1992), there was a 19 percentage-point reduction in private-sector certification success rates compared to the preceding period under card check. When mandatory elections were repealed in 1993 and card check was re-instituted, success rates returned to their previous levels (Riddell 2004, 495-6). Virtually the entire drop in union success rates during the mandatory election period could be explained by the switch from card check to mandatory elections, with employer opposition (measured by unfair labor practices) estimated to be twice as effective under elections as under card check, and accounting for at least 25 percent of the total reduction in certification success rates (Riddell 2004, 509).

The study on Ontario examined the outcomes of certification applications from 1993 to 1998, with roughly 3 years under card check and 3 years under mandatory elections. After the change to mandatory elections, there was a decline in overall certification success rates by 8.4 percentage points. This study concluded that, controlling for factors such as bargaining unit size, type of employment, and initial level for support, a unionization attempt under the mandatory vote regime was 21 percentage points less likely to be successful than one under card check (Slinn 2004).

Another study, looking at data for nine Canadian provinces from 1978 to 1996, found that mandatory elections reduced union success rates in certification campaigns by 9 percentage points compared to what they would have been under card check (Johnson 2002). Following up on this, and looking at the period from 1980 to 1998, further research estimated that Canada’s unionization rate would have been 3.3 to 4.6 percentage points lower by 1998 if all jurisdictions had mandatory elections from 1980, and 1.0 to 1.2 percentage points higher if all had card check from 1980 (Johnson 2004).

Success rates decrease further when the time limits to hold quick elections are exceeded (Johnson 2004, 46-49). Employer objections to the certification application or unfair labor practice filings (by employees, the union, or employer), for example, can, in the absence of expedited hearings, delay elections and lower the ultimate probability of union victory (Campolieti, Riddell, and Slinn 2007, 42-46).

Finally, further evidence that employer opposition has been a leading cause of the decline in the unionization rate in the United States can be found by comparing the experiences of the public and private sectors. Public-sector employers have less incentive to oppose unionization than private-sector employers,²⁷ and so opposition to unionization is not nearly as prevalent or intense in the public sector, “where workers organize relatively free from the kind of coercion, intimidation, and retaliation that so dominates in the private sector” (Bronfenbrenner 2009, 2).²⁸ Research shows that nearly all (96 percent) private-sector employers in the United States engaged in anti-union campaigns over the period 1999 to 2003 but only about half (52 percent) of public-sector employers did (Bronfenbrenner 2009, 23). Additionally, as mentioned above, some states have card check as the means to certification, which greatly diminishes the opportunity to oppose unionization. The result of this can be seen in the union membership rate of the two sectors: while the rate in the private sector has experienced a precipitous drop over the past 30 years, the rate in the public sector has actually increased slightly, as shown in Figure 4.

Figure 4.

Union membership rates for the public and private sectors in the United States, 1977-2011.

The United States is not alone in this situation. Though Canada’s overall unionization rate has held fairly steady over the past half century, the private-sector unionization rate has declined since the early 1980s, though not nearly as much as it has in the United States (see Figure 5). This is the same period in which some provinces—covering approximately 75 percent of the Canadian workforce—switched from card check to mandatory elections and is consistent with the idea that mandatory elections allow for more employer opposition—and thus lower union success rates—in unionization campaigns.²⁹ As Riddell (2004, 494) noted, “Many observers believe it is no coincidence that unions declined concurrently with the abandonment of the card-check system.”

Figure 5.

Union membership rates for the public and private sectors in Canada, 1981-2011.

Table 2 compares the percentage-point changes in the unionization rates by sector in the United States and Canada. In the period from 1981 to 2004, the United States’ overall rate declined by 8.9 percentage points, while it declined by 7.0 percentage points in Canada. Similarly, the private-sector unionization rate declined by 10.8 percentage points in the U.S. and 9.8 percentage points in Canada. On the other hand, the public-sector rate in the U.S. actually increased slightly over this time period, while it remained unchanged in Canada.

Table 2.

Change in Union Membership Rates, 1981-2011 (percentage points).

	1981-2004		1997-2011
	U.S.	Canada	U.S.	Canada
All	–8.9	–7.0	–2.3	–1.4
Private	–10.8	–9.8	–2.8	–5.9
Public	2.1	0.0	–0.2	–1.4

Source: See Figures 4 and 5.

Looking at the more recent period, from 1997 to 2011, we see less change in the rates. In the U.S., the overall rate decreased by 2.3 percentage points, in Canada, by 1.4 percentage points. But in contrast to the earlier period, the United States’ private-sector rate actually decreased less than Canada’s, in fact by only half as much (2.8 compared to 5.9 percentage points). Finally, the public-sector rate in the U.S., rather than increasing during this period as it did from 1981 to 2004, was essentially unchanged, while Canada’s decreased by 1.4 percentage points.

While Canada, in the later period, had a greater percentage-point decline in its private-sector unionization rate, Table 3 shows that, in proportional terms, the U.S. actually had larger declines in both periods. From 1981 to 2004, the private-sector rate decreased by 57.8 percent in the U.S. and 32.9 percent in Canada. From 1997 to 2011, this decrease slowed significantly in the U.S. but only slightly in Canada. Overall, the unionization rate in the U.S. declined by 41.6 percent from 1981 to 2004 and by 16.3 percent from 1997 to 2011, and in Canada it declined by 18.6 percent from 1981-2004 and by 4.5 percent from 1997 to 2011. Again, the public-sector unionization rate in both countries changed little.

Table 3.

Change in Union Membership Rates, 1981-2011 (percent).

	1981-2004		1997-2011
	U.S.	Canada	U.S.	Canada
Overall	–41.6	–18.6	–16.3	–4.5
Private	–57.8	–32.9	–28.9	–26.9
Public	6.1	0.0	–0.5	–1.9

Source: See Figures 4 and 5.

While the relatively stable public-sector rates slightly mask the overall decline in unionization rates in both countries, the decline in the private sector is certainly driving it.

Criticism of Card Check

Employers in the United States have two main arguments against card check, both stemming from their supposed concern over employees’ best interests.

The first is that some employees would be pressured or deceived into signing authorization cards by pro-union employees or union organizers, and a secret-ballot election would give them the anonymity required to express their true opinions. This ignores the fact that union intimidation would be an unfair labor practice, as it currently is in Canada and that, in Canada, complaints of such activity have been “rare” and findings that it actually occurred have been “rarer” (Godard, Rose, and Slinn 2009, 117). Furthermore, survey data from workers in the United States who have gone through both election and card-check campaigns show similar results, with the researchers noting that “management pressure on workers to oppose unionization was considerably greater than pressure from co-workers or organizers to support the union in both card checks and elections” (Eaton and Kriesky 2009, 157).

The other common criticism of card check is that employees cannot make a fully informed choice when unions are formed by this method. They argue that unions would have an unfair advantage in convincing employees that unionization is in their best interest. By requiring first a petition for an election and then an election (typically delayed as much as possible by employers), employers argue they would have enough time to present employees with their position—in other words, to run an anti-union campaign.³⁰

If this were true, one would expect to see a large number of decertifications³¹ during card-check regimes. If employees were duped by their fellow workers or union organizers into signing authorization cards, presumably they would realize this fairly quickly and would want to decertify the union in order to continue the employment relationship they had prior to unionization. However, data from British Columbia and Ontario—again, the two provinces containing more than half of the Canadian workforce and where both card check and mandatory election systems have existed—do not show any indication of this. Figure 6 shows decertifications as a percent of the prior year’s certifications in both provinces, under both systems. Contrary to what we would find if employers’ criticism were correct, the level of decertifications was almost always higher during the periods of mandatory elections, not during the periods of card check. In British Columbia, the percentage of decertifications to the prior year’s certifications averaged 46.1 percent overall, 32.0 percent during card check, and 53.5 percent during mandatory elections. Ontario had much lower levels of decertifications to certifications but shows a similar trend. The percentage of decertifications to the prior year’s certifications averaged 13.1 percent overall, 11.0 percent during card check, and 15.4 percent during mandatory elections. Similar figures result whether the lag (in this case, one year) is two years, three years, or more.

Figure 6.

Decertifications as a percent of prior year’s certifications, British Columbia and Ontario, 1981-2011.

Reducing Opportunity for Employer Opposition

Nearly 30 years ago, Paul Weiler, in a seminal article (1983) on the subject, argued that the incentives for employers to break the law were too great, and rather than combating these with increased penalties, reducing the opportunities for employer to engage in illegal activity was the appropriate approach. He argued for what he called “instant elections,” based off the practice in Nova Scotia, which in 1977 became the first Canadian jurisdiction to have mandatory elections held shortly after filing a petition showing support for unionization.

The research presented here lends credence to the view that decreasing the opportunities to commit unfair labor practices will be more effective than increasing disincentives.³² However, the evidence of employer opposition in Canada suggests that even “instant” elections are not quick enough. Given the opportunity to engage in illegal activity that elections provide (through delays in the process), employers have shown that they will take it. Card-check authorization may be the best possible way to curb illegal employer opposition in the unionization process in the United States. The Employee Free Choice Act (EFCA)—organized labor’s unfulfilled legislative hope for the Obama administration—would have done this.

First Contract Arbitration

A union is not fully effective unless it is able to negotiate a collective bargaining agreement with the employer of its members. A collective bargaining agreement—or contract—is a legally enforceable document that spells out such bread-and-butter issues as wages, raises, cost of living adjustments, health insurance benefits, and paid time off as well as such non-monetary issues as grievance and disciplinary procedures, scheduling policies, work rules, performance reviews, and a host of other things, many of which can be unique to particular industries or individual workplaces.

In the United States, a large percentage of workers who organize unions at their workplace are unable to obtain a first contract after their union has been certified. Recent research has found that just under half of newly certified unions in the U.S. are unable to negotiate a contract two years after certification.³³ Previous research found higher success rates in the past, indicating that unionized workers are having an increasingly difficult time reaching the ultimate goal of collective bargaining, as shown in Figure 7.³⁴

Figure 7.

First contract success rates in the United States, 1955-2004.

Part of the decrease in contract success rates is because negotiating a first contract can be quite difficult:

First, the magnitude of the task is large—the parties are negotiating and codifying all the terms and conditions of employment, not simply modifying an existing collective agreement. Second, this task occurs in the context of a bargaining relationship that is immature. Third, the negotiators are often inexperienced. Fourth, these problems may be compounded by hostility that developed during organizing that spills over into negotiations. Fifth, bargaining unit members may have unrealistic expectations concerning what the union will be able to achieve in a first agreement. Finally, and perhaps most importantly, even after a union has been certified, an employer determined to prevent unionization may use various strategies, both legal and illegal, at the bargaining table to oppose unionization. (Johnson 2010, 586)

The other part of the decrease, however, has to do with employer opposition to unions even after they have been certified. Despite the fact that employers and employees are mandated by the NLRA to negotiate “in good faith,” research focusing on data from 1999 to 2004 shows that the already-dismal success rates were even lower in those negotiations where unfair labor practice charges were filed (Ferguson 2008, 6).³⁵ Much of this can be explained by employers having little incentive to negotiate in good faith: if they are found not to be doing so, they are simply ordered back to the bargaining table.³⁶

Canada provides an example of one possible solution to this particular problem, in the form of “first contract arbitration.”³⁷

First Contract Arbitration in Canada

First contract arbitration (FCA) was initially introduced in Canada in 1974, as a way to prevent disruptive work stoppages resulting from first contract negotiations³⁸ and to ensure that workers who voted to unionize were able to negotiate a contract despite continued employer opposition (Godard, Rose, and Slinn 2009). Today eight of the 14 jurisdictions in Canada have some form of FCA, covering approximately 85 percent of the workforce (Slinn and Hurd 2011, 80).³⁹

Generally speaking, either a union or an employer (or both) can apply to the appropriate authority for FCA if bargaining has come to an impasse or certain conditions have been met. Depending upon the enacting legislation, access to the FCA process can be granted automatically or after a review has found it to be merited. If the case proceeds to FCA, the first step is often conciliation or mediation to help the parties reach a voluntary agreement (if this had not already been required as one of the conditions for application). Should this fail, the last resort is arbitration, in which a neutral arbitrator or arbitration panel will hear from both sides and then impose a legally binding contract that will usually include any provisions the parties were able to agree on.

Though this is the basic outline of FCA, specific implementations vary. Researchers have identified four different FCA models that have developed in Canada: “exceptional remedy,” “no-fault,” “automatic access,” and “mediation intensive” (Slinn and Hurd 2011, 41).⁴⁰

The exceptional remedy, or “fault,” model was the first type of FCA introduced in Canada, in British Columbia in 1974. It was subsequently replaced with the mediation intensive model in that province in 1993, though Quebec (since 1978), the federal jurisdiction (also since 1978), and Newfoundland and Labrador (since 1985) currently have it. It is the most restrictive form of FCA and requires there to be demonstrated bad-faith bargaining or an irreparable breakdown in negotiations. In the federal jurisdiction and in Newfoundland and Labrador, there is a double screening process in order for applications to FCA to be accepted, with applications first being made to the Minister of Labour, who then decides whether or not to refer the matter to the labor board, which then decides whether or not the application has merit and if arbitration is appropriate. In Quebec, the application is made to the Minster of Labour, who then decides whether or not to send it to arbitration. In these cases, the arbitrator first acts as a mediator and only arbitrates if this is unsuccessful (Murray and Cuillerier 2009, 89).

The no-fault model is less restrictive than exceptional remedy and has been in place in Ontario since 1986 and in Saskatchewan since 1994. There is no double screening process, and rather than having to prove bad-faith bargaining or a bargaining impasse, negotiations must be found to be unsuccessful for one of several statutory reasons. In Ontario, the labor board cannot decline to direct the parties to arbitration if one of these has been proven, whereas the board may refuse to intervene in Saskatchewan. In Saskatchewan, at least 90 days must have passed since certification of the union before arbitration can begin and the labor board can first direct the parties to conciliation if they have not already gone through such a process.

The automatic access model is less restrictive than either the exceptional remedy or no fault models and is available in two Canadian provinces—Manitoba and Nova Scotia. Manitoba originally had the exceptional remedy model for two years before abandoning it, in 1984, for this form, which had been the first and only of its kind until Nova Scotia passed similar legislation at the end of 2011.⁴¹ There is no discretionary screening process; two conditions must be met: that a certain amount of time has passed since certification of the union and that the parties were unable to reach an agreement even after conciliation. If the union and the employer cannot agree on an arbitrator, the labor board has the option of requiring that the two parties continue negotiations, perhaps with the guidance of a conciliator, but must ultimately impose a contract if they fail.

The mediation intensive model is also less restrictive than either the exceptional remedy or no fault models and only exists in British Columbia (since 1993). It is not conditional on the bargaining atmosphere as in the exceptional remedy and no fault models, and there is no time requirement as in the automatic access model. Like the automatic access model, there is no discretionary screening process; two conditions are required for application to the labor board to be made (and granted) to the FCA process—that the parties have not reached an agreement and that a strike vote has been carried out successfully. At that time, a mediator is appointed, who has 20 days to help the parties reach an agreement. At the end of this period, the mediator must report back to the labor board and recommend terms of the agreement for the parties to consider and/or a process to conclude negotiations, which may include further mediation, arbitration, or the authorization of a work stoppage.

The Employee Free Choice Act and Criticisms of First Contract Arbitration

FCA was one of the main provisions in the unpassed EFCA in the United States. While not a perfect fit,⁴² the EFCA version (which would have applied to private-sector workers currently covered by the NLRA) came closest to the automatic access model in Canada. If no agreement had been reached 90 days after the beginning of first contract negotiations, EFCA would have allowed either the employer or the union to notify the Federal Mediation and Conciliation Service (FMCS), an independent federal agency created in 1947. The FMCS would then try to mediate an agreement between the two parties for the next 30 days. If an agreement could still not be reached, the FMCS would refer the dispute to an arbitration panel that would be tasked with deciding the terms of the first contract, which would then be imposed for a period of two years.⁴³

While it could be argued that the time periods in EFCA could be expanded or that more emphasis could be placed on mediation, these were typically not the arguments made by opponents of the legislation.⁴⁴ Critics of the FCA provision of EFCA—usually employers, employers’ associations, or employers’ lawyers—had a different set of concerns. Two of these deserve special attention.⁴⁵ The first claim is that FCA would discourage voluntary collective bargaining by making one or both parties rely on the arbitration process rather than doing the hard work of negotiating. Another concern is that it would hand the reins of private business over to government arbitrators, resulting in onerous or unworkable conditions.⁴⁶

Neither of these concerns appears to be borne out by the Canadian experience with FCA. In Canada, when FCA legislation was first being considered, employers opposed it for many of the same reasons that U.S. employers opposed its provision in EFCA (Slinn and Hurd 2011, 44). However, nearly 40 years after the passage of the first FCA law, it is no longer considered a controversial piece of Canadian labor legislation (Slinn and Hurd, 2011, 41-45). Research has even shown that employers have constituted up to one-third of all FCA applications in one Canadian province (Slinn and Hurd, 2011, 55).

A study of the jurisdictions with FCA in Canada from 1986 to 2008 found that there was an application to the FCA process for about 4 percent of all first contract negotiations. However, not all applications are granted—just 1.4 percent of all first contract negotiations are directed to FCA (Slinn and Hurd, 2011, 60-61). Many of those cases that applied to the FCA process resulted in voluntarily negotiated agreements, and this was particularly the case in British Columbia, which emphasizes mediation in the FCA process (Slinn and Hurd, 2011, 77).

Another study looked at data from the years 1976 to 2005 and also shows that FCA is rarely sought. The ratio of FCA applications to the previous year’s certifications ranged from an average of just 1 percent (in the federal jurisdiction) to 17 percent (in Manitoba) (Johnson 2010, 597).⁴⁷ It also showed the imposition of first contracts was even rarer: the ratio of imposition of first contracts to the previous year’s certifications ranged from an average of 0.3 percent (in the federal jurisdiction) to 8 percent (in Manitoba) (Johnson, 2010, 599).⁴⁸

This research also looked at work stoppage incidence in relation to first contract negotiations. It found that work stoppages were likely to be reduced “because FCA provides an incentive for the parties to reach agreement rather than risk the possibility of a third party imposing the terms and conditions of the contract.” Rather than discouraging voluntary collective bargaining, FCA in fact encourages it (Johnson, 2010, 599).

As to the second concern—that arbitrators “may impose sweeping terms and conditions of employment that will lead to disastrous results for both employees and employers”⁴⁹—research has shown that arbitrators are usually conservative in the terms and conditions they write into contracts.⁵⁰ Additionally, one study looked to Manitoba, which has the distinction of being the Canadian jurisdiction with the highest rates of application to FCA and imposition of first contracts as well as one of the jurisdictions with the automatic access FCA model—the type most similar to that proposed in EFCA. It found that the rate of failure of those businesses that had a first contract imposed by an arbitrator between 2001 and 2007 was actually lower than the overall business failure rate (Eisenbrey and Eagan-Van Meter 2010). Additionally, it is worth noting that there is nothing stopping the two parties from negotiating and modifying a contract once it has been imposed (Zack 2009, 3).

Finally, rather than “government arbitrators,” most often arbitrators are from the private sector. In both the United States and Canada, arbitrators are skilled professionals who have experience in labor-management arbitration. In the U.S., “the Federal Mediation and Conciliation Service (which is given rulemaking authority under [EFCA]) and the American Arbitration Association have many decades of experience in creating and maintaining rosters of arbitrators whom by education, experience, and the recommendation of both labor and employer professionals, have met the qualifications required to serve as expert, reliable, and neutral arbitrators” (Zack, 2009, 3). Additionally, the employer and union are usually first given the opportunity to choose an arbitrator together before one would be assigned. Or, alternatively, the parties could choose one arbitrator each to represent their interests and then jointly select a neutral one to create a three-person arbitration panel.⁵¹

Labor analysts in the United States have been arguing in favor of FCA at least as far back as the early 1990s.⁵² The chorus has only grown since that time, reaching its crescendo with EFCA. Given the barricades that employers erect in the face of employees’ wishes to collectively bargain and the benefits of FCA to the Canadian industrial relations system, it appears to be a policy that would greatly enhance labor relations in the United States by ensuring that employees who vote to unionize are able to obtain a contract.

Conclusion

This article has focused on two of the largest differences in labor policy between the United States and Canada—the union certification process and first contract arbitration. There are other policy differences that, while important, have not been discussed. For instance, some Canadian jurisdictions ban temporary or permanent striker replacement, whereas no such ban exists in the United States.⁵³ Nearly half of the U.S. states have so-called right-to-work legislation,⁵⁴ which prohibits employers and unions from including “union shop” provisions in contracts, requiring all employees covered by a union contract to pay representation fees to the union. In Canada, the situation is nearly the reverse—where such clauses are not explicitly included in contracts, legislation implicitly includes them.

Examining all of the differences between the industrial relations frameworks of the two countries would be a much larger project than that undertaken here. However, the legal process for unionization and FCA are perhaps the most important policy differences and illustrate the vast difference in labor policy between the United States and the country most similar to it. Compared to their Canadian counterparts, U.S. workers have much more difficulty exercising their right to freely join and form unions and participate in collective bargaining, in large part due to ill-restrained employer opposition.

The legislation most able to remedy this situation—the Employee Free Choice Act—may currently be politically unfeasible, but it’s difficult to argue that another policy change would make as significant a stride towards this goal. Nonetheless, there are other reform options. The biggest opposition to EFCA from the Right involved its provision for card check; trying to enact its two other provisions (FCA and unfair labor practice reforms) on their own might prove less contentious. Setting aside card check, limiting the amount of time between a petition and an election to just 5 or 10 days would be the next best way to reduce the opportunity for illegal employer opposition in unionization campaigns—in other words, bringing the Canadian “quick vote” system to the United States.⁵⁵ And reforms such as creating “a civil right to unionize”⁵⁶ or passing “just cause” legislation⁵⁷ would also be steps in the right direction.

However, with little support from the Obama administration or the Democratic Party and opposition from congressional Republicans, perhaps the best starting point is for organized labor and its allies to focus less on legislative solutions and more on movement building at the grassroots level. The possibilities are numerous but include ramping up resources put into smart and innovative unionizing campaigns, deepening involvement in community organizing, devoting more energy to educating the public about the benefits of unionization, and placing more emphasis on state and local elections. Somewhat more ambitiously, perhaps it is time once again to examine the potential for a Labor Party.⁵⁸ Such independent strategies may be the only way to begin to generate more robust protections of workers’ rights to unionize.

Footnotes

Acknowledgements

The author thanks John Schmitt, Dean Baker, Alan Barber, Nicole Woo, Jim Stanford, and an anonymous reviewer for helpful comments and suggestions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Kris Warner was until recently a research assistant at the Center for Economic and Policy Research in Washington, D.C., where a working version of this paper appeared in August 2012. He is currently a student at the University of Maryland, College Park, studying U.S. labor history.

References

Adams

Roy J.

1994. Union certification as an instrument of labor policy: A comparative perspective. In Restoring the promise of American labor law, ed. Friedman

Sheldon

Hurd

Richard W.

Oswald

Rudolph A.

Seeber

Ronald L.

, 260-72. Ithaca, NY: ILR Press.

Akyeampong

Ernest. B.

2000. Non-unionized but covered by collective agreements. Perspectives on labour and income. http://www.statcan.gc.ca/studies-etudes/75-001/archive/e-pdf/5374-eng.pdf.

Benz

Dorothee

. 1998. Scaling the wall of employer resistance: The case for card-check campaigns. New Labor Forum (Fall/Winter): 118-28.

Block

Richard N.

1994. Reforming U.S. labor law and collective bargaining: Some proposals based on the Canadian system. In Restoring the promise of American labor law, ed. Friedman

Sheldon

Hurd

Richard W.

Oswald

Rudolph A.

Seeber

Ronald L.

, 250-59. Ithaca, NY: ILR Press.

Bronfenbrenner

Kate

. 1994. Employer behavior in certification elections and first-contract campaigns: Implications for labor law reform. In Restoring the promise of American labor law, ed. Friedman

Sheldon

Hurd

Richard W.

Oswald

Rudolph A.

Seeber

Ronald L.

, 75-89. Ithaca, NY: ILR Press.

Bronfenbrenner

Kate

. 2009. No holds barred: The intensification of employer opposition to organizing. Washington, DC: Economic Policy Institute and American Rights at Work Education Fund.

Bronfenbrenner

Kate

Warren

Dorian

. 2011. The empirical case for streamlining the NLRB certification process: The role of date of unfair labor practice occurrence. New York: Columbia University, Institute for Social and Economic Research and Policy.

Brudney

James J.

2005. Contractual approaches to labor organizing: Supplanting the election paradigm? Paper presented at the 57th annual meeting of the Labor and Employment Relations Associations, Philadelphia, January 7-9.

Campolieti

Michele

Riddell

Chris

Slinn

Sara

. 2007. Certification delay under elections and card-check procedures: Empirical evidence from Canada. Industrial & Labor Relations Review 61 (1): 32-58.

10.

Card

David

Freeman

Richard B.

1993. Introduction. In Small differences that matter: Labor markets and income maintenance in Canada and the United States, ed. Card

David

Freeman

Richard B.

Cambridge, MA: National Bureau of Economic Research. http://www.nber.org/books/card93-1.

11.

Chaison

Gary N.

Rose

Joseph B.

1994. The Canadian perspective on workers’ rights to form a union and bargain collectively. In Restoring the promise of American labor law, ed. Friedman

Sheldon

Hurd

Richard W.

Oswald

Rudolph A.

Seeber

Ronald L.

, 241-49. Ithaca, NY: ILR Press.

12.

Clark

R. Theodore

Jr.

2009. Letter to Congress. February 13. http://www.uschamber.com/sites/default/files/hill-letters/090213arbitration.pdf.

13.

Clemens

Jason

Veldhuis

Niels

Karabegovic

Amela

. 2005. Explaining Canada’s high unionization rates. Canada: The Fraser Institute. http://www.fraserinstitute.org/research-news/display.aspx?id=13024.

14.

Dickens

1980. Union representation elections: Campaign and vote. PhD diss., Department of Economics, Massachusetts Institute of Technology.

15.

Duffy

Paul

Johnson

Susan

. 2009. The impact of anti-temporary replacement legislation on work stoppages: Empirical evidence from Canada. Canadian Public Policy 35 (1): 99-120.

16.

Eagan-Van Meter

Patrick

Eisenbrey

Ross

. 2009. First-contract arbitration facts: The Canadian experience. Washington, DC: Economic Policy Institute. http://www.epi.org/page/-/pdf/ib256.pdf.

17.

Eaton

Adrienne E.

Kriesky

Jill

. 2009. NLRB elections vs. card check campaigns: Results of a worker survey. Industrial & Labor Relations Review 62 (2). http://digitalcommons.ilr.cornell.edu/ilrreview/vol62/iss2/2.

18.

Eisenbrey

Ross

Eagan-Van Meter

Patrick

. 2010. Business success and first contract arbitration. Washington, DC: Economic Policy Institute. http://www.epi.org/publication/ib275/.

19.

Farber

Henry S.

2005. Union membership in the United States: The divergence between the public and private sectors. Working Paper No. 503, Princeton University, Industrial Relations Section. http://dataspace.princeton.edu/jspui/handle/88435/dsp015999n338s.

20.

Farber

Henry S.

Krueger

Alan B.

1992. Union membership in the US: The decline continues. Cambridge, MA: National Bureau of Economic Research. http://www.nber.org/papers/w4216.pdf.

21.

Ferguson

John-Paul

. 2008. The eyes of the needles: A sequential model of union organizing drives, 1999-2004. Industrial and Labor Relations Review 62 (1): 3-21.

22.

Flanagan

Robert J.

2005. Has management strangled U.S. unions? Journal of Labor Research 26 (1): 33-63.

23.

Freeman

Richard B.

1985. Why are unions faring poorly in NLRB representation elections? In Challenges and choices facing American labor, ed. Kochan

Thomas A.

, 45-64. Cambridge, MA: MIT Press.

24.

Freeman

Richard. B.

1987. Contraction and expansion: The divergence of private sector and public sector unionism in the U.S. Cambridge, MA: National Bureau of Economic Research.

25.

Freeman

Richard B.

2007. Do workers still want unions? More than ever. Washington, DC: Economic Policy Institute. http://www.sharedprosperity.org/bp182/bp182.pdf.

26.

Freeman

Richard B.

Medoff

James L.

1984. What do unions do? New York: Basic Books.

27.

Godard

John

Rose

Joseph B.

Slinn

Sara

. 2009. Should Congress pass the Employee Free Choice Act? Some neighborly advice. Just Labour: A Canadian Journal of Work and Study 15:116-20. http://www.justlabour.yorku.ca/volume15/pdfs/09_godard_rose_slinn_press.pdf.

28.

ICTWSS Database. 2011. Version 3.0. http://www.uva-aias.net/208.

29.

Johnson

Susan

. 2002. Voting or card-check: How the union recognition procedure affects organizing success. Economic Journal 112 (479): 344-61.

30.

Johnson

Susan

. 2004. The impact of mandatory votes on the Canada-U.S. union density gap: A note. Industrial Relations 43 (2): 356-63.

31.

Johnson

Susan

. 2010. First contract arbitration: Effects on bargaining and work stoppages. Industrial and Labor Relations Review 63 (4): 585-605.

32.

Kochan

Thomas

Zack

Arnold

. 2009. Negotiations, mediation key to first contract arbitration. Roll Call, June 15. http://www.rollcall.com/news/-35842-1.html.

33.

Logan

John

. 2006. The union avoidance industry in the United States. British Journal of Industrial Relations 44 (4): 651-75.

34.

Malin

Martin

. 2012. The legislative upheaval in public-sector labor law: A search for common elements. ABA Journal of Labor and Employment Law 27 (2): 149-64.

35.

Mayer

Gerald

. 2007. Labor union recognition procedures: Use of secret ballots and card checks. Washington, DC: Congressional Research Service. http://digitalcommons.ilr.cornell.edu/cgi/viewcontent.cgi?article=1566&context=key_workplace.

36.

Morissette

René

Schellenberg

Grant

Johnson

Anick

. 2005. Diverging trends in unionization. Perspectives on Labour and Income 6 (4): 5-12.

37.

Murray

Gregor

Cuillerier

Joelle

. 2009. The sky is not falling: Unionization, Wal-Mart, and FCA in Canada. Just Canada: A Canadian Journal of Work and Society 15 (Special Edition): 78-98.

38.

National Labor Relations Board. 2011. Performance and accountability report. Washington, DC: National Labor Relations Board. http://www.nlrb.gov/sites/default/files/documents/189/nlrb_2011_par_508.pdf.

39.

Prosten

1978. The longest season: Union organizing in the last decade, a.k.a. how come one team has to play with its shoelaces tied together? In Proceedings of the thirty-first annual meeting of the Industrial Relations Research Association. Chicago: Industrial Relations Research Association.

40.

Riddell

Chris

. 2001. Union suppression and certification success. Canadian Journal of Economics 34 (2): 396-410.

41.

Riddell

Chris

. 2004. Union certification success under voting versus card-check procedures: Evidence from British Columbia, 1978-1998. Industrial and Labor Relations Review 57 (4): 493-517.

42.

Riddell

Chris

. 2010. The causal effect of election delay on union win rates: Instrumental variable estimates from two natural experiments. Industrial Relations 49 (3): 371-86.

43.

Riddell

W. Craig

. 1993. Unionization in Canada and the United States: A tale of two countries. In Small differences that matter: Labor markets and income maintenance in Canada and the United States, ed. Card

David

Freeman

Richard B.

Chicago: University of Chicago Press. http://www.nber.org/books/card93-1.

44.

Robinson

Edward Ian

. 1990. Organizing labour: Explaining Canada-U.S. union density divergence in the post-war period. PhD diss., Yale University, New Haven, CT.

45.

Roomkin

Juris

1978. Unions in the traditional sectors: The mid-life passage of the labor movement. In Proceedings of the thirty-first annual meeting of the Industrial Relations Research Association. Chicago: Industrial Relations Research Association.

46.

Schmitt

John

Mitukiewicz

Alexandra

. 2012. Politics matter: Changes in unionisation rates in rich countries, 1960-2010. Industrial Relations Journal 43 (3): 260-80.

47.

Schmitt

John

Warner

Kris

. 2009. The changing face of labor. Washington, DC: Center for Economic and Policy Research. http://www.cepr.net/documents/publications/changing-face-of-labor-2009-11.pdf.

48.

Schmitt

John

Zipperer

Ben

. 2009. Dropping the ax: Illegal firings during union election campaigns, 1951-2007. Washington, DC: Center for Economic and Policy Research. http://www.cepr.net/documents/publications/dropping-the-ax-update-2009-03.pdf.

49.

Singh

Parbudyal

Jain

Harish

. 2001. Striker replacements in the United States, Canada, and Mexico: A review of the law and empirical research. Industrial Relations 40 (1): 22-53.

50.

Slinn

Sara

. 2004. An empirical analysis of the effects of the change from card-check to mandatory vote certification. Canadian Labour and Employment Law Journal 11:259-301.

51.

Slinn

Sara

. 2005. An analysis of the effects on parties’ unionization decisions of the choice of union representation procedure: The strategic dynamic certification model. Osgoode Hall Law Journal 43 (4): 407-50.

52.

Slinn

Sara

. 2008. No right (to organize) without a remedy: Evidence and consequences of the failure to provide compensatory remedies for unfair labour practices in British Columbia. McGill Law Journal 53:687-737. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1100003.

53.

Slinn

Sara

. 2012. First contract arbitration comes to Nova Scotia. Conversation on Work. http://conversationsonwork.wordpress.com/2012/01/05/first-contract-arbitration-nova-scotia/.

54.

Slinn

Sara

Hurd

Richard W.

2009. Fairness and opportunity for choice: The Employee Free Choice Act & the Canadian model. Just Labour: A Canadian Journal of Work and Society 15 (November): 104-15. http://ssrn.com/abstract=1580266.

55.

Slinn

Sara

Hurd

Richard

. 2011. First contract arbitration and the Employee Free Choice Act: Multi-jurisdictional evidence from Canada. Advances in Industrial and Labor Relations 18:41-86.

56.

Uppal

Sharanjit

. 2011. Unionization 2011. Perspectives on Labour and Income. http://www.statcan.gc.ca/pub/75-001-x/2011004/article/11579-eng.pdf.

57.

U.S. Bureau of Labor Statistics. 2013. Union members summary. http://www.bls.gov/news.release/union2.nr0.htm.

58.

U.S. Dunlop Commission (Commission on the Future of Worker-Management Relations). 1994. Final report. Washington, DC: U.S. Commission on the Future of Worker-Management Relations. http://digitalcommons.ilr.cornell.edu/key_workplace/2/.

59.

Weiler

Paul

. 1983. Promises to keep: Securing workers’ rights to self-organization under the NLRA. Harvard Law Review 96 (8): 1769-1827.

60.

Wilson

Rand

. 2012. The labor law reform we need. Labor Notes, July 11. http://www.labornotes.org/2012/07/labor-law-reform-we-need.

61.

Wood

Stephen

Godard

John

. 1999. The statutory union recognition procedure in the Employment Relations Bill: A comparative analysis. British Journal of Industrial Relations 37 (2): 203-44.

62.

Zack

Arnold

. 2009. First contract arbitration: Issues and design. Cambridge, MA: Labor and Worklife Program at the Harvard Law School. http://www.law.harvard.edu/programs/lwp/people/staffPapers/zack/Arbitration_LERA_Blog_3_13_09.pdf.