class actions, disabled rights, copyright, attorneys general, online speech, law schools, obesity, New York, mortgages, legal blogs, safety, CPSC, pharmaceuticals, patent trolls, ADA filing mills, international human rights, humor, hate speech, illegal drugs, immigration law, cellphones, international law, real estate, bar associations, Environmental Protection Agency, First Amendment, insurance fraud, slip and fall, smoking bans, emergency medicine, regulation and its reform, dramshop statutes, hotels, web accessibility, United Nations, Alien Tort Claims Act, lobbyists, pools, school discipline, Voting Rights Act, legal services programs


November 2009 Archives

John Endean

What if America's labor law were more like Canada's? Were our Congress to enact into law a more union-friendly legal code of the sort long familiar to our northern neighbors, what sorts of consequences would we expect? At present, 29.4 percent of workers in Canada are represented by unions, as contrasted with 12.4 percent of workers in the U.S. Would adopting a more "Canadian" legal regime close much of that gap, or only a little of it? And what would be the consequences for employee well-being, for managerial efficiency, and for the health of the U.S. economy generally?

These questions are not new ones among those who follow labor policy, but they have taken on fresh interest given the enormous stir in Washington over the proposed Employee Free Choice Act (EFCA). The top legislative priority of organized labor, and potentially the most significant piece of labor legislation since the Wagner Act of 1935, EFCA consists largely if not entirely of policy initiatives that follow a "Canadian" path:

* "Card check". Today, in most cases, installing a union to represent workers at a place of business requires a majority vote of the workers by secret ballot. EFCA's best-known and most controversial provision would require recognition of a union upon its presentation of signatures on union cards from a majority of the workers in a proposed bargaining unit. (1) The card-check system has a long track record in Canada.

* Imposed arbitration of first contracts. Once organization is accomplished, EFCA would compel management to reach a first contract with the new union, by providing for mandatory arbitration and imposition of a contract by a government-appointed arbitrator should negotiations not result in a contract by a certain point. Some Canadian employers both public and private are subject to imposed arbitration at negotiation impasse; in the United States up to now, such requirements have ordinarily been imposed only on some public employers.

* "Quickie" elections. As the unpopularity of eliminating the secret ballot has become clear, organized labor and its supporters have begun to cast around for "compromise" EFCA provisions aimed at bolstering unions' organizing efforts in other ways. One such idea is to speed up greatly, perhaps to 10, 12 or 15 days, the holding of elections following a union petition, which currently in the U.S. are held a median of 39 days later. In Canada, by contrast, there is usually only a five-day window before elections. Shorter periods before a vote are generally considered unfavorable for employers because it gives them little time in which to assemble a case against unionization and make it known to workers; the union, by contrast, will ordinarily have had weeks or months to make its case to workers in private persuasion before it surfaces with its election demand.

Quickie elections, in contrast to card check, are often thought to have "moderate appeal". Thus, William B. Gould IV, an influential legal scholar, former counsel to the United Auto Workers, and former chairman of the National Labor Relations Board (NLRB), supports EFCA in general principle but has criticized card-check and recently proposed quickie elections as part of a "better approach" that might command bipartisan support:

Secret ballots to resolve union representation rights are the way to go, and Obama should meet the Republicans halfway by saying so - and then add this all-important coda: Elections should continue only if the law ensures that voting is conducted expeditiously - for instance, within one or two weeks of the filing for a union's petition seeking recognition. This is the case in Canada, whereas in the United States, the resolution of union drives currently takes months and sometimes years. Quick elections are the key to meaningful reform because delay is the principal way in which labor law stacks the deck against employees. It allows employers to engage in one-sided anti-union campaigns of intimidation and coercion, with little possibility for remedy. (2)

Of all countries that might provide examples for labor law reform, Canada is the most similar to the United States culturally and politically. It is also the most familiar to American managers (many of its firms, especially in the industrial heartland of Ontario, are owned by or affiliated with American corporations, the well-known cross-national integration of Big Three auto manufacturing being only one example.) How relevant is the Canadian labor experience, and what can it teach us about the achievability of EFCA's constituent parts and the costs and benefits they might bring? (3)

Not one but multiple systems. In contrast to our system of labor law in the U.S., in which the federal government occupies most of the field and sharply limits the 50 states' scope for divergence, Canada genuinely shares the regulation of labor relations and union certification between Ottawa and the provinces, with the provinces given the lead. (4) About ten percent of the total Canadian workforce is covered by federal labor law. This includes federal government workers as well as private workers in certain industries deemed national, which include banking, shipping, telecommunications, and inter-provincial trucking.

Canada's federal labor code is a card check system with first-contract mediation and binding arbitration. There are no secret ballot elections. Instead, if more than 50 percent of the workers in a proposed bargaining unit sign cards, the union is certified by the Labour Relations Board. In short, this system does prescribe something a lot like the EFCA's proposed regime for a tenth of the Canadian workforce.

What about the ninety percent of workers not covered by Canada's federal labor law? They are subject to the labor laws of the provinces in which they reside. Up until 1976, all of the provinces used card check. Beginning in that year, however, changing political and economic circumstances have led some provinces to rethink the methods of union selection.

Today, six of the ten provinces - Alberta, British Columbia, Newfoundland and Labrador, Nova Scotia, Ontario, and Saskatchewan - require a secret ballot. The four others - Manitoba, New Brunswick, Prince Edward Island, and Quebec - use a card check system. Six provinces also mandate in some form first contract mediation and binding arbitration: British Columbia, Manitoba, Newfoundland, Ontario, Quebec, and Saskatchewan. Overall, of workers in industries covered by provincial labor law, about 68 percent of the Canadian work force lives in provinces with a mandatory secret ballot, and the other 32 percent in provinces with card check.

The numbers fluctuate because card check can be a political football in provincial politics. New Brunswick, for example, first adopted a secret ballot and then reverted to card check as contending political parties succeeded each other in office. Similarly, British Columbia adopted secret ballot elections in 1984, returned to card check in 1993, then readopted the secret ballot in 2001.

There is thus no single "Canadian" model for union organization. If there is a "median" or "most typical" law among the diverging Canadian examples, it is probably the "expedited secret ballot" system that prevails in six provinces including the most populous, Ontario.

Ontario's expedited secret ballot. Ontario switched from card check to a five-day secret ballot certification process in 1995. This change was part of a larger program of tax, budget, and regulatory reforms called the Common Sense Revolution by its architect, Mike Harris, Ontario's conservative Premier.

Upon a showing that 40 percent of the workers in a proposed bargaining unit have signed cards expressing an interest in joining a union, a secret ballot vote must occur within five business days. That's not a lot of time, especially since unions have the initiative in triggering the process and can choose the time and circumstances they consider most favorable. Businesses may not even know an organizing effort is in progress until the union, having secretly obtained the number of signatures necessary to force an election, files its application for certification.

Once a union submits an application, companies must submit a formal response, including a list of relevant employees, to the Labour Relations Board and to the union within two business days. This mandated response can itself be a costly legal scramble and paper chase, and makes it even more likely that managers will be distracted during the few days that will be their only formal chance to make their case on the organization vote.

In that latter task, managers are far more limited by what they can say about the impact of unionization than are their American counterparts, a fact that often comes as a surprise to American companies with Canadian subsidiaries. In one notorious case, for example, from 1996, disgruntled employees of a Wal-Mart operation in Windsor, Ontario, approached the United Steelworkers of America seeking representation. Enough cards were signed to force an election. In the election, however, 79 percent of the employees voted against the union. The Employee Labour Relations Board proceeded to uphold union objections to Wal-Mart's American-style interference with the unionization process. Bizarrely, at least from an American perspective, the Board based its finding in part on Wal-Mart's silence when employees asked if the store would be closed if the unionization drive succeeded. The Board found the company's lack of comment to have had a "chilling effect" upon the union campaign. (5) By way of remedy, it did not (as one might have expected) merely throw out the election results that had gone against the union: it ordered the union installed to represent the workers, majority vote or no.

From a management standpoint, the Ontario system does have some mitigating features. For example, if a union loses an organizing vote, it must wait a year before trying to organize the same company again. The one-year ban applies not only to the losing union but to all other unions as well.

More important, Ontario no longer imposes mandatory arbitration automatically if a company and union cannot agree on a first contract. The 1995 legislation that supplanted card check with the five-day vote also modified the then-existing mandatory arbitration provision by putting in place a four-part test for evaluating first contract negotiations. This four-part test comes close to a bad-faith bargaining hurdle. In effect, if a company can show that it is bargaining with the union in good faith, it can avoid mandatory arbitration.

Accustomed as they are to campaign periods of about a month before an organizing election, most American managers would likely regard the Ontario secret ballot system, with its abbreviated, five-day campaign period, as a thumb on the scale in favor of union organization. According to a labor lawyer in Canada interviewed for this note, Ontario managers are not in revolt against the five-day system because "you get used to it." The possibility of quick organization becomes just one more "crisis management" issue, with all the transaction costs that crises inevitably entail.

Union "density" and the legal background. Union density - the percentage of total workers who belong to unions - is greater in Canada than in the United States. According to Statistics Canada, about 29.4 percent of all Canadian workers belong to unions. In the United States, the equivalent figure is about 12.4 percent.

In both countries there is a sharp disparity between union density in the public and private sectors. In Canada, 71 percent of public sector workers belong to unions, while only about 16 percent of workers in the private sector are organized. A similar ratio prevails, but at lower rates, in the United States, where union density is about 39 percent in the public sector and 7.6 percent in the private. The comparatively robust presence of unions in the public sectors of both countries reflects the disinclination of government managers to contest unionization, in part because the costs of organization - higher wages, expensive benefits, and restrictive work rules - are indirectly spread among the public at large. (6) It is also worth noting that the public sector employs a significantly higher percentage of the workforce in Canada than it does in the United States.

No one factor explains the overall density "gap" between Canada and the United States. Unquestionably card check has facilitated unionization north of the border, which tends to confirm the feeling of labor leaders in the United States that it would prove helpful to them here. (7) One often-quoted study, by Professor Susan J.T. Johnson of Wilfrid Laurier University, estimated that the greater use of card check in Canada accounts for somewhere between 17 to 24 percent of the difference in union density between Canada and the United States. (8)

Available evidence suggests that the adoption in some provinces of a secret ballot, even when accompanied by a relatively brief campaign period, has made organization more difficult. Perhaps the most striking example is in New Brunswick where the success rate of union organizing fell 19 percent when the secret ballot was put into place and rose by about the same amount when the card check regime was later restored. (9) A study of organization in Ontario found that "the overall proportion of successful certification applications [was] substantially lower under the mandatory vote than it had been under the card-check system." (10)

Note that there is surprisingly little backing for the sometimes-heard assumption that Canada (in supposed contrast with the U.S.) is a country where unionism is simply uncontroversial and popular with the broad populace. With a favorable legal framework in place for many years, a union movement that represents only 16 percent of private sector workers cannot exactly claim a decisive mandate from the Canadian working public. The Canadian experience following many provinces' introduction of secret ballot elections also suggests that when workers are allowed to vote on whether to join a union - when, in other words, they regard joining a union as a matter of individual choice in which competing considerations are brought to their attention - they are measurably less inclined to join. This is a difficult point for unions to accept.

The flagging spirit of Canadian unionism. Unionism in Canada's public sector appears for the moment to be secure. But only the growth in public sector unionization has kept Canada's overall density rate near 30 percent; Canadian private sector unions are struggling, by contrast, with what growth there has been in union membership outstripped by the greater proliferation of nonunion jobs. Pradeep Kumar of Queen's University, a sympathetic observer of the Canadian labor movement, has argued that in general "the data appear to portray a picture of a stagnant labour movement with declining density in a wide range of areas, particularly in private service industries with expanding employment, and with a false sense of security due to continuing union strength in the public sector." (11)

Even the most visible instance of new private unionization in recent years is indicative of this weakness. The giant auto-parts supplier Magna International, which had long resisted unionization, reached a 2007 agreement with the Canadian Auto Workers (CAW): in order to achieve this long-sought goal, however, the CAW gave up the right to strike, amended its grievance procedures, and permitted the company to screen candidates for union representative - so-called "employee advocates" -- at each plant before they are ratified by employee vote.

Some in Canada's labor movement are preoccupied with the hope of turning around this trend by prevailing on provincial governments to restore card check. (12) New Brunswick aside, they have had scant success. When the Liberals returned to power in Ontario in 2003, there were rumblings about restoring card check across the board. That did not happen, and instead the old system was restored in 2005 only for workers in construction workers (a move, oddly enough, condemned as "sexist" and an "atrocity" by one union because the construction industry has "a predominantly male workforce"). (13) Although Canadians are famously prickly about their social and cultural independence from the United States, it is probably true that the best boost for a return to card check in Ontario and other provinces would be the adoption of card check by the United States, Canada's largest trading partner.

In this context, calls for a return to card check may be a distraction from the more important matter of making union membership relevant for a new generation of Canadians who may not see belonging to a union as self-evidently desirable. Perhaps there is a lesson here for American labor unions as well.


(1) In the first half of 2008, the union win rate in NLRB private sector elections was 66.8 percent. This win rate has been tracking upward, with one exception, for the last five years. In 2003 the union win rate was 58.3 percent; in 2004, 58.6 percent; in 2005, 61.3 percent; in 2006, 61.4 percent; and in 2007, 60.5 percent. See "Union Win Rate in NLRB Elections Increased Substantially in First Half 2008," BNA Daily Labor Report, 217 DLR, January 28, 2009, pp. C-1 - C-2.

(2) William B. Gould, IV, "How Obama Could Fix Labor Law," Slate, August 28, 2008.

(3) Recently three American companies sympathetic to labor law reform - Costco, Starbucks, and Whole Foods - formed an organization called the "Committee for a Level Playing Field for Union Elections." The Committee has as its centerpiece the maintenance of secret ballot elections with a shortened campaign period, along the lines of what Gould identifies as the Canadian model.

(4) Before 1925, collective bargaining legislation was the responsibility of Canada's federal government. In 1925, the United Kingdom Privy Council, in Toronto Electric Commissioners v. Snider established priority of provincial rather than federal jurisdiction over most labor and employment issues. Subsequently, the Constitution Act of 1867 delineated the separation of powers between the federal and provincial governments.

(5) See Douglas Gilbert and Brian Burkett, "Canada's Labor and Employment Laws," June 2001. Gilbert and Burkett are Canadian management-side labor lawyers and their piece can be found at

(6) In addition, government employment is by nature fairly static and captive in the sense that it cannot be "offshored." Unlike private sector workers, government employees are not buffeted by the pressures of international competition, mergers and acquisitions, technological change, or bankruptcy. Organizing government workers and bringing new members on board is thus fairly routinized and predictable. It is, in other words, easier.

(7) In addition, unlike the United States, Canada permits mandatory union membership in collective agreements as a condition of employment. And in contrast to so-called "right-to-work" states in the U.S., Canada also permits mandatory dues payments, again as a condition of employment. Jason Clements, Niels Veldhuis, and Amela Karabegovic, "Explaining Canada's High Unionization Rates, Fraser Alert, August 2005. This piece can be found at:

(8) Susan Johnson, "The Impact of Mandatory Votes on the Canada-U.S. Union Density Gap: A Note," 43 Indus.Rel. 356 (2004), quoted in Anne Layne-Farrar, "An Empirical Assessment of the Employee Free Choice Act: The Economic Implications," The Alliance to Save Main Street Jobs, March 3, 2009, p. 16.

(9) Chris Riddell, "Union Certification Success Under Voting Versus Card-Check Procedures: Evidence from British Columbia, 1978 - 1998," Canadian Journal of Economics, vol. 34, no. 2, quoted in Jason Clements, Niels Veldhuis, and Amela Karabegovic, "Explaining Canada's High Unionization Rates," Fraser Alert, op. cit.

(10) Sara Slinn, "The Effect of Compulsory Certification Votes on Certification Applications in Ontario: An Empirical Analysis," Canadian Labour and Employment Law Journal, vol. 10, no. 3.

(11) Pradeep Kumar, "Whither Unionism: Current State and Future Prospects of Union Renewal in Canada," June 2008, available online at: See also Pradeep Kumar, "is the Movement at a Standstill?", Our Times, vol. 27, issue 5, October-November, 2008.

(12) See, e.g., Bruce Allen, "On and After the Magna Vote," New Socialist, See also, Mine Mill598/CAW Organizing Report, April 27, 2005,

(13) The "sexist" and "atrocity" language can be found in a draft advocacy letter to members of the Canadian Parliament that is included in the "Labour Law Reform - Lobby Kit" created in 2005 by the United Steelworkers District 6. Available at:

* * *

John Endean is the president of the American Business Conference, a Washington, D.C.-based coalition of leaders of midsize growth companies. This paper, original to Point of Law, was commissioned by the Manhattan Institute as the first in a planned series of Institute papers on labor policy. It was published November 9, 2009.

* * *



Rafael Mangual
Project Manager,
Legal Policy

Manhattan Institute

Published by the Manhattan Institute

The Manhattan Insitute's Center for Legal Policy.