AS THE election dust settles, attention turns to President-elect Obama's governing agenda. If he prioritizes the labor-law changes he favored as a senator and candidate, he will inflict serious, lasting costs on American workers and the economy.

The size of the federal deficit will constrain the taxing and spending initiatives of the new administration. A more liberal Congress, however, and special-interest groups, will place great demands on Mr. Obama. Organized labor, which accumulated significant IOUs during the campaign, win lead the way, which means demands for radical changes to labor laws.

The Employee Free Choice Act, favored by then-Senator Obama, passed the House and nearly cleared the Senate. With a near-filibuster-proof Senate, the act is almost certain to be re-considered. This measure fundamentally changes core principles of labor law. Most seriously, it removes the requirement for a secret-ballot vote to certify a union as the exclusive bargaining agent of workers.

Currently, any union trying to gain certification first needs the support of 30 percent of workers in the form of signed union cards or a petition. Once this hurdle is cleared, the National Labor Relations Board (NLRB) conducts a secret-ballot vote to certify the union. Workers then anonymously and privately vote to decide if they want union representation. If 50 percent plus one of the workers vote favorably, the union is certified.

Under the Employee Free Choice Act, workers will be unionized based simply on the signing of a union card. With "card check" in place, workers lose the fundamental democratic right to choose a union through secret-ballot voting.

Academics friendly to labor have argued that the actual effects will be marginal. Labor leaders such as the SEIU's Andy Stern have suggested that more than 1 million new members will be added each year. Some conservative commentators have argued that the changes will upend the country's labor market. So who is right?

Several Canadian provinces have switched from secret ballot to card check - and vice versa - which provides a natural experiment. The evidence from Canada indicates that changing how we certify unions will have a serious and worrying impact.

Professor Christopher Riddell of Queen's University has completed work on the effects of changing certification methods in British Columbia between 1978 and 1998. He concluded that unionization success rates increased by roughly 19 percent after card check was introduced. He also found that union success fell by the same amount when secret ballot voting was introduced.

Contrary to union rhetoric, the evidence indicates that employer intimidation is quite small. The reduced success of unions under secret-ballot voting is not explained by employer intimidation. Indeed, it is far more likely that card check certification will increase the intimidation of workers, not by employers but by unions.

Democrats are also considering mandatory arbitration for first contracts in circumstances when an agreement cannot be reached. This would entail an unheard-of role for the government in private-sector bargaining. Given that the NLRB and the Department of Labor will be stacked with Democrat appointments, the incentive for any newly installed union is to push for enormous concessions by the employer. Increases in labor disputes and strikes are more than likely.

The changes also impose costs on the economy. An increasing body of academic research has concluded that higher unionization rates are associated with lower economic performance generally. Studies have found that more regulated, higher unionized labor forces are characterized by higher unemployment, less job creation, and, critically, reduced investment.

All American workers will bear the costs, regardless of whether they are affected by President Obama's radical changes in labor law. The United States will become a less hospitable, less attractive place to do business and invest. Americans may wonder if that's what they voted for on Nov. 4, 2008.

Jason Clemens is the director of research at the Pacific Research Institute in San Francisco.