I applaud Richard Freeman for spurring this discussion. There is no question that the National Labor Relations Act (NLRA) is presently an utter failure in serving its explicit mission of protecting and encouraging workers' right to have unions and engage in collective bargaining. Indeed, as Richard has noted previously, the Act has institutionalized a process that effectively gives management veto power over whether or not workers become organized.

In addition, I agree with Richard that the federal labor preemption doctrine operates as a block on creativity and experimentation that in some instances might well benefit workers at the state and local levels. The AFLCIO itself has in recent years supported numerous legislative efforts that, while organizing-friendly, have been sufficiently outside the scope of the NLRA to survive preemption challenges: worker retention laws, labor peace ordinances, bans on use of public monies in state contractors' employees' organizing campaigns, extension of bargaining rights to categories of employers not covered by the Act, and project labor agreements, to name a few. In other cases, however, such ordinances have been struck down. It is always tempting, therefore, to speculate on how much better off workers might be if there were no federal preemption constraints at all.

That said, I would raise three sets of concerns about Richard's proposal relating to (1) its underlying political assumptions, (2) its reliance on public sector experience as indicative of what we might expect in the private sector, and (3) the unlikelihood of an overall positive outcome.

Political Assumptions

First, there is simply no reason to believe that corporate employers would not oppose this version of labor law reform every bit as strongly as they oppose an NLRA overhaul. The management community is almost universally satisfied with the present system—why would it not be given that it exercises such complete control over workers' right to unionize and bargain collectively under existing law? Furthermore, the uneven playing field aside, national employers would surely resist the notion of adopting a different labor relations paradigm in each of the fifty states. Thus, as a practical matter, if the labor movement and its allies were able to muster sufficient political strength to get the Freeman proposal through Congress and the White House, they would surely have enough influence to effectuate the kind of comprehensive labor law reform that we would like to see at the national level.

Second, again purely as a practical matter, the proposal ignores the very real issue of resources. It is expensive enough when labor has to defend one "paycheck deception" ballot initiative in California, or push for one "labor peace" ordinance in Milwaukee. Were we to have to start from scratch in fifty states in order to legislate mini-NLRA's in each one—or even were we to write off the "red states" and focus only on a dozen or so with the best prospects for passing sympathetic state labor laws—the labor movement would find itself financially outmatched virtually everywhere.

Reliance on Public Sector Experience

Richard relies heavily on experience in states with developed public sector labor laws and points to a correlation between those states and the ones with relatively high union density. The reality, however, is that the public sector laws in those states are typically patterned after the NLRA and—on paper—are rarely better in any significant way than their federal counterpart. (Indeed, the main difference normally is that public sector versions do not afford as full, if any, a right to strike.) Further, of much more importance than Richard suggests, union density is higher in states with "good" public sector labor laws mainly because public employers rarely interfere with their workers' decision to unionize. State legislative bans on captive-audience meetings, requirements to make names and addresses of bargaining unit members available, or agreements to stay neutral during a campaign are simply not as necessary. Finally, shutdowns, closures, and moves offshore do not occur in the public sector, nor would threats or predictions of such carry any weight.

Unlikelihood of a Positive Outcome

The bottom line question, of course, is whether under current circumstances we would really be better off without federal labor preemption, with each state free to enact its own superseding versions of the overall portions of the NLRA if it chose to do so. And for the following reasons, as well as that of inadequate resources described earlier, I think not.

First, I fear the risk of a race to the bottom would be significant. Even before the presence of international and global competition entered the mix, we have seen how labor costs have influenced companies' decisions to relocate from union-friendly northern states to union-scarce states in the south. The effects of globalization will only make those decisions more likely in any industry with location flexibility.

Second, as courts have been reminding us, many of the protections that workers need from employer interference cannot be extended for reasons that go beyond federal labor preemption. Neutrality requirements, for example, are arguably restraints on employer speech, constitutionally preempted, and cannot be legislated by either the federal or state governments.

Labor's major efforts to improve matters, therefore, would probably focus on card-check recognition provisions, which may bring more loss than gain. After all, under the NLRA, at present, card check is not unlawful, even though not mandatory. Under the state-by-state approach, states would presumably be free to ban voluntary recognition agreements altogether; for that matter, they could ban private sector collective bargaining completely if they wished—as a few already do in the public sector. Would labor have any real opportunity to achieve a meaningful card-check recognition system in more states than in those in which it would lose it entirely?

Finally, at best the outcome would result in a pronounced ghettoization of union-friendly states, one that would undoubtedly reduce the labor move-ment's national political strength considerately. Even if union density increased considerably in a handful of states, as Richard predicts, its impact in presidential and congressional elections, for example, would be reduced overall if its density declines in an even larger number of other states.

Alternatives

This unreceptive a critique would be somewhat unfair without suggesting some possible alternatives. After all, as Richard suggests, with the state of worker protections under current federal law as bad as it is, we really cannot afford to be too critical of any proposal that might encourage experimentation with new legal structures.

I would start with a variation on Richard's theme, one that he also alludes to: the NLRA as a minimum standard. Under this proposal, the NLRA could be preempted but only in one direction; states would be permitted to pass their own legislation that strengthened worker protections of the right to organize and bargain collectively, but not legislation that weakened such rights. This would be analogous to the authority that states have to be more but not less protective with other national labor and employment standards, for example, minimum wage and overtime protections under the Fair Labor Standards Act, or health and safety protections under the Occupational Safety and Health Administration.

A problem with applying this same approach with the NLRA is definitional, however, since it may be somewhat more difficult to define what a minimum standard is when it comes to recognition procedures. Nonetheless, this should be a challenge worth confronting, as the NLRA already contains an explicit, one-way, non-preemption provision in one area of the law: union security provisions. Balancing that with another one-way non-preemptive arrangement allowing states to regulate more protectively when it comes to selection of representatives would provide a logical balance.

Finally, one might also consider looking at preemption from the other direction, that is, should at least some aspects of the NLRA be considered preempted by international labor law? Numerous United Nations and International Labor Organization (ILO) conventions and declarations have long held the right to organize and the right to bargain collectively to be basic human rights. Most recently, the ILO's 1998 Declaration on Fundamental Principles and Rights at Work reaffirmed that these both are among a handful of universal worker rights that must be respected. Considerable ILO jurisprudence has evolved over the years as to both the right to organize and the right to bargain collectively. No less an independent authority than Human Rights Watch conducted a lengthy study in 2000 and found the United States significantly failing to conform with international norms on freedom of association at work and the right to bargain collectively. There is little doubt that American workers today would be better off were their rights measured by international ILO standards than they are under the NLRA. Perhaps a topic for next year's conference should be preemption of the NLRA, rather than by the NLRA.