Friday, January 16, 2009

President-Elect Obama on EFCA

President-Elect Obama recently discussed the Employee Free Choice Act with the Washington Post editorial board. The audio is here, and the relevant portion starts at the 56-minute mark.

Former Board Member Peter Kirsanow also spoke about the bill on January 17 with the host of America's Business, a production of the National Association of Manufacturers. A link to the audio is here, and Kirsanow's interview starts around 21:30.

Monday, January 12, 2009

Follow-Up on Proposed Reform

Joe Brock, from Laboring Away at the Institute, posted a comment to my proposal for labor law reform. Below, I break his comment into segments and respond. His comments are in italics and my responses are in bold type.

I've been looking over your proposal with some interest, and I have some comments that i'd like to share. Your statistics are obviously well researched, and I think these statistics would be surprising to the general public who might not have the benefit of the numbers. For example: You correctly note that the median time between petition and election was 39 days in 2007. I think that would surprise many who have come to the conclusion in listening to the unions argument that it routinely took much longer. I think that 39 days from petition to election is about what it SHOULD take to make such a potentially career changing decision.

You focus on the half of elections held within 39 days and say the law does a good job. I focus on the half of elections that take more than 39 days (including 7% that take more than 56 days and the small minority that take many months) and say that the law can do a substantially better job. There is no need to wait months. We can do better.

7-21 days is clearly not enough time if you want your employees or potential members to do their own research based on the facts being offered in a campaign. Who you choose to represent you will impact your income and possibly your personal future more so than who you choose to be your President.

Twenty-one days is long enough for employees to perform independent research. Employees would be free to wait that long before casting their ballots. Employees who think they have enough information would be free to cast their ballots 7 days after petition (but no sooner). Employers are thus guaranteed 7 days to communicate their views. That is plenty of time. Moreover, nothing would prevent employers from communicating generally about unionization before any petition is filed. Employees appreciate as well anyone that their decision is important and that each side’s views should be taken with a grain of salt.

Holding an election without a decision on the apropriateness of the unit being clarified is a non-starter with me. As you correctly point out, these issues are rare, and the Board does a great job in getting the parties to reach agreement on this when it becomes an issue. I think that if there was going to be a "trial run" it would open up the potential for MORE questions of appropriateness, allowing the employer a free bite at the apple, or a peak under the covers, as it were. Win the "trial run" and drop the question. Lose the election and appeal forever. It would get abused as part of an election strategy.

What if the Board were to conduct the election, impound the ballots, fully resolve any dispute over unit appropriateness, and finally conduct the tally of ballots? That procedure would alleviate your concern about an increase in unit-appropriateness disputes because employers would have to litigate the issue before knowing the election result. That procedure would also prevent unit-appropriateness disputes from greatly delaying elections and thereby minimize the temporal opportunity for coercion by both sides.

I'm also not on board with Internet and telephone voting. little or no delay in the election process can be attributable to the physical process of setting up a time and place. 28 days doesnt normally become 40 because the Board agent has a scheduling conflict. Besides, i've been involved with many union elections. The process currently in place is damn near impossible to manipulate. You walk into a room with observers from both sides, you fill out a ballot in secret, put it in a box that never leaves anyones sight. The Board agent knows how many ballots went out, and they're printed on unusual paper. They count them in front of everyone. It's the fairest election process i've seen. In fact, it's the ONLY fair process as far as unions go. This is why they want to change it. Internet and telephone voting make me nervous, and I can see union organizers giving incorrect phone numbers on the cards, or manipulation of internet, etc.

The NMB has used telephone voting for 6 years and internet voting for over 1 year without any report of a security breach. Fortune 500 companies use the same technology for shareholder votes, and I’m not aware of any reported security breaches there. In short, I don’t think there’s a reasonable concern that unions will hack the system. (Also, keep in mind that the Board has for a very long time conducted mail-ballot elections. Do you think that internet and telephone voting is less secure than mail balloting?)

I recommended telephone and internet voting because, unlike manual balloting, it allows for a 2-week voting period that starts a week after the petition is filed. The Board can’t accomplish that with manual balloting.

Now, moving onto your idea about "limited" bargaining rights. Perhaps I misunderstand your proposal, but doesn't the concept of representing only those who supported the union fly in the face of a secret-ballot vote? Are you espousing that employees publicly declare their allegiance to either the union or the employer? Sounds like an EFCA end around and it has the same problems of any public vote, or card-check.

The law currently permits members-only bargaining, absent a 9(a) exclusive representative. I'm merely proposing requiring it in this limited context to eliminate the potential for years of post-election delay in the unionization process.

The important point is that the election has already been held, interim membership will not affect the election tally, and the incentive for coercion of employees has thereby been greatly reduced. The ULP process will be available for any coercion that did occur.

Again, the delay between vote and certification of the unit is NOT the problem that the unions are making it to be, as your statistics correctly point out. Most elections are certified in 10 days and bargaining can then begin.

Most election results are certified shortly after the election. Interim, members-only bargaining won’t affect employers or unions in those cases. In the minority of cases where a long delay occurs between election and certification, interim, members-only bargaining will promote collective bargaining, one of the Act’s enumerated purposes.

The REAL problem unions are faced with in the collective bargaining process is that the union is coming to a negotiation with nothing to offer an employer. If you went into negotiations with a car dealer with NO money, and NO credit, you'd get NO car. The union brings no customers, and no capital for an employer. Why would they agree to any improvements if they didn't see a business benefit to it?

Unions offer employees (their consumers) an opportunity to require their employer to sit at the table and listen to their concerns. Some employees consider this a valuable commodity. Other employees don’t. Employees, speaking through their representative, sometimes have some good input for employers. Employers sometimes improve working conditions based on that input. Sometimes they improve working conditions based on economic pressure exerted by their unionized employees.

I could go on, but you probably stopped reading a while ago. I certainly commend the thinking outside the box for your ideas, even though I couldn't quite agree with them.

Thanks for your comments.

Sunday, September 14, 2008

A Proposal for Telephone & Internet Voting, Speedy Elections, and Interim, Members-Only Bargaining

Delay is a significant problem under the NLRA. The Board took more than 100 days to resolve 21% of representation cases in the fourth quarter of FY 2007, and initial bargaining can be further delayed by an appeal to a circuit court. NLRB Performance & Accountability Report FY 2007 at 39-40.

At least two types of delay can occur during the representation process, thereby postponing initial bargaining: (1) the delay between the date an election petition is filed and the date the NLRB conducts the election; and (2) the delay between the date of election and the date of a final determination that the employer is legally obligated to bargain with an elected union. The first type of delay occurs in all representation cases. The second type of delay occurs in a small minority of cases. However, that delay can be extremely long, and it severely frustrates employees’ collective-bargaining rights. Both types of delay are largely a consequence of current law, not the fault of Board personnel.

To speed up the representation process, I propose that the Board start conducting elections by telephone and the internet, as does the National Mediation Board. Second, I propose holding elections before resolving disputes over unit appropriateness. Finally, I propose a statutory amendment that would impose a limited duty to bargain immediately upon a tally revealing that a majority of employees voted in the union’s favor. Below is an elaboration of the problem and my proposals. I’m looking forward to anyone’s comments or counterproposals.

Delay In Holding Elections
In the vast majority of cases, at least 28 days elapse between RC petition and election. The median was 39 days in FY 2007. Memo GC 08-01, at 2. In seven percent of cases, it took longer than 56 days. Id. In a very few cases, it can take quite a while. See, e.g., TransCare Paratransit, 29-RC-11482 (2008) (8 months); Marymount Manhattan College, 02-RC-23151 (2008) (17 months).

Long delays between petition and election are sometimes attributable to a dispute over the appropriateness of the petitioned-for unit. Parties were unable to reach an election agreement in 8.8% of RC cases in 2007 and 11.9% of RC cases in 2006. Memo GC 08-01, at 6. The Board’s current practice is to resolve a dispute over unit appropriateness before holding an election. See Sections 101.20 and 101.21 of the Board's Rules and Regulations. Understandably, it takes the Board some time to resolve the factual and legal issues presented. The election is held afterward.

Delay between petition and election creates an opportunity for parties to coerce employees before they have recorded their sentiment. Even absent coercion, long delay tends to frustrate an organizing movement as employees’ interest in unionization might naturally wane. Thus, to promote collective bargaining and free choice, employee sentiment should be recorded as quickly as possible after the filing of a petition (while ensuring that all parties have notice and an opportunity to discuss unionization before a vote is taken).

Facilitating Speedy Elections
To facilitate speedy elections, I first recommend that the Board adopt the National Mediation Board’s telephone and internet voting procedures. See NMB Representation Manual, Section 13; Internet Voting Comment Period, 34 NMB 200 (2007); Introduction of Internet Voting/Mock Election, 34 NMB 71 (2007); Telephone Electronic Voting, 29 NMB 482 (2002). The NMB has conducted elections by telephone since 2002 and by the internet since October 2007. The NMB assigns each voter two codes to be used in conjunction to keep voting anonymous and secure. Thus far, the NMB’s technology has maintained ballot secrecy.

Using telephone and internet voting, the Board could feasibly hold elections over a period between 7 and 21 days after the filing of the RC petition. Upon filing, the Board could quickly notify the employer, require the employer to immediately post an election notice, obtain an Excelsior list, and send voting instructions to eligible employees. Section 9(c) (1) requires the Board to hold a hearing before conducting an election to decide whether a question of representation affecting commerce exists. A very limited hearing could be held on the sixth day after a petition. The sole issue would be whether the employer affects commerce, a fact that is very frequently the subject of a stipulation. (If Section 9 requires a more substantial hearing -- which is not evident from its text -- I propose a statutory amendment that would permit my proposed procedure.) If 7 days is impracticable, the voting period could be 14 to 21 days.

My second proposal is to change current practice and start holding elections before resolving a dispute over the appropriateness of the petitioned-for unit. This change would enable the Board to hold nearly all elections within 7 (or 14) to 21 days after a petition is filed.

The general rule of “elections first” would have a limited exception. An election should not be held in a clearly inappropriate unit. Regional Directors would have unreviewable discretion to resolve unit appropriateness before holding an election where they find that the petitioned-for unit is clearly inappropriate (e.g., a petition for a unit of managers, or of employees of multiple employers, or employees of an RLA carrier).

A critic might argue that my proposal will result in some wasteful elections. In some cases, the Board will ultimately decide that a petitioned-for unit is not appropriate and therefore it will dismiss the petition or direct an election in a different, appropriate unit. I would respond that current law entails a similar waste. Conducting a hearing on unit appropriateness before an election is a waste when the union ultimately loses the election.

Delay Between Election and Final Determination of Duty to Bargain
As discussed above, the second type of delay is the delay between the election and a final legal determination that a union is the exclusive collective bargaining representative of an appropriate unit of employees (“post-election delay”). Post-election delay can be extremely long. In the meantime, employees don’t get the benefit of collective bargaining.

Post-election delay usually occurs because an employer has filed determinative challenges and/or election objections. Before certifying a union, the Board must resolve challenges and objections, and that process takes time. In some cases, it can take quite a while. Bloomfield Healthcare Center, 352 NLRB No. 39 (2008) (22 months between election and Board certification); Lily Transportation Corp., 342 NLRB No. 121 (2008) (8 months); Reliable Trucking, Inc., 349 NLRB No. 79 (2007) (20 months).

Even after the Board issues a certification of representative, some employers refuse to bargain. It takes time for the GC to file a motion for summary judgment and for the Board to issue a bargaining order. Casino Aztar, 352 NLRB No. 41 (2008) (4 months between certification and Board’s bargaining order); Carroll College, Inc., 350 NLRB No. 50 (2007) (22 months).

Even after the Board issues its bargaining order, employers can and do appeal the Board’s decision to a court of appeals. Again more time elapses. U-Haul Co. of Nevada, Inc. v. NLRB, 490 F.3d 957 (D.C. Cir. 2007) (19 months between Board’s bargaining order and court’s enforcement order); Sprain Brook Manor Nursing Home, LLC v. NLRB, 255 Fed. Appx. 529 (D.C. Cir. 2007) (14 months). In U-Haul and Sprain Brook, it took 4 years and 2 years, respectively, between the filing of the petition and the court order to bargain.

Only a minority of cases involve such lengthy post-election delay. By my calculation, employers filed determinative challenges and/or objections in approximately 7.6% of cases in which the tally favored a union in 2006, and approximately 6.4%, 7.3%, and 7.7% of such cases in 2005, 2004, and 2003, respectively. See, e.g., 2006 Annual Report at Tables 11B, 11C, and 13.

Employers withdrew their challenges or objections in a minority of those cases. Most cases required a hearing and a decision by an RD. As stated above, some cases were appealed to the Board and some further appealed to a court. While this entire group of cases is a relatively small percentage of total cases, the lengthy delay involved is a significant problem under the Act. Employees are deprived of their freely chosen bargaining representative for a long time. This is true regardless whether the employer has appealed based on a good-faith belief in the merit of its arguments or whether it has appealed solely or primarily to delay bargaining.

The wait is particularly troubling because, in 2006, the Board overruled employers' elections objections in 86.8% of the RC cases in which it ruled on them. 2006 Annual Report Table 11D. That percentage was 92.5% in 2005, 91.8% in 2004, and 91.4% in 2003. Thus, in most cases, the long wait results in a union certification. (I found no statistics regarding how often the Board certifies a union after resolving an employer's determinative challenges. I speculate that a union is certified in well over half of those cases).

Proposal to Provide Immediate, Limited Bargaining Rights
To speed up initial bargaining. I propose a statutory amendment that would impose a limited duty to bargain immediately upon a tally revealing that a majority of unchallenged voters cast their ballots in favor of union representation, provided that the union has not challenged a determinative number of ballots. Tallies would occur 22 days after the filing of the RC petition.

Upon such a tally, an employer would have to immediately bargain with the union on a members-only basis until the Board resolves the employer’s determinative challenges, election objections, and/or any dispute over the appropriateness of the unit. While waiting for a final legal determination, employers and unions would be required to sit at the table and bargain with open minds about union members' terms and conditions. When a final legal determination ultimately issues, the parties will have a head start on bargaining.

(A word about why I propose members-only bargaining during this interim period: while I support the Act's principle of majority rule, I think it would be unwise to permit a union to represent non-supporters until a final determination is made that a majority of employees have freely elected the union in an appropriate unit. Under my proposal, employers and non-member employees would be able to deal directly with each other.)

An exception to the limited bargaining duty would exist. No limited bargaining duty would arise, even if a tally favors the union, when the employer disputes unit appropriateness on grounds that the unit includes statutory supervisors, guards and non-guards, employees of multiple employers, etc. Thus, while an election can be held in a unit of putative supervisors, no duty to bargain will arise in such a unit unless and until the Board certifies the union.

Under my proposal, if and when the Board certifies the union, the employer must bargain with it as the exclusive representative of all unit employees. On the other hand, if and when the Board concludes that the unit is inappropriate, sustains election objections, and/or resolves determinative challenged ballots against the union, the limited duty to bargain would be extinguished.

The limited duty to bargain on a members-only basis would have limited benefit to employees if the law were to forbid employers and unions from reaching interim agreements. Thus, my proposal calls for the law to clearly permit parties to execute members-only CBA’s and to apply them to members only. The amended statute would also permit (but not require) an employer to offer non-members the same terms and conditions as exist under the interim CBA.

Skeptical employers might complain that unions will be able to secure limited bargaining rights by committing objectionable conduct. Skeptical unions might complain that employers will retain up to 21 days to coerce employees. To reduce coercion, the proposed statutory amendment would impose a fine on parties (employers and unions alike) found to have coerced employees during an organizing drive. The fines must have sufficient deterrent effect. Likewise, the amendment would impose a fine to deter parties from shirking their interim duty to bargain.

Skeptics of my proposal might also argue that it is unwise to require an employer to bargain with a putative union victor when, in some cases, the employer will ultimately prevail on its objections, challenges, or unit-appropriateness arguments. But, keep in mind that the Board overrules employers' objections in about 90% of RC cases. On balance, I think that the law would promote collective bargaining more effectively if it imposed a limited, temporary duty to bargain than to require employees to wait months or years after a valid election for initial bargaining to start.

I realize that interim, members-only bargaining will present issues. How, if at all, should the Board apply the unilateral change doctrine? What effect, if any, will an interim agreement have after certification? Professor Charles Morris, who advocates for a general duty to bargain on a members-only basis, has argued that the Board will be able to handle such issues effectively. Charles J. Morris, The Blue Eagle at Work: Reclaiming Democratic Rights in the American Workplace 236-37 (2005). I agree and will leave my thoughts on such issues for another day.

To conclude, I think my proposal has a number of advantages. It speeds up elections and initial bargaining. It retains the secret-ballot election, which ensures, to the greatest extent possible, that employees will record their choice free of coercion. Moreover, it ensures that all unit employees and the employer will have at least 7 days to speak non-coercively about the relative advantages or disadvantages of unionizing before employees cast their ballots. Finally, while I like the merits of my proposal, it also might provide some common political ground. I’m sure that it can be improved. I’m hoping to get some input.

Wednesday, July 23, 2008

Orrin Hatch on EFCA

Here is a some video from CNBC's Squawk Box of Senator Orrin Hatch criticizing the Employee Free Choice Act.

Monday, May 12, 2008

Agreements Not To Organize

As discussed at Workplace Prof Blog and Debris, the Wall Street Journal has reported on an agreement between SEIU and UNITE HERE on the one hand and two employers on the other that the unions won't seek to represent certain groups of those employers' employees. SEIU and UNITE HERE currently represent some of those employers' other employees.

The Board has given effect to an agreement not to organize a group of workers if "the promise be express, for a reasonable period of time and the result of bargaining between equals." Lexington House, 328 NLRB 894, 897 (1999) (citing Briggs Indiana Corp., 63 NLRB 1270 (1945). In Lexington House, the Board refused to process an RC petition where the petitioning union had promised the employer that it would not seek to represent the petitioned-for unit for 12 months. The Board found that the agreement's restriction was not contrary to the Act's policies.

It would be interesting to know what SEIU and UNITE HERE received in return for their promise. Increased wages for the currently represented groups? Better health care? Increased access to still other groups of unrepresented employees? It would also be interesting to know whether the unions had had any interest in organizing or plans to organize the off-limits group. I don't subscribe to the WSL, so can't access the full article.

Friday, March 30, 2007

More from Estlund and Hurtgen on the Employee Free Choice Act

After reading the written testimony to the Senate HELP Committee, I sent five questions to Cynthia Estlund and one question to Peter Hurtgen. They kindly agreed to let me post their answers.

Questions to Cynthia Estlund
1) How exactly would the EFCA's card-check provisions reduce employer coercion?

A: Card check limits the employer’s opportunity to wear down the employees over weeks or months, and it would limit the opportunity to extend that period through procedural maneuvers. (Union avoidance consultants, which most employers hire when faced with a union organizing drive, are quite candid that they often need several weeks or more to erode the union’s majority.) Because majority sign-up is a rolling process, the impact of the employer’s campaign may also be diffused by the fact that some of the employees the employer is targeting have already signed up. The Eaton & Kreisky study (cited in my testimony) shows that, one way or another, employees report much less pressure from any source in card check than in election campaigns (and much less pressure from the union than from the employer in either type of campaign).

2) Are you concerned that, under the EFCA, a law-abiding employer can lose the opportunity to engage in non-coercive speech on the disadvantages of unionization?

A: There generally is time for the employer to convey information. First of all, employers are entitled to (and some do) tell employees their views on unionization in general from the first day on the job. As for information about the particular union, except for some very small units, the card check campaign doesn’t stay secret from the employer.
In any event, employees don’t obviously have any less information about the union before signing a card than they generally have about the employer before taking a job. You can only learn so much before starting the job. So, too, the best way to learn about what it’s like to have a union is having a union. If you don’t like what you learn over time, and if a majority of your co-workers share your views, you can simply tell the employer that, and the employer can or even must withdraw recognition. When employees are dissatisfied with their employer, by contrast, their only real option is to quit (which of course is no less an option when employees are dissatisfied with the union).
Finally, let’s keep in mind that employees who actually have union representation overwhelmingly say they prefer to keep it – 87-90 percent. Opponents of card check often seem to assume that it is some huge and fateful step for employees to choose union representation. But we make a lot of choices in life that are harder to undo, and that have at least as important consequences, with a lot less information and certainly without having to undergo a highly adversarial campaign over its merits.

3) Would you support legislation that directly prohibits employers from engaging in non-coercive speech on the disadvantages of unionization?

A: Apart from constitutional and other problems with such a prohibition, I have doubts about its enforceability, and the impact on already horrendous delays at the NLRB. (If I were able to snap my fingers and enact labor law reform, I would be more inclined to limit employer’s ability to compel employees to submit to captive audience meetings and/or their ability to exclude union organizers from the workplace than to regulate what they can say.) EFCA takes a different, less interventionist approach that allows organizers and employees to choose a different campaign structure instead of further restricting employer conduct.

4) You declined to express an opinion on Section 3 of the EFCA, which addresses mandatory interest arbitration. Do you support Section 3?

A: I do. Interest arbitration is not an ideal end to collective bargaining. But it’s necessary response to a serious problem, and it’s a better spur to good faith bargaining than what current law provides. (And that is mostly how it works in Canada, where the very large majority of cases that could be subject to interest arbitration end in voluntary agreements.)

The situation that first contract arbitration is meant to address is familiar and pretty disastrous: Employers, after failing in their union avoidance effort, can continue the fight, stall and delay, aiming not toward an agreement but toward impasse, knowing that all the Board will do is order more bargaining, and all the employees can do is strike, which will expose them to permanent replacement and eventually decertification. (In this they are following the fairly standard advice of those near-ubiquitous union avoidance consultants.) Many employees who manage to surmount the employer’s first anti-union effort lose out in the second phase. They end up with no contract and no tangible gains from their hard-fought campaign; after a year or two or three, demoralization and turnover often doom the union’s majority. It’s a very sad situation that calls for a significant change in the law.

5) Do you think that the EFCA's new remedies are sufficient to adequately deter employer coercion? Or, do you think that even greater remedies are required?

A: First, a lot of conduct that is legal is very coercive, especially over time and in the very intense form that it is often engaged in, given the employer’s overwhelming power over employees and the workplace. But even looking just at illegal forms of employer coercion, the new remedies are probably not a sufficient deterrent. The penalties are likely to be used sparingly. Treble backpay will mainly affect one especially coercive tactic – discriminatory discharges – and would still make the average backpay award under $12,000. That is surely not enough to deter the determined employer. Injunctive relief, and esp. preliminary reinstatement of fired activists, is important and potentially very helpful in undoing some of the harm to the organizing campaign.

The problem is that many employers seem to believe (wrongly, I think) that the fate of their business depends on keeping the union out (and their highly-paid consultants fan the flames of anti-union sentiment). It’s hard to know what it would take to adequately deter employers in that situation from taking advantage of their managerial prerogatives, and their ability to predict terrible consequences for the workplace as a whole (which they will be around to help bring about if they “lose”), to induce employees to vote “no.” That’s the thinking behind EFCA’s creation of the alternative structure of majority sign-up or card check. It’s not a perfect solution, but I doubt there is one.

The challenge is to change the labor relations climate – social norms, in part – so that more employers (like Cingular, Costco, Kaiser-Permanente) decide simply to deal with the fact of collective bargaining, and to normalize unionization again, so that it becomes one way of moving forward instead of a tooth and nail fight.

Question to Peter Hurtgen
Q: You declined to express an opinion on Section 4 of the EFCA, which provides for new remedies. Do you support or oppose Section 4? Do you think that the NLRA’s current remedies are adequate to effectively deter employer coercion?

A: I oppose section 4 and I believe present Board remedies are adequate or could be without an amendment to the Act.

Monday, March 26, 2007

Senate Hearing on Employee Free Choice Act

The Senate Committee on Health, Education, Labor, and Pensions is holding a hearing on the Employee Free Choice Act on Tuesday, March 27 at 9:30 a.m.

The witnesses will be:
Errol Hohrein, an employee who participated in an organizing drive
Cynthia Estlund, Professor, NYU Law School
Lawrence Mishel, President, Economic Policy Institute
Peter Hurtgen, Morgan, Lewis & Bockius LLP, former Chairman and Member of the NLRB, and former Director of the FMCS