Service Employees International Union, AFL-CIO v. Labor Relations Commission

729 N.E.2d 1100, 431 Mass. 710, 2000 Mass. LEXIS 351, 168 L.R.R.M. (BNA) 2486

• Court: Massachusetts Supreme Judicial Court
• Decided: June 15, 2000
• Status: Published

Opinion

Spina, J.

In this case the Labor Relations Commission (commission) decided by a divided vote that a public employer did not violate its duty pursuant to G. L. c. 150E, §§ 6, 10, to bargain in good faith with the exclusive representative of its employees when the employer surveyed the employees about their use of sick leave. Commonwealth of Mass., 25 M.L.C. 48 (1998). We vacate the order of the commission. We hold that a public employer may not survey its employees about mandatory subjects of collective bargaining if the employees belong to a bargaining unit represented by a union at a time when the union is engaged or preparing to engage in collective bargaining with the employer. We do not decide whether such a survey would be permissible at other times.

1. Facts and prior proceedings. The pertinent facts are undisputed. The survey in this case asked employees how many sick leave days they took each year; whether they took sick leave several days at a time or one day here, one day there; what circumstances generally led them to take sick leave; how they felt about employees who malingered; what would make them less likely themselves to malinger (for instance, whether requiring a doctor’s note after a certain number of absences would improve matters); and what incentives they thought would encourage better attendance and discourage malingering. The survey also invited “general comments.” See id. at 49.

A memorandum accompanied the survey. The memorandum said that the “Administration” (presumably that of then Governor Weld) was determined either to “impos[e] methods aimed at reducing excessive sick leave use” or to eliminate sick leave altogether. The memorandum described the surveyed employees’ sick leave usage as “among the worst in the entire [S]tate.” It said that the average number of days used by employees was eleven, that the “desirable” number was “5-6,” and that it was “imperative” to improve the former number “as soon as possible.” The survey appears to have been largely unsuccessful. Of the few employees who responded to it, a good many did so frivolously. See id.

Some of the surveyed employees belonged to a bargaining unit represented by the plaintiff union. At the time the survey was distributed, the union was negotiating with the Commissioner of Administration and Finance for the Commonwealth in order to arrive at a new collective bargaining agreement for the unit. Sick leave and sick leave use were major subjects of negotiation. During discussions, the union and the commissioner exchanged a number of proposals about sick leave policy. See id.

Some months after the negotiations ended, changes were announced in the procedures by which sick leave was reported and verified. Both the union and employees in the bargaining unit objected to the announced changes. The changes were not made. See id. at 50.

The union charged the commissioner with having committed a practice prohibited by G. L. c. 150E, § 10. The commission issued a complaint against the commissioner on the charge and held a hearing on the complaint. After an administrative law judge issued recommended findings of fact,¹ the commission adopted the findings and dismissed the complaint, holding that the survey did not constitute “direct dealing” forbidden by G. L. c. 150E. See id. at 50-51.

As it often does in construing G. L. c. 150E, the commission relied for guidance on decisions of the National Labor Relations Board (NLRB) construing the National Labor Relations Act (NLRA), 29 U.S.C. §§ 151 et seq.² According to the commission, the NLRB determines whether a given survey constitutes direct dealing by asking whether the “timing and nature” of the survey are “sufficiently linked” to continuing negotiations to warrant the conclusion that “the employer circumvented a union at the bargaining table and dealt directly with employees represented by the union.” Commonwealth of Mass., supra at 50. The commission rejected the union’s argument that intent to undermine a union’s bargaining position is not a necessary element of direct dealing. See id. at 51 & n.3, citing Lawrence Sch. Comm., 3 M.L.C. 1304 (1976); Blue Hills Regional Sch. Dist. Comm., 3 M.L.C. 1613 (1977).

After examining the timing and nature of the sick leave usage survey, the commission concluded that the record did not “establish a direct connection” between the survey and “a position the Commonwealth sought to advance at the bargaining table.” Commonwealth of Mass., supra at 51. As evidence for this conclusion, the commission stated that (1) the drafters of the survey had not known what the union and the commissioner were discussing in negotiations; (2) the purpose of the survey was to obtain information as to how to reduce excessive resort to sick leave; (3) no evidence showed that the survey referred to specific bargaining proposals or “questioned employee[s] about specific issues that were the subject of the parties’ negotiations”; (4) no “direct” evidence showed that the commissioner intended to use survey results as a basis for bargaining proposals; (5) employees’ response to the survey was minimal and largely frivolous. Id.

The commission likened the survey in this case to the survey in Logemann Bros., 298 N.L.R.B. 1018 (1990), which asked employees for suggestions on how to improve plant efficiency. The NLRB held that the survey was motivated by a legitimate business concern on the part of the employer and that the survey was akin to a suggestion box arrangement. See Commonwealth of Mass., supra at 51.

The commission appears to have made its decision on two principal grounds: that the survey was unrelated to “specific” proposals or issues to be discussed in negotiations between the commissioner and the union; and that the employer did not intend to use the survey to “arm itself for negotiations.” Id. at 51. See id. (survey was “motivated by the legitimate business concern of reducing sick leave usage”). The commission did not explain how the first ground of decision was compatible with its factual finding that sick leave was a subject of discussion in the negotiations.

The dissenting commissioner was of the view that direct dealing does not merely include “specific employer intent to erode the Union’s bargaining position over a particular proposal then pending during negotiations.” Id. She would have held instead that surveys that ask about subjects of negotiation between the employer and the union constitute direct dealing if they are conducted when the parties are negotiating or preparing to negotiate. See id. at 53.

The union sought review of the commission’s decision in the Appeals Court pursuant to G. L. c. 150E, § 11. We transferred the case here sua sponte.

2. Public employers’ surreys of employees represented by a union pursuant to G. L. c. 150E. Although we give weight to the commission’s experience and authority, see G. L. c. 30A, § 14, we do not affirm commission decisions that are inconsistent with law. See G. L. c. 30A, § 14 (7) (c); Plymouth v. Civil Serv. Comm’n, 426 Mass. 1, 5 (1997); Boston Police Superior Officers Fed’n v. Labor Relations Comm’n, 410 Mass. 890, 892 (1991) (no deference may be paid to commission’s decisions when we review for error as a matter of law). See also Massachusetts Community College Council v. Labor Relations Comm’n, 402 Mass. 352, 353 (1988).³

In interpreting G. L. c. 150E, the commission has often relied on Federal decisions applying parallel provisions of the NLRA. See Burlington v. Labor Relations Comm’n, 17 Mass. App. Ct. 402, 405 (1984); Blue Hills Regional Tech. Sch. Dist., 9 M.L.C. 1271, 1276 (1982). Cf. City Manager of Medford v. State Labor Relations Comm’n, 353 Mass. 519, 522-523 (1968). Although we have held in particular cases that the differences between G. L. c. 150E and the NLRA warranted departures from Federal decisions interpreting the latter enactment, see Graham v. Quincy Food Serv. Employees Ass’n & Hosp., Library & Pub. Employees Union, 407 Mass. 601, 612-613 & n.9 (1990), we perceive no reason for such a departure in this case. See Trustees of Forbes Library v. Labor Relations Comm’n, 384 Mass. 559, 562 n.2 (1981).

Section 6 of G. L. c. 150E imposes upon the public employer the obligation to negotiate in good faith with the union that the members of a collective bargaining unit have chosen as the unit’s exclusive representative. Section 10 (a) (5) in turn makes it a prohibited practice for an employer to refuse to discharge the duty imposed on him by § 6.⁴ This duty necessarily entails the duty to refrain from circumventing the union by dealing directly with bargaining unit employees as to mandatory subjects of negotiations. See Town of Randolph, 8 M.L.C. 2044, 2052 (1982) (duty to refrain from direct dealing is “corollary” to employer’s duty to bargain with exclusive representative), cited in Trustees of the Univ. of Mass. Medical Ctr., 26 M.L.C. 149, 160 (2000); Lawrence Sch. Comm., 3 M.L.C. 1304, 1311 (1976); Medo Photo Supply Corp. v. NLRB, 321 U.S. 678, 684 (1944). Cf. G. L. c. 150E, § 5 (exclusive representative is empowered “to act for and negotiate agreements covering all employees in the unit”). To define the limits of direct dealing is therefore to do no less than to define the meaning of the exclusive representation concept, a basic building block of labor law policy under G. L. c. 150E.

Direct dealing is impermissible for at least two related reasons. First, direct dealing violates the union’s statutory right to speak exclusively for the employees who have elected it to serve as their sole representative. This right necessarily includes the power to control the flow of communication between the employer and the represented employees concerning subjects as to which the union is empowered to negotiate.⁵ Second, direct dealing undermines employees’ belief that the union actually possesses the power of exclusive representation to which the statute entitles it. See Americare Pine Lodge Nursing & Rehabilitation Ctr. v. NLRB, 164 F.3d 867, 875 (4th Cir. 1999) (“direct dealing is characterized by actions that persuade employees to believe that they can achieve their objectives directly through the employer and thus erode the union’s position as the exclusive bargaining representative”).

These justifications for the direct dealing prohibition also provide guidance in delineating its limits. As courts and the NLRB have repeatedly held, an employer does not violate the NLRA by merely communicating its bargaining position to its employees. See Americare Pine Lodge Nursing & Rehabilitation Ctr., supra at 875-876, and cases cited; Obie Pac., Inc., 196 N.L.R.B. 458, 459 (1972). See also 29 U.S.C. § 158(c) (“The expressing of any views, argument, or opinion . . . shall not constitute or be evidence of an unfair labor practice [under the NLRA], if such expression contains no threat of reprisal or promise of benefit”). Cf. Boston Sch. Comm., 6 M.L.C. 1363, 1370 (1979). Counsel for the union acknowledged at oral argument that a survey that would otherwise amount to direct dealing is not forbidden if the union consents to it. See Toledo Typographical Union No. 63 v. NLRB, 907 F.2d 1220, 1222 (D.C. Cir. 1990), cert. denied, 498 U.S. 1053 (1991). Neither of these kinds of communication threatens the union’s role as exclusive representative of the members of the bargaining unit in dealings with the employer.

Surveys of employees as to mandatory subjects of negotiation are a different matter, at least if bargaining discussions have begun or are expected to begin. Exchanging information about employees’ views on these subjects is a crucial element of any negotiation between an employer and a union. Employers who solicit this information directly from employees vitiate the union’s role as the exclusive voice of employees in negotiations. They also obtain a valuable index of employees’ willingness to consider this or that combination of bargaining terms from the employer. See Obie Pac., Inc., 196 N.L.R.B. 458, 459 (1972) (“Part of the task facing a negotiator ... is effectively to coalesce an admixture of views of various segments of his constituency, and to determine, in the light of that knowledge, which issues can be compromised and to what degree. A systematic effort by the other party to interfere with this process by either surreptitious espionage or open interrogation constitutes clear undercutting of this vital and necessarily confidential function of the negotiator”). The questions in the survey need not correspond to proposals actually raised in negotiations in order to have this effect.⁶

The commission implicitly held that intent to erode the bargaining position of the union is a necessary element of direct dealing. See Commonwealth of Mass., 25 M.L.C. 48, 51 & n.3 (1998). We disagree, for two reasons. First, intent of this kind is extraordinarily difficult to prove. We do not think that the General Court intended to require unions to shoulder such a heavy burden of proof in defending the very integrity of their status as exclusive representatives of the bargaining units that elected them. Second, unions are harmed by direct dealing regardless of whether it is intended by the employer. See Lawrence Sch. Comm., 3 M.L.C. 1304, 1312 (1976) (direct dealing “undermines the effectiveness of the bargaining representative in general”); Waddell Eng’g Co., 305 N.L.R.B. 279, 281 (1991) (“Whether intended or not, even absent union animus, such conduct tends to undermine the status of the employee representative and is inconsistent with the principles of good-faith bargaining”). When direct evidence of an employer’s intent to gain an advantage in negotiations is available, this evidence can certainly be relevant to whether the employer’s activity was direct dealing. Cf. Leland Stanford Jr. Univ., 240 N.L.R.B. 1138, 1138 n.1 (1979). But evidence of this sort is not necessary. An employer’s communication with its employees is direct dealing if “its purpose or effect” is “the erosion of the [ujnion’s status as exclusive bargaining representative” (emphasis added). E.I. Dupont de Nemours & Co., 301 N.L.R.B. 155, 156 (1991). Employers are responsible for the foreseeable consequences of surveys regardless of their subjective intent. Cf. City of Boston, 20 M.L.C. 1603, 1607 (1994) (“Unilateral changes in mandatory subjects of bargaining are per se violations of [§] 5 of the Law and the employer’s good faith is not relevant”).

The commission’s opinion cites two early commission decisions for the proposition that “the notion of direct dealing outlined in our prior decisions contemplates intentional action by an employer.” Commonwealth of Mass., 25 M.L.C. 48, 51 n.3 (1998), citing Lawrence Sch. Comm., 3 M.L.C. 1304 (1976); Blue Hills Regional Sch. Dist. Comm., 3 M.L.C. 1613 (1977). The former case, if anything, suggests the contrary of this proposition. See Lawrence Sch. Comm., supra at 1311-1312. The latter case has to do with a unilateral change in a subject of negotiation, not with direct dealing. It too contains language that suggests that the employer’s intent is irrelevant to whether the employer’s conduct is permissible. See Blue Hills Regional Sch. Dist. Comm., supra at 1615 (“The employer must be held accountable for the for[e]seeable consequences of its action”). We therefore reject the commission’s interpretation of these decisions.⁷

Applying the principles we have articulated to the survey issued in this case, we conclude that the survey was unlawful. Paid leave generally, including sick leave, is a term or condition of employment and is thus a mandatory subject of collective bargaining under G. L. c. 150E, § 6. See City of Gloucester, 26 M.L.C. 128, 129 (2000); Town of Hull, 19 M.L.C. 1780, 1784 (1993); NLRB v. Katz, 369 U.S. 736, 744 (1962). So are policies that determine the availability of leave for employees, such as the method whereby an employee is required to verify his claim of eligibility for leave. See Commonwealth of Mass., 22 M.L.C. 1459, 1462-1463 (1996); City of Springfield, 12 M.L.C. 1051, 1054 (1985); NLRB v. BASF Wyandotte Corp., 798 F.2d 849, 852-853 n.1 (5th Cir. 1986); Southern Cal. Edison Co., 284 N.L.R.B. 1205, 1210-1211 (1987). Cf. Town of Weymouth, 11 M.L.C. 1448, 1455 (1985) (whether employee must undergo examination by employer’s physician in order to obtain disability leave is mandatory subject of bargaining).⁸ The employer issued the survey at a time when the union and the employer were engaged in collective bargaining. It follows that the survey was direct dealing. Because intent is not a necessary element of direct dealing, it is not relevant to our decision that the employees who devised the survey had little if anything to do with the collective bargaining negotiations in which sick leave was discussed, that the survey was never discussed during the negotiations, and that the survey appears to have wholly failed at achieving the purposes for which it was devised. These considerations may affect the commission’s choice of remedies for a direct dealing violation. See Millis Sch. Comm., 23 M.L.C. 99, 100 (1996) (noting that commission has “broad authority to fashion appropriate orders to remedy unlawful conduct”); Barnstable Sch. Comm., 5 M.L.C. 1358, 1361 (1978); Shenango Steel Bldgs., 231 N.L.R.B. 586, 586, 589 (1977). They do not affect whether direct dealing has occurred in the first place. See Lawrence Sch. Comm., 3 M.L.C. 1304, 1311 (1976) (“If the discussions concerned a proper subject of collective bargaining, the Committee . . . violated Section 10[a][5]”). Cf. Harris-Teeter Super Mkts., Inc., 310 N.L.R.B. 216, 217 (1993) (“it [is] of no consequence” to the direct dealing determination “that the Respondent did not change its position after hearing the employees’ negative reactions to the proposal”).

National Treasury Employees Union v. Federal Labor Relations Auth., 826 F.2d 114 (D.C. Cir. 1987), is authority to the contrary. That case, however, involves the interpretation of the Federal public sector collective bargaining statute, 5 U.S.C. §§ 7101-7135, which differs considerably from both the NLRA and G. L. c. 150E. The preamble to the Federal statute, 5 U.S.C. § 7101, requires that it be interpreted to further both the rights of unions as exclusive representatives of employees and the cause of efficiency in government. See National Treasury Employees Union v. Federal Labor Relations Auth., supra at 116, 120-123. The agency that administers the statute is given considerable discretion “to reconcile the tensions that will at times exist” between these “twin objectives.” Id. at 121. Like the NLRA, see id. at 122, G. L. c. 150E strikes a different balance. Cf. Library of Congress v. Federal Labor Relations Auth., 699 F.2d 1280, 1287 (D.C. Cir. 1983).

We disagree with the commission’s conclusion that the facts of Logemann Bros., 298 N.L.R.B. 1018 (1990), are closer than those of Harris-Teeter Super Mkts., 310 N.L.R.B. 216 (1993), to the facts of this case. See Commonwealth of Mass., 25 M.L.C. 48, 50-51 (1998). In Harris-Teeter Super Mkts., the employer orally asked employees for comments on whether scheduled work days should be changed: a mandatory subject of collective bargaining. The employer was in the middle of negotiations with the union at the time the solicitation occurred. The Board held that it was impermissible to “solicit[] the sentiment of the employees on a subject to be discussed at the bargaining table.” Id. at 217. The survey in Logemann Bros, told employees: “Do you think there are better ways to do the job? How about our equipment and machines? Can we do things ‘smarter’?” Id. at 1019. The employer used employees’ survey responses in its bargaining sessions with the union. See id. at 1018. The NLRB declined to invalidate the survey, holding that the questions in the survey were “general and innocuous” and that the survey was an “open-ended solicitation for employees’ suggestions.” Id. Even though the question about equipment in the Logemann Bros, survey referred to working conditions, a mandatory subject of bargaining, the reference lacked specificity. Surveys or suggestion boxes that ask in highly general terms, if at all, about mandatory subjects of collective bargaining are quite different from surveys of the sort that was distributed in this case. “Given the timing and the nature of the . . . survey, it cannot be deemed a mere general survey -— unrelated to collective bargaining — of employee views” (footnotes omitted). Alexander Linn Hosp. Ass’n, 288 N.L.R.B. 103, 106 (1988), enforced in relevant part sub nom. NLRB v. Wallkill Valley Gen. Hosp., 866 F.2d 632 (3d Cir. 1989).

The union asks us to hold that surveys on mandatory subjects of collective bargaining are unlawful even when no negotiations are occurring or expected to occur. The union notes that its role as exclusive representative of employees continues not only during negotiations but throughout the life of the collective bargaining agreement. See Higher Educ. Coordinating Council, 25 M.L.C. 37, 40 (1998). We decline to decide this question. Compare Harris-Teeter Super Mkts., 310 N.L.R.B. 216, 217 (1993); East Tenn. Baptist Hosp., 304 N.L.R.B. 872, 873 (1991), enforcement denied in part on other grounds, 6 F.3d 1139 (6th Cir. 1993).

We acknowledge that the commission, “in proceeding on a case-by-case basis, should be permitted to make refinements and even new rules in light of past experience.” West Bridgewater Police Ass’n v. Labor Relations Comm’n, 18 Mass. App. Ct. 550, 555 (1984). Yet new rules and refinements on old rules may not be fashioned at the expense of governing statutes and their basic policies. General Laws c. 150E makes the union the exclusive representative of the employees in its bargaining unit. To allow an employer to create other channels of communication with employees as to mandatory subjects of collective bargaining would be to make the union an exclusive representative of employees in name only. Cf. Anderson v. Selectmen of Wrentham, 406 Mass. 508, 512-513 nn.8-9 (1990).

Because the Labor Relations Commission erred as a matter of law, the order of the commission dismissing the complaint is vacated. The case is remanded to the commission for proceedings consistent with this opinion. See Boston Police Superior Officers Fed’n v. Labor Relations Comm’n, 410 Mass. 890, 894 (1991); County of Suffolk v. Labor Relations Comm’n, 15 Mass. App. Ct. 127, 133 (1983).

So ordered.

1 The chief counsel of the commission, who represents the commission in this appeal, served as administrative law judge in this proceeding. The union has not challenged the propriety of this practice, so we do not decide the issue, but we note that obvious problems would arise if the commission had disagreed with the findings that the chief counsel made in his capacity as administrative law judge. Cf. Labor Relations Comm’n v. Fall River Educators’Ass’n, 382 Mass. 465, 470 (1981).

2 The commission had never decided in a published opinion whether a survey of employees represented by a union constitutes direct dealing.

3 Our task of reviewing the commission’s decisions would be eased if the commission, or the commercial publisher that reports its decisions, were to publish a list of past decisions that subsequent decisions (whether by the commission or by courts) have overruled. Neither Lexis nor Westlaw, it would seem, includes the decisions of the commission in its otherwise ample selection of databases. The Social Law Library provides a searchable database of most of the commission’s decisions, but this does not include anything like Lexis’s Shepard’s system (or Westlaw’s KeyCite system) for checking citations. Cf. Price v. Cole, 31 Mass. App. Ct. 1, 6 (1991). The commission bears some responsibility for ensuring that litigants and reviewing courts are able to determine what the current law is. The commission may wish to consider petitioning the General Court for funds for this purpose.

4 Numerous commission decisions have characterized violations of G. L. c. 150E, § 10 (a) (5), as “derivative” violations of § 10 (a) (1), which forbids public employers from interfering, restraining, or coercing any employee in the exercise of any right guaranteed under c. 150E. See, e.g., Commonwealth of Mass., 25 M.L.C. 48, 51 (1998); City of Everett, 2 M.L.C. 1471, 1475 (1976), aff’d, Labor Relations Comm’n v. Everett, 1 Mass. App. Ct. 826 (1979). Sections 10 (a) (1) and 10(a)(5) are parallel to §§ 8(a)(1) and 8(a)(5), respectively, of the NLRA. Federal case law makes clear that a violation of § 8(a)(5) is necessarily a violation of § 8(a)(1). See Microimage Display Div. v. NLRB, 924 F.2d 245, 250 (D.C. Cir. 1991); G. Heileman Brewing Co. v. NLRB, 879 F.2d 1526, 1533 (7th Cir. 1989); NLRB v. Swedish Hosp. Medical Ctr., 619 F.2d 33, 35 (9th Cir. 1980); NLRB v. Newark Morning Ledger Co., 120 F.2d 262, 265, 267 (3d Cir.), cert, denied, 314 U.S. 693 (1941). Overnite Transp. Co. v. NLRB, 372 F.2d 765, 769 (4th Cir.), cert, denied, 389 U.S. 838 (1967), which is authority to the contrary, appears to rely on a misreading of n.9 in NLRB v. Katz, 369 U.S. 736, 742 (1962).

5 General Laws c. 150E, § 5, creates a narrow exception to this general rule. “An employee may present a grievance to his employer and have such grievance heard without intervention by the exclusive representative of the employee organization representing said employee, provided that the exclusive representative is afforded the opportunity to be present at such conferences and that any adjustment made shall not be inconsistent with the terms of an agreement then in effect between the employer and the exclusive representative.”

6 Surveys that constitute “direct dealing” with employees as to subjects of collective bargaining are distinguishable from surveys that attempt to undermine a union’s efforts to be recognized or retained by official election as a bargaining unit’s exclusive representative. The former violate § 8(a)(5) of the NLRA and derivatively violate § 8(a)(1) of the NLRA. The latter independently violate § 8(a)(1) of the NLRA. See, e.g., Tom Wood Pontiac, Inc., 179 N.L.R.B. 581, 581 (1969) (survey intended to discover “to learn what complaints the employees had which might cause them to want a union” infringed employees’ rights under NLRA), enforced, 447 F.2d 383 (7th Cir. 1971). What we say in this case has nothing to do with the second kind of survey. We note, however, that some surveys fall within both these categories. See, e.g., Northwest Pipe & Casing Co., 300 N.L.R.B. 726, 733 (1990); St. Joseph’s Hosp., 247 N.L.R.B. 869, 877 (1980).

7 As for decisions of the NLRB, the commission cited only five of those in its opinion. At oral argument, counsel for the commission stated that the five decisions were the only direct dealing cases involving employee surveys that the NLRB had decided in its history. This assertion is manifestly erroneous. Numerous NLRB decisions on surveys claimed to be direct dealing are cited throughout this opinion.

Some of the decisions on whether particular surveys amount to direct dealing appear to assume or imply that intent to harm the union is a necessary element of direct dealing. See, e.g., Vans Grocery Co., 320 N.L.R.B. 53, 56 (1995); Laminated Prods., Inc., 294 N.L.R.B. 816, 820 (1989); United Techs. Co., 274 N.L.R.B. 1069, 1070-1071 (1985), enforced on other grounds sub nom. NLRB v. Pratt & Whitney Air Craft Div., 789 F.2d 121 (2d Cir. 1986); Obie Pac., Inc., 196 N.L.R.B. 458, 458-459 (1972); Continental Oil Co., 194 N. L.R.B. 126, 126, 131-132 (1971). Cf. West Hartford Educ. Ass’n v. De-Courcy, 162 Conn. 566, 593 (1972). Others do not. See, e.g., Wyandanch Engine Rebuilders, Inc., 328 N.L.R.B. 119 (1999); Detroit Edison Co., 310 N.L.R.B. 564, 565, 572-573, 576 (1993); American Commercial Lines, 296 N.L.R.B. 960 (1989), enforced in relevant part, Inland Tugs v. NLRB, 918 F.2d 1299 (7th Cir. 1990); Bob’s Big Boy Family Restaurants, 264 N.L.R.B. 432, 434 (1982); M.A. Harrison Mfg. Co., 253 N.L.R.B. 675, 675, 680, 684-685 (1980), enforced, 682 F.2d 580 (6th Cir. 1982), overruled on other grounds by Unbelievable, Inc., 318 N.L.R.B. 857, 859 (1995). Not a few of these decisions are unclear on the point. Many contain passages that taken alone could be read to mean the opposite of the proposition for which we have cited them. See, e.g., Obie Pac., Inc., supra. Some of the decisions speak of intent but by this seem to mean merely intent to deal directly. Still others proceed to “show” intent by mere evidence of the timing and nature of a survey.

8 Indeed, the Commissioner was obliged to give the union notice and an opportunity to bargain before any changes could be made in sick leave policy. See, e.g., Town of Hull, 19 M.L.C. 1780, 1784 (1993); School Comm. of Newton v. Labor Relations Comm’n, 388 Mass. 557, 572 (1983), citing, inter alia, NLRB v. Katz, 369 U.S. 736 (1962).