By Jim Cline
Every arbitration contains lessons to apply to contract negotiations. Sometimes the lessons involve an extension of standards ways that arbitrators approach issues, and sometimes they involve unique situations or novel arguments on existing issues.
In the last issue of our Newsletter, we detailed the January 28 Island County Corrections award by Arbitration Michael Cavanaugh. Cavanaugh’s decision reinforced some generally understood patterns that we should be mindful of. But he also approached unique and developing issues in ways that were less easy to predict and may provide helpful guidance moving forward.
In this article, we will identify and discuss at least 15 of the lessons that can be learned from Cavanaugh’s decision and his reasoning. Many of these include how to expect arbitrators to resolve certain questions and how those may apply to your current contract negotiations. We will conclude with some general lessons that apply to how parties should prepare for and approach arbitrations (and contract negotiations in an interest arbitration context).
1. Historic comparables matter — but less so with the passing of time.
Arbitrators have repeatedly stated that historic comparables, whether rendered by a prior arbitration award, or by the parties’ agreement in negotiations, should be provided considerable weight. In this case, Arbitrator Cavanaugh also demonstrated that the passing of time and shifting demographic data might render those “historic” comparables less controlling. Parties seeking to deviate from historical comparables should be prepared to present solid and clear data, including how the patterns have changed over time.
2. Plus 100%/-50% comparability selection range is viable — when properly presented and explained.
This issue has been a recurring one over the history of the state interest arbitration statute. Knowledgeable union advocates have consistently urged the application of a 2-1 selection range. Applying this range results in the data being evaluated on a band that is 100% higher than the target jurisdiction and 50% lower. This percentage band results in jurisdictions that are “twice as large and half as small.”
In the early years of the statute, law enforcement units were generally represented by the Teamsters who uncritically accepted the management-friendly “+50%/-50%” approach. On occasion, some out-of-state union advocates also followed this formula. The critical issue here is that by only going up 50%, many fewer favorable matches for the union get swept into the pool of potential comparables. Without getting too technical, there is a sound statistical reason this occurs, a mathematical argument that many arbitrators began to grasp.
Arbitrator Cavanaugh has previously been on record with the clear majority on this issue. While a minority of arbitrators have been persuaded to follow the management-friendly test that only goes up 50%, as indicated, this has often been the product of unions not effectively presenting the rationale for the 2-1 range.
3. Geographic proximity matters — but cannot overcome significant size disparities.
In this case, the Guild presented a “reach” case to include a highly favorable Whatcom County comparable. But on both population and tax base factors, Whatcom County is well outside the generally accepted 100% upper band. The Guild was left arguing that an employee could plausibly commute between either Bellingham and Coupeville, and that, as a result, they shared a common labor market. Cavanaugh credited that argument but still concluded that the disparities in size were too significant.
Arbitrators have reached beyond the 2-1 band usually when two conditions were present — additional comparables were necessary to create a reasonably sized “set,” and the jurisdictions resided in a well-recognized common labor market. The Tri-Cities is a commonly cited example of this type of labor market. Whatcom and Island County are geographically close but do not share an adjacent border, so the Guild argument here was more of a stretch.
4. Labor market will receive less consideration than comparables.
Despite finding that Whatcom County was not a comparable, Cavanaugh concluded that Whatcom County would still be “relevant” as part of a labor market consideration. But it is not clear that he actually applied Whatcom County’s higher wages in rendering his decision. This pattern of ruling a proposed comparable is not a “comparable,” but then stating that it would be weighed as an “other factor” has occurred a number of times before. Each time, it seems that the arbitrator effectively provides that local but rejected comparable little, if any, weight.
5. Straightforward wage analysis is more persuasive than “total compensation” or other complex calculations.
Parties may have some legitimate and logical reason for arguing total compensation data, but Arbitrator Cavanaugh followed the dominant path of previous decisions — “total comp” data generally receives little, if any, consideration by arbitrators.
The problem with “total compensation” analysis is that it is often presented in too complex a manner to track all the internal calculations. Another problem is that there is no agreement on what constitutes “total” compensation, leading to the inclusion or exclusion of arguable elements. And perhaps unsurprisingly, when these models are presented in arbitration, they almost invariably seem to include or exclude those elements that improve the calculation of the proponent. (Perhaps this is an extension of the human nature principle that people only tell you what they want you to know!)
This pattern of self-serving wage reports has led our firm, as we did here, to disparage the end product by pejoratively labeling it “partial” total compensation. And in this case, as we have seen previously, the vast number of “moving parts” in these calculations seem to produce a significant number of entry and calculation errors that we’ve effectively identified as a basis to reject the end product. Rebutting these charts is even more persuasive when you can acknowledge, as here, that several of the errors were in the union’s favor (even when the result was not).
6. Employers claiming a limited ability to pay have a difficult to surmount burden.
In the current economic situation, it is unlikely that a Washington public employer could claim a legitimate limited ability to pay argument. Fueled with unanticipated sales tax revenues and American Rescue funding, local government reserves are generally much higher than necessary for budget planning needs.
Island County attempted to make an argument about its low “levy rate.” The issue here is a technical one stemming from the complex Washington property tax system. The Guild argued that the County’s argument turned reality on its head and that the levy rate was “low” only because the County’s property tax base was high — the 4th highest per capita assessed valuation in the state.
Cavanaugh provided brief space to what never amounted to a credible claim. His discussion about the unwillingness of taxpayers to raise the levy not being controlling is an interesting one that will be cited in future cases.
7. Wage differentials among county deputy classifications may warrant closer attention in negotiations and arbitration.
The Corrections Guild presented a detailed analysis of the differential between Correction and Patrol Deputies, something generally overlooked in most county arbitration hearings. Arbitrator Cavanaugh seems to have given very great weight to this data. What was compelling wasn’t just the current contract differential but how it had grown substantially over time as the Patrol deputies scored increases in some of the post-2008 contract years where the CO wages were frozen.
A review of statewide county contracts reveals no well-defined pattern of differentials. It’s apparent that county negotiations generally focus on external comparables. One lesson from this case is that future negotiations may involve more scrutiny of internal settlement patterns. Cavanaugh also noted that in two of the counties, the CO union had expressly attached future wage increases to patrol deputy wages.
8. CPI will be an important factor — but may result in unpredictable weight.
Ever since last summer’s unexpected large spike in inflation, we have predicted volatility in settlement patterns. As Cavanaugh noted here, one factor is uncertainty about how long the higher CPI trend would continue. Another important factor is that many contracts settled — usually low — without the benefit of seeing the summer 2021 CPI spike.
We’ve approached 2022 negotiations with the presumption that summer 2021 increases were the ones most relevant to 2022 contracts. This approach follows the practice in contracts that contain CPI formulas to apply CPI on a lagging basis (to the following year). Unusually, here Arbitrator Cavanaugh attempted to apply the year-end CPI formula to the year of the wage increase. That is an interesting approach, and there is some logic to it if you think of the wage adjustment as designed to correct for current year inflation. But this approach doesn’t follow the traditional pattern in contract negotiations. In fact, it has the practical limitation of the CPI number now being “knowable” until long after negotiations have begun.
Regardless, another takeaway here is that Cavanaugh’s determination that 4% was fair for 2022 might lead some employers to resist increases above that level. But the most recent and emerging pattern of contract settlements suggests that position is not viable. Cavanaugh had to face the reality that half of the comparables had settled on an average close to 2%, and that 4%, as he expressed it, was twice that level of increase.
9. Settlement trends in the Covid period are uneven and may be less controlling than normal.
Related to the volatility created by the unexpected turns in CPI is the variance in settlement patterns. As Kate Kremer and I recently discussed in our webcast, 2021-22 settlements have been greatly impacted by their timing. Were they before or after the pandemic? Before or after the economy recovered from the pandemic? Before or after the June 2021 CPI report?
Generally, arbitrators have given significant weight to comparable settlement patterns. The reason is easy to see – even if there is no argument for “catch up,” the settlement pattern is what’s needed just to “keep up.” Cavanaugh here appears to indicate that in this climate, settlement patterns should be given reduced consideration. He may be right, and it will be interesting to see how other arbitrators approach this question.
10. Staffing and recruitment issues will receive consideration — but not as much as you might anticipate.
The written decision discussed staffing and recruitment/retention issues that the Island County Jail faced, but not to the extent that actual data would have warranted. The Island County Jail had severe recruitment and retention issues. It was forced to change work schedules simply to avoid an unmanageable amount of overtime and considered provisional hires. In the context of the evidence presented in this hearing, the arbitrator could have cited that evidence and issued an even larger wage increase. The fact that he did not underscores what seems to have transpired in other recent arbitrations: Arbitrators have noted that some recruitment and retention issues exist but have seemingly not been very persuaded by that in generating the wage award.
No doubt, unions need to continue to press this issue in negotiations and arbitration. But do not be surprised when employers attempt to refute the claim and arbitrators react somewhat indifferently. The evidence on recruitment and retention will need to be clear and compelling to move the dial.
11. Retroactivity, for both current and former employees, is the norm in arbitration awards.
There seems to be some confusion often on either or both sides of the table on the obligation to pay retroactive wage payments. As we’ve explained in the Washington Public Safety Representative’s Manual, retroactivity on wages is presumed when the contract is silent. See Barclay v. City of Spokane, 83 Wn.2d 698, 700, 521 P.2d 937 (1974). We have had to address this issue many times in the past, but if your contract is silent on whether the wages are owed to departed or retired employees, and the contract is written to be retroactive to the first day of the agreement, the law is that they are those former employees are to be paid retroactivity for those months of service. Therefore, to preclude the obligation to make such retroactive payments, the employer must expressly negotiate for such an exclusion.
Given the presumptive duty to pay former employees, employers have often argued in arbitration not to make the Award retroactive, at least to those former employees. The retroactivity of the contract itself is a virtually uniform aspect of every Washington arbitration award, and it has been denied in a small number of rare cases, such as where a severe financial inability to pay existed. Barring such circumstances, retroactivity is a “given” on Washington labor contract arbitrations.
That said, employers have often acknowledged the obligation for the contract to be retroactive but have sought to exclude the former employees. Generally, but not uniformly, arbitrators have rejected that request. The typical arbiter rationale is that those former employees provided a critical service and should not be deprived of payment because the parties were unable to arrive at a timely agreement.
In Island County, the Guild faced a situation where the County had historically bargained such retroactivity out of the contract and the previous contract contained that exclusion. The Guild put on a forceful case and noted the substantial impact denial of retroactivity would have on these employees. What needs to be borne in mind, is that usually the issue isn’t just the retro payment. More often, the more significant impact is on the “final average salary” and therefore the pension.
For whatever reason, probably due to the increase in employee turnover, we’ve seen more employers in 2022 trying to resist paying retro to former employees. While some of your members may not care, the impact on your former members can be tremendous, and the precedent it sets as to all members is a bad one. Just keep in mind that once this practice starts, the employer will want it to continue, and it will negatively affect, to a greater or lesser extent, almost everyone’s pension in the end.
12. Special pays and premiums can be difficult to win by arbitration.
As we’ve analyzed past arbitration decisions, we’ve detected a clear pattern of arbitrators granting moderate wage increases while highly resistant to adding new premiums. That doesn’t mean that they never grant them. It just means that even when backed by strong comparability data, new premiums can be difficult to achieve.
In this case, the parties compromised and TAed on enhancing special premiums, but arbitrated shift differential and education pay. Comparability data strongly supported both proposals. The fact that the Guild prevailed on even one of those two proposals is something of a win. The approaches arbitrators sometimes take on this subject do vary. Some are willing to add premiums when the support is clear-cut. That especially seems to be the case as to education premiums. Notably, this arbitrator concluded that the evidence was not as compelling for correction officers to receive a premium for education as it was for patrol deputies.
The best approach is to negotiate in a compromise posture on these issues. Often getting your “foot in the door” on an initial premium and building on it later is a sounder strategy that taking it to arbitration and getting a no that may lock you out on the issue for a few contract cycles.
13. In arbitration, reasonable, data-supported proposals, are impact the overall result.
The final three lessons relate to lessons for the advocates at hearings. It is often best to temper your “protected position” proposal. That doesn’t mean presenting the minimum you’d accept. Some negotiations buffer is fine as long as it’s in the range of reasonableness. When parties submit more extreme proposals, some arbitrators appear to tend to react negatively and in a way that makes a parties’ entire case suffer. If you are proposing significant wage or premium increases, you have to be armed with data that makes the proposal appear somewhat reasonable. Where your members insist on an enhancement not well supported by the comps, you have to be ready to explain to the arbitrator why the local circumstances still make that a reasonable ask.
14. In arbitration, strong data presentation matters.
I’ve never stopped being surprised by what arguments an arbitrator found to be successful and which ones they’ve seemingly ignored. If you present a thin case, you’re likely to get a thin result. Arbitration is serious business. You want to come ready with a detailed, competently presented case. If you don’t, your contract may suffer for years.
15. Parties need to be ready to meet a heavy burden when proposing to change contract language.
Noneconomic proposals are more challenging to prevail on than wage proposals. The party supporting a change in contract language bears the burden of proof. Arbitrators often treat this as a high burden. Generally, it’s not enough to show that the language is supported by many comparables. You have to explain what the problem is that warrants revising contract language.
In the Island County case, the arbitrator considered the comparable language and agreed with the Guild that the inability to challenge disciplinary transfers was unreasonable. While the Guild hadn’t identified a past arbitrary transfer, Arbitrator Cavanaugh understood that the future risk of that occurring was significant.
On another language item concerning the appeal of medical separations, he did not think the Guild had proved that the current language was unworkable. While he recognized the ambiguity in the language, he concluded that the issue was best addressed in future grievance arbitrations. In short, he didn’t want to jump in to clean up language that he wasn’t convinced was fully “broken.”
