DECISION AND ORDER
BACKGROUND AND PROCEDURAL HISTORY
The U.S. Department of the Treasury, Internal Revenue Service (IRS or Agency), filed the instant request for Federal Service Impasses Panel (Panel) assistance under the Federal Service Labor-Management Relations Statute (Statute), 5 U.S.C. §7119, concerning a dispute over the structure of its performance award program. The IRS is responsible for the administration of the US Tax Revenue Code. Its mission is to “[p]rovide America's taxpayers top quality service by helping them understand and meet their tax responsibilities and enforce the law with integrity and fairness to all.” The National Treasury Employees Union (Union or NTEU) represents a bargaining unit consisting of more than 66,000 employees at the IRS, who encumber a variety of positions, including: Revenue Agents and Officers, Tax Examiners, Tax Analysts, Customer Service Representatives, and Information Technology Specialists. The parties are governed by a national collective bargaining agreement (CBA) that went into effect in 2018, and expires in 2021.
The parties’ national performance awards program for bargaining unit employees has been in place since 2001. It has developed over time and is memorialized in Article 18 of the parties’ CBA, as well as in a series of agreements executed from 2001 through 2014. Collectively, these agreements are known as the “National Performance Awards Agreement” or “NPAA.” Over the years, the parties have negotiated supplemental agreements that modified the NPAA. Article 18 of the CBA sets forth the general parameters of the awards program, including funding and eligibility requirements, and is currently not open for negotiations.
Under the NPAA, the award pool structure is set by three criteria: (1) Organization –Business Operating Division and Appendix A; (2) Occupational Series; and (3) Geography. The organizational component is based on the organizational divisions of the IRS, which includes 13 Business Operating Divisions (BODs). The organizational component is further subdivided by the functions, or areas of related work, within a BOD. For example, the Information Technology Division BOD contains “User and Network Services,” “Applications Development” and so on, each of which is its own “Appendix A” pool. Every bargaining unit employee in the IRS is assigned to an Appendix A award pool, which accounts for 115 pools across the IRS.
The second criteria, occupational series, is based on the top 10 occupational categories in the IRS with an 11th category for employees that do not fall within the top 10 entitled, “other.” The final awards pool criterion consists of the geographic location of the employees by state or territory. Thus, employees in the same BOD, Appendix A pool, and occupational series compete for an award with other similarly situated employees in a common geographic area. For example, there is an award pool for the Small Business/Self Employed (BOD), Examination (Appendix A pool), Revenue Agents (Occupational Series) located in Phoenix, Arizona (Geographic area/territory). The parties stipulate that there are approximately 822 total award pools as of 2018.
Pursuant to Article 18 of the parties’ CBA, the Agency funds the awards program at no less than one percent of the total annual bargaining unit salary. The amount of an employee’s award is based on a shares system, with the employee’s grade level and performance appraisal rating determining the number of shares they receive; the higher an employee’s performance score and grade, the higher the performance award. Employees in each pool are ranked based on their average critical job element (CJE) score in their annual appraisal, which ranges from one to five, with five being the highest. Employees with a CJE of 4.6 or above are guaranteed a performance award. Employees with a CJE of at least 3.4, that have an appraisal rating in the top 55 percent of their pool, and have at least 12 continuous months of service are eligible for a performance award. Employees with less than a 3.4 CJE are ineligible for an award.
Due to various reorganizations, realignments, and restructuring of the Agency, the parties attempted to negotiate a simplified award pool process and structure. From 2009 to 2011, the parties formed a joint committee to try and streamline the award pool process and structure to no avail. Pursuant to the agreed upon dispute resolution procedure in the CBA, the parties in 2014, requested the assistance of a Factfinding Panel. Article 47, Section 2.G of the CBA states in relevant part, “[I]f the parties fail to reach agreement at the end of the bargaining period, the parties agree to use the following procedure to resolve any remaining disputes:
- Either party may contact the designated Factfinder that has been selected by the national parties to advise the Factfinder of the dispute. The parties will submit their final proposals and any supporting documentation to the Factfinder…prior to the initial mediation session.
(c) The Factfinder is empowered to assist the parties in reaching agreement. The Factfinder shall determine the appropriate resolution process, including last and best offers (Article-by-Article or issue-by-issue) or amendment of final offers.
(d) Following mediation and factfinding…the Factfinder will issue a recommendation to resolve the dispute....
(e) Any disputes remaining after submission to the Factfinder will be resolved pursuant to 5 USC §7119, or other appropriate provisions of 5 USC §7101, et seq. The party that moves such remaining disputes to the statutory impasse resolution process carries the burden of proof regarding the reasons the Factfinder’s report does not resolve the issue at impasse.”
On March 18, 2014, the Factfinding Panel issued a recommendation to the parties regarding the structure of the award pools:
The award pools presently in effect are complex and cumbersome to administer. More importantly, they are markedly and increasingly inconsistent with the Agency’s organizational structure. The pools are overdue for adjustment to a system which is simpler and which will reflect the Agency’s organization. The Panel recommends that the pools be kept in place pending renegotiation through an expedited bargaining process that will put the new pools in place prior to April 1, 2015 or such other date as will allow awards to be calculated and paid for FY15.
As a result of the recommendation, the parties executed an agreement that required them to commence negotiations over a new bargaining unit award pool structure. The parties bargained in 2016 and 2017, but were unable to reach agreement during the negotiations. Pursuant to Article 47, Section 2.G of the CBA, the parties submitted the dispute to another Factfinder. A mediation session was held with Factfinder Sean Rogers on May 14 to 15, 2019, which did not produce an agreement. The parties moved to Factfinding with Mr. Rogers on June 24 to 25, 2019, and submitted their last best offers for consideration.
On July 29, 2019, the Factfinder issued a decision, recommending that the parties adopt the Union’s last best offer; that the parties form a four-person so-called “garage team” to automate the award pools; and he recommended that the duration of the new award pool structure should coincide with the term of parties’ next CBA.
Following receipt of the Factfinder’s recommendations, the Agency notified the Union that it was unable to accept the recommendations. The Agency filed the instant request for Panel assistance on September 27, 2019. On January 8, 2020, the Panel voted to assert jurisdiction over the request for assistance and ordered the parties to a Written Submissions procedure with an opportunity for rebuttal statements. The parties timely provided their submissions and final offers, which are attached to this Decision, along with the Factfinder’s recommendation.
FINAL OFFERS AND POSITIONS OF THE PARTIES
- Agency’s Position
- Agency’s Final Offer
- Administrative Burden
- Agency’s Final Offer
The Agency states that it seeks to implement a more simplified award pool structure by removing the Appendix A pools and eliminating geography as factors. Instead, the award pools under the Agency’s proposal will be formed based on two factors: (1) placing an employee into one of the 13 BODs; and (2) based on the top 11 occupational series. All employees in the same BOD and occupational series – regardless of geographic location – would comprise an award pool and compete nationally for awards. The current eligibility requirements would remain the same: employees with a CJE of 4.6 or above would be guaranteed to receive an award; while employees with a CJE of at least 3.4 that have an appraisal rating in the top 55 percent of their pool and have at least 12 continuous months of service would be eligible for a performance award. Employees with less than a 3.4 CJE would not be eligible for an award.
Based on 2018 data, the Agency contends that its final offer would reduce the total number of award pools from 822 to 51, making it much easier to administer. Each year, the Agency states that the award process requires approximately eight to nine months to administer; the majority of that time is spent configuring the complex award pools. As part of its submission, the Agency created a timeline detailing the amount of work required to administer the current award pools. In this chart, it indicates that over the course of eight months, in order to administer the system each year, two employees spend 4,000 hours to manually input the award pool data. The Agency states that when it reorganizes, which occurs regularly, it impacts the Appendix A and geography sub-pools, requiring significant pool reconfiguration. However, by eliminating those sub-pools as factors, the Agency contends that reorganizations or realignments within a BOD would no longer necessitate award pool changes, saving the IRS hundreds of hours of time and resources. As a result, the two employees will spend only 900 hours under the Agency’s proposal inputting the data over the course of two months.
The Agency provided an affidavit from a Senior Human Resources Specialist who primarily oversees and implements the award pool configurations. In her affidavit, she details the three steps involved to build the award pools: (1) updating the award poll configuration spreadsheets; (2) running the award pool builder software; and (3) verifying the validation report. Within each of these steps, she states that there are several sub-steps that must be followed in order to administer the system. For example, she states that in the event of restructuring or reorganizations, adjustments to the prior year’s Appendix A pools may be necessary in order to build new award pools, or delate obsolete pools to properly reflect the organization. To create the sub-pools in which employees ultimately compete against each other for performance awards, the Appendix A pools are broken down based on occupational series and geography. Figure 2 within the affiant’s statement depicts these “sub-pools” by illustrating one of the IRS’s BODs - the Small Business/Self Employed Division (SBSE)- which had 97 sub-pools in 2018 after applying the occupational series and geographic categories to the pool. The 97 sub-pools within the SBSE BOD only account for a little over 10 percent of the total pools in 2018, which the Agency must administer. The Agency states that by removing Appendix A and geography as factors, its proposal simplifies the award pool structure by utilizing a two-step process to create award pools: it uses only BODs and occupational series, which reduces the number of total award pools to 51, thereby decreasing the administrative burden on the Agency in managing the system.
- Awards Highest Performers
The Agency contends that its proposal rewards the highest performers pursuant to 5 C.F.R. § 451.104(h), which states, “[p]rogams for granting performance-based cash awards on the basis of a rating of record at the fully successful level (or equivalent) or above, as designed and applied, must make meaningful distinctions based on levels of performance.” The Agency points to the Federal Register, which indicates that making a meaningful distinction in performance means employees with higher performance ratings should receive larger awards than those with lower ratings. The Agency states that its proposal does result in 1.16 percent fewer overall awards; however, the Agency argues that the overall funding of awards does not change, and higher performing employees are better recognized by awarding them larger award amounts.
Under the current award pool structure and the Union’s offer, the Agency states that there are circumstances where a higher performing employee is deprived of an award while a lower performing employee in the same BOD and occupational series receives an award. The Agency provided a simulation of the parties’ final offers, which indicates that under the Union’s offer, a higher rated employee in the same BOD and occupational series as another similarly situated employee does not receive an award, while other employees in the same BOD and series with lower scores received an award. This is because the IRS has office locations across the country and remote management is prevalent; employees under the same occupational series, BOD, and manager may be located in different geographical areas and, therefore, placed in different Appendix A pools. An employee may end up receiving an award, while a similarly situated employee does not.
For example, if two Revenue Officers work remotely in the SBSE BOD, but one works in Boston and one works in Dallas, the geographic location of the employees will influence whether they receive an award; the two Officers will be in two different award pools – one for Boston and one for Dallas. If the Boston Officer has an appraisal score of 4.4, but his pool has a CJE cutoff score of 4.6, he will not receive an award. If the similarly situated Dallas Officer has an appraisal score of 3.8 and her pool has a CJE cutoff score of 3.6, this Officer might receive an award if they fall within the top 55 percent of their pool. The Agency states that under its offer, there are no scenarios where a higher performing employee would not receive an award, while a lower performing employee in the same BOD and occupational series did receive an award. Whether one or both of the Revenue Officers receive an award would depend on their CJE scores and the average appraisal score of the award pool members; not the area of the country that the employee is located.
- Arguments against Union’s Final Offer
- Award Pool Administration
By requiring the Agency to administer two award pools – one using the current three category criteria and another grouping employees with an average CJE score of 4.6 or higher by BOD and occupational series – the Agency states that the Union’s proposal will utilize inconsistent structures and add to the complexity and problematic administration of the award program. Under the pool that maintains the current award pool structure, the Agency states that award pools must still be configured and reconfigured annually, using BOD/Appendix A pools, occupational series, and geography. As previously noted by the Agency, the award pools using the current structure requires multiple steps and revisions to ensure the Appendix A and their sub-pools are appropriately configured and employees placed in the correct pool. The Agency will also have to manage a second pool - the new “4.6 pool,” which it states will require extensive software reprograming and accompanying costs to create the additional award pools.
To illustrate the accompanying costs associated with the Union’s offer, the Agency prepared a chart indicating the time required to administer the award pools under the Union’s proposal. The data indicates that it will take two employees 4320 hours per year to administer the awards pool under the Union’s proposed structure. The Union’s proposal will actually require the Agency to spend more time administering the award pools than the current program; the Agency attributes this to the Union’s two-category pool structure. The Agency states that administering two types of pools – some utilizing geography and others using a 4.6 CJE cut off score as factors – will lead to grievances and unfair labor practice charges alleging that the Agency failed to properly manage each pool and fairly reward employees.
The Agency provided an affidavit from an IRS Lead IT Specialist whose primary duty is to oversee and implement enhancements to the National Performance Awards Program (NPAP) database. He stated that the current process requires the manual mapping of each employee to the appropriate pool, which entails integrating data from two distinct computer systems: the NPAP and the Treasury Integrated Management Information System (TIMIS). NPAP is used in award pool administration and TIMIS is used to store and track official personnel and payroll data for employees. NPAP currently interfaces with TIMIS by importing pertinent data to the award system. He states that under the Union’s offer, it would require several months to reprogram, test, and implement the software to administer two award pool systems. However, under the Agency’s offer, he states that it would be much easier to implement, since existing software can be used and updated by simply removing two categories from the award pool configuration.
Finally, the Agency argues that the language contained in the Union’s offer only eliminated the bargaining obligation related to the “non-4.6 pools.” The Agency, however, would still be required to bargain with the Union when reorganizations or realignments impact the “4.6 pools.” Thus, the Agency states that the Union’s offer does not completely relieve it of its bargaining obligation when administering the award pool process.
The Union objects to requiring employees to compete for awards nationwide because employees perform different duties under different position descriptions (PDs). The Agency defends its proposal by stating that the two Revenue Officers mentioned above may have different duties, but they fall under the same occupational series, so their skills are the same; therefore, they are not harmed by being placed in one pool. The Agency further argues that under the current award pool structure and one of the Union’s proposals, some award pools already contain colleagues in the same occupational series with different PDs. For example, the Agency states that in FY 2018, award pools in SBSE consisted of employees in occupational series 0526, but these employees did not have identical specialties or work under the same PDs; yet, they competed against each other for an award. Thus, the Agency argues that under the Union’s proposal, employees with different PDs will still compete for awards within the same pool.
The Union, in its position statement indicated there is a disparate rating culture across the IRS, arguing that employees under the same occupational series receive different performance ratings based on their geographic location. The Agency provided a simulation of award distributions with and without geography as factors in the award pool calculation. The Agency states that the data shows there is no significant impact on the number of award recipients by removing geography.
- Award Pool Data
The Agency produced an affidavit from a re-employed annuitant assigned to a Management and Program Analyst Position in the Worklife, Benefits, and Performance (WBP) Division. Her duties involve working on special projects for the Agency, such as award negotiations. She explains that the Union’s data, that it relied upon to formulate its proposal, is flawed for two reasons. First, the Union ran simulations of the Agency’s proposal using an unidentified computer program, not the NPAP system that actually calculates bargaining unit awards. She states that the Agency’s simulations, using the NPAP database, produced an accurate reflection of the award results under the Agency’s proposal. When she compared the results of the Agency’s simulation to the Union’s, she found that the Union’s did not match the NPAP figures. For example, the Union found specific employees ineligible for an award, where the Agency’s simulation found those same employees eligible.
Secondly, the affiant states that the Union did not include the total employee population when it ran simulations of the Agency’s proposal. For example, in Fiscal Year (FY) 2018, she states that the Union’s submission for the Wage and Investment BOD was missing 4,047 bargaining unit employees; therefore, the conclusion that the number of awards would drop by 250 under the Agency’s proposal is not accurate; the Agency’s proposal would actually yield one more award than the Union’s in that BOD. Similarly, the claim made by the Union that 2,296 bargaining unit employees who received awards in FY 2018 would not have received awards based on the Agency’s proposal is incorrect. The affiant states that based on her calculations using NPAP, the overall number of awards in FY 2018 would only drop by 1.16 percent under the Agency’s final offer, which equates to 741 employees. She further states that despite this loss, the Union failed to recognize the 825 higher-rated employees who would have gained an award in 2018 using the Agency’s proposed structure.
The affiant also found similar discrepancies in the Union’s FY 2015 and FY 2009 data. For example, in FY 2015, the Union stated that 2,603 employees who received awards under the current awards pool would not receive an award under the Agency’s offer. However, she found that number incorrect and the actually number was 2,550 employees. Moreover, she states that the Union failed to recognize the 1,798 employees who would have gained an award in 2015 under the Agency’s proposal.
- Factfinder’s Recommendation
For the reasons previously stated, the Agency indicated it could not accept the Factfinder’s recommendation, which recommends that the parties adopt the Union’s final offer and adds a full-time four-person team to develop a computer automated award pool process by the end of FY 2020. The Agency claims that the requirement to implement a four-person team to automate the award process disregards the Agency’s security concerns that such a system can be implemented. The Agency states that the award pool process has already been automated to the maximum extent possible without jeopardizing the security of the Department of Treasury’s information technology system – TIMIS – a system that the IRS does not own. The Agency further claims that the requirement to create an automation team is a permissive subject of negotiations involving the technology, methods, and means of performing work under section 7106(b)(1) of the Statute, and is also non-negotiable because it impermissibly interferes with the Agency’s right to direct employees and assign work under section 7106(a)(2)(A) and 7106(a)(2)(B) of the Statute.
- Union’s Final Offer
The Union’s offer uses two categories to determine awards. The first of which combines all employees with a CJE cut-off score of 4.6 or above into “Consolidated Pools” by BOD and occupational series, eliminating Appendix A pools and geography. The Union states that there is no evidence of a rating bias among the “4.6. pools” because those employees automatically receive an award; therefore, the geographical component becomes unnecessary in that pool. The Union also removed Appendix A pools from this grouping in an effort to mirror the Agency’s offer.
In the Union’s second category of pools, it groups the rest of the employees whose CJE scores are lower than 4.6 by the same three criteria structure used for the current award pool structure. The Union states that the “non-4.6 pools,” maintains the geographic component because employees in this pool are not guaranteed an award; therefore, they need the geographic “safeguards” in place to minimize the disparate ratings that are experienced by similarly situated employees in different locations. The Union states that its two-category award pool structure would result in a reduction of the 822 award pools to 199 total award pools: 152 “non-4.6 award pools” and 47 “4.6 or higher award pools.”
The Union contends that by having geographically discrete pools, the negative impact of “disparate rating cultures” is minimized. To determine whether the adverse rating cultures have persisted today, the Union compared the impact of the Agency’s final offer, which eliminates geography and Appendix A pools to the current award pool structure using FY 2018, FY 2015, and FY 2009 data. For each fiscal year, the Union determined that thousands of employees who received an award under the current structure, would not receive awards under the Agency’s final offer. The Union contends that this information indicates that the geographical component is necessary in the administration of an award pool program.
Under the Agency’s proposal, the Union argues that 2,296 employees located in different areas who received awards in FY 2018 would not have receive awards based on the Agency’s proposed structure. For example, the Union states that over 300 employees locate in either Philadelphia, Austin, Kansas City, and other locations throughout the country received awards in FY 2018, but would not receive an award under the Agency’s final offer. For FY 2015, the Union contends that 2,604 employees who received awards under the current structure would not receive awards based on the Agency’s proposed award pool. For example, several hundreds of employees located in San Francisco, Pittsburgh, and Fresno received awards in FY 2015, but would not receive awards under the Agency’s offer. Finally, for FY 2009, the Union states that 3,123 employees who received awards under the current system would not receive awards under the Agency’s offer. For example, the Union again states that several hundreds of employees located in Atlanta, Philadelphia, Fresno received awards in FY 2009, but would not receive an award under the Agency’s proposal. Under the Union’s proposal, it states that eligible employees would not lose out on an award.
- Administrative Burden
The Union states that one of its pools is similar to the Agency’s final offer, i.e., the consolidated “4.6 pool.” The Union’s “4.6 pool” eliminates geography and Appendix A pools, just as the Agency’s proposal, but groups the 34,000 employees that have received a 4.6 CJE score or higher in FY 18 into one pool. Under the Union’s second pool, the remaining 32,000 employees are placed in the “non-4.6 pools,” which the Agency will continue to administer the same award system that’s currently in place. The Union states that under its proposal, the Agency is not required to negotiate with the Union when it reorganizes under the “non-4.6 pools,” and when it reorganizes under the “4.6 pools” as well. Thus, the elimination of 670 pools and the elimination of the Agency’s obligation to bargain over award pool reconfigurations whenever it reorganizes will alleviate the administrative burden imposed on the Agency. As a result, the Union states that there will likely be less errors when processing performance awards and consequently, less grievances and unfair labor practice charges.
- Factfinder’s Recommendation
The Union points to the Factfinder’s recommendation for support of its proposal. The Factfinder made three recommendations: (1) the parties should adopt the Union’s final offer; (2) the parties should adopt the recommendation to create a team to automate the annual award process by FY 2020; and (3) the parties’ agreement on the award pool structure should have the same duration as the parties’ next term agreement. With respect to the Factfinder’s second and third recommendation, the Union states that it is not asking the Panel to resolve those issues. Therefore, the Union is requesting that the Panel only impose the portion of the Factfinder’s recommendation that adopts the Union’s final offer.
In support of adopting the Union’s proposal, the Factfinder reasoned that the Union’s offer reduces the total number of award pools from 822 to 199, helping to simplify the award pool structure. The Factfinder indicated that although the Agency’s proposal reduces the number of award pools to 51, in doing so it decreased the regional character of the business units, particularly with regard to the different job duties within a location. He found that the Union presented convincing evidence that many employees in the same BOD assigned to the same occupational series work under significantly different position descriptions in different locations. As a result, he stated that a national awards program would unfairly group many bargaining unit employees with significantly different job duties into the same award pool. The Factfinder further found that the Agency’s elimination of geography as a criterion would render the 51 award pools overbroad in scope and would raise questions of fairness and objectivity of ratings and evaluations. Finally, the Factfinder noted that although the Agency would initially have to create the two-category award pool structure to adopt the Union’s proposal, after the first year, the administrative work load and time for managing the awards program would be reduced, since there would only be 199 pools to administer. He stated that the Union’s offer to release the Agency from its bargaining obligation would also reduce the Agency’s costs in administering the system.
- Arguments Against the Agency’s Final Offer
- National Awards Program
Under the Agency’s proposal, it contains two components: (1) the 13 BODs; and (2) the top 11 occupational categories. As an example, the Union states that there would be a single pool for the Wage and Investment BOD Customer Service Representatives. All GS-0962s in this BOD would compete against each other nationwide to be in the top 55 percent of the pool, despite differences in duties and geographic ratings from one location to the next. The Union states that employees are not evaluated on a national level; therefore, it is not justified to require them to compete nationally for a performance award. Employees within the same series are under different PDs and perform different work. For example, the Union asserts that a GS-1169 Revenue Officer position could be an employee working in the field, working in an advisory capacity, or working in the centralized Offer-In-Compromise program. The position is the same series, but under a different PD, with different managers and different program goals. The Union argues that by the Agency removing geography from its proposal, it ignores the different job duties within a location and the resulting pools would unfairly group many employees with different duties into the same nationwide award pool.
- Rewarding Highest Rated Employees
In response to the Agency stating that its proposal will reward the highest performing employees, the Union argues that the Agency conflates “highest performing employees” with “highest rated employees.” The Union states that there is no evidence to show that a higher rating directly correlates with higher performance. The Union asserts that the Agency has not introduced evidence substantiating the claim that there are no scenarios under its offer where a higher performing employee would not receive an award, while a lower performing employee in the same BOD and series would receive an award. The Union on the other hand, states that it introduced evidence that showed geographic location significantly impacts the level of CJE rating that employees will receive. Therefore, the Union argues that it necessary to maintain the geographic component for the “non-4.6 award pools.”
The Union also claims that the Agency is attempting to make changes to the eligibility requirements set forth in Article 18, which is not open for negotiations by modifying its proposal to ensure that lower rated employees do not receive an award over higher rated employees. In Article 18, Section 2.A., it states that employees with an average CJE score of at least 3.4 and who have at least 12 continuous months of service at the IRS, are eligible for a performance award. The Union points out that under FLRA case law, the Agency must wait to make changes until Article 18 is open for negotiations in September 2020.
The parties tried unsuccessfully to negotiate a more efficient award pool structure since 2009. In 2014, pursuant to Article 47, Section 2.G of the CBA, the parties requested the services of a Factfinding Panel in order to assist them in reaching an agreement over a more effective award pool process. The Factfinding Panel ordered the parties to bargain a new award pool process that was easier to administer and consistent with the Agency’s organizational structure. The parties tried once again to negotiate a new award pool process, but, after failing to reach an agreement, they selected a Factfinder in May 2019, to provide mediation and factfinding assistance. The mediation did not produce an agreement; therefore, the parties submitted their last best offers to the Factfinder for a recommendation on the award pool structure.
In July 2019, the Factfinder issued his decision with recommendations that the parties adopt the Union’s final offer, that the parties create a four-person team tasked to automate the award pool process, and that the duration of the new award structure should coincide with the term of the parties’ next CBA. The Agency refused to adopt the Factfinder’s recommendation, so it filed a request for Panel assistance. In accordance with the Section 2.G(e) of the parties’ CBA, the “party that moves such remaining disputes to the statutory impasse resolution process carries the burden of proof regarding the reasons the Factfinder’s report does not resolve the issue at impasse.”
The Agency has demonstrated that the Factfinder’s recommendations fail to resolve the impasse with which the parties have struggled since at least 2009. It is evident that, each year, the current award pool process requires approximately nine months to administer and that much of that lag is due to the fact that the system has to be reconfigured every year to account for changing geographical factors and Agency reorganizations. It is also evident that these annual reconfigurations currently require manual calculations using data from both NPAP and TIMIS, that there is no means available to integrate those two systems, and that there is no identifiable alternative to manual calculations. Finally, it is evident that the Union’s proposal would continue to require that the Agency continue these manual calculations but, prior to implementation, would require additional software reprogramming and months of testing.
The Factfinder’s solution for these complicated administrative challenges was to create a new, four-person garage team comprised of two representatives from each party, and to require that this team find an as-yet-unidentified way to automate the calculations necessary to efficiently administer the award pool system.
Whatever the merits of this recommendation, it does not resolve the impasse between the parties. Meanwhile, as described above, the Agency’s proposal removes geographic factors that have made annual reconfigurations necessary, addressing the root of the impasse.
Additionally, the Agency submitted evidence demonstrating that the Union’s proposal would produce results contrary to the purpose of the award system. For example, the Agency produced an affidavit from a re-employed annuitant whose duties involve working on special projects for the Agency, which included the negotiations over the award pool structure. Based on her statement, the data that the Union relied on in support of its proposal to the Factfinder appears to be inaccurate, and therefore, the Factfinder’s reliance on this inaccurate data is unsupported. The NPAP system, which is used by the Agency to calculate bargaining unit awards, found employees eligible for a performance award that the Union found ineligible. She found that the Agency’s proposal, which eliminates geography, would only reduce the number of employees who received an award by 1.16 percent, or 741 employees based on FY 2018 data. That number is far less than the 2,296 employees that the Union stated would not receive an award under the Agency’s proposal. In sum, the Union’s inaccurate data overestimated the negative impact of the Agency’s proposal. While 741 employees is a sizable amount, when factoring in the additional 825 “highly” rated employees that would end up receiving an award under the Agency’s proposal, the gap between the parties’ two proposals is not nearly as wide as the Union purports, and may in fact result in more employees receiving an award.
The Union argued to the Factfinder that high performance does not correlate to high ratings; that there is no evidence to show that a high rating means high performance. However, in order to be a highly rated employee, that employee must be performing at a high level. Therefore, a performance award system should reward these employees. An awards program should make meaningful distinctions in performance by rewarding employees who are higher performers with larger awards than those with lower ratings. The Agency’s proposal makes meaningful distinctions among employees because it ensures that highly performing employees receive awards. The Union expressed concern that some employees under the Agency’s proposal may not receive an award because the focus will be on the highest performing employees.  However, the Panel believes that the Agency’s award proposal will incentivize employees to work harder and perform better in order to ensure that they achieve a performance award.
The Agency presented data to the Factfinder under one of the Union’s proposals that there are circumstances where a higher performing employee does not receive an award while a lower performing employee in the same BOD and occupational series does receive an award. The Agency attributed this mainly due to geography. Therefore, the Agency has proposed to nationalize the performance award system to ensure that highly rated employees do not miss out on awards due to factors unrelated to performance, like geography.
The Union’s proposal, however, opposes a national award system and maintains the geographic component for employees that are not guaranteed an award. The Union argued that for these employees it’s necessary to maintain geography because similarly situated employees are not rated equally from location-to-location. The Factfinder found in favor of adopting the Union’s proposal because the Union “presented convincing evidence” that grouping employees into a national awards pool, as the Agency proposes to do, would “raise questions of fairness and objectivity of ratings and evaluation which are national in scope.” The Panel disagrees. First, the Panel does not believe the Union’s data is convincing, as the Union’s data has been shown to be inaccurate. Secondly, the Panel is not convinced that the Agency’s national-focused (not geography-based) program is unfair and lacking in objectivity. Therefore, maintaining a geography-based component, as the Union proposes, is not be necessary.
Most of the evidence offered to the Panel was also provided to the Factfinder. In its review of the evidence, the Panel disagrees with the Factfinder that the Union’s evidence was convincing. The Union did not present the Panel with evidence that “convincingly” indicated employees would suffer significant negative impact if geography were removed from the awards system. As previously noted, the Panel found the three affidavits that the Agency produced, whose testimony was also provided to the Factfinder. Not only did these witnesses provide statements contradicting the Union’s proposal, but they provided data that actually invalidated the Union’s evidence. Specifically, the data the Union principally relied upon to justify to the Factfinder that retaining geography as a component of the award pool structure was discredited by one of the affiants. Given this, the Panel will not defer to the Factfinder’s recommendation to adopt the Union’s proposal. Instead, the Panel will accept the Agency’s proposal because the Agency met its burden in demonstrating that the Factfinder’s recommendation did not resolve the impasse between the parties. Additionally, the Panel found that the Agency presented more convincing arguments and evidence to support the adoption of its proposal. As such, the Agency’s proposal, not the Union’s, is most consistent with an effective and efficient award system and, therefore, it should be adopted.
Pursuant to the authority vested in it by the Federal Service Labor-Management Relations Statute, 5 U.S.C. § 7119, and because of the failure of the parties to resolve their dispute during the course of proceedings instituted under the Panel’s regulations, 5 C.F.R. § 2471.6(a)(2), the Federal Service Impasses Panel under § 2471.11(a) of its regulations hereby orders the adoption of the Agency’s final offer to resolve the impasse.
By direction of the Panel.
Mark A. Carter
April 9, 2020
 The parties memorialized the award pool structure in “Appendix A” of the 2001 NPAA; hence the name.
 Agency’s Figure 7 at 0164.
 Agency’s Attachment 3; Affidavit of Nicole Burton.
 72 FR 1267-01 (2007).
 Agency’s Attachment at 0161.
 Agency’s Attachment at 0162.
 Agency Attachment at 0164.
 Agency Attachment 4; Affidavit of Dan Tippets.
 Agency Exhibit 13 at 0165.
 Agency Attachment 5.
 Agency’s Exhibit 10 at 160.
 Agency Attachment 4; Affidavit of Dan Tippets.
 Union’s Attachment 10, 11, and 12
 The Union has indicated that it is not necessary for the Panel to address the Factfinder’s garage team or duration recommendations, Union’s Statement of Position at 12. However, it is clear that the garage team recommendation was made precisely because the Union’s proposal perpetuated—indeed, exacerbated—the difficulties inherent to the current award system.
 The Agency also presented arguments that the Factfinder’s recommendation to create a four-person team to automate the award system may be non-negotiable (i.e., legally unenforceable) because it interferes with management’s right to assign wok under section 7106(a)(2)(B) of the Statute. Because the Panel has ordered adoption of the Agency’s proposal, it is unnecessary to address the Factfinder’s recommendation as to the garage team or the duration.
 The Union also argued that the Agency’s proposal is an attempt to circumvent their bargaining obligation with the Union by modifying award eligibility requirements. However, the Agency’s offer does not amend the eligibility requirements for employees to receive an award; employees must still obtain at least a 3.4 CJE score to be eligible. What the Agency’s proposal does do is ensure that employees who are highly rated and, thus, highly performing receive an award.
 All three affiants did not testify during the factfinding; however, the information contained in their affidavits was communicated to the Factfinder.