In response to the 1999 crisis caused by the mass resignation of the European Commission, the Commission introduced a series of administrative reforms based in large part on New Public Management models. A centerpiece of those reforms was a new staff appraisal process linking numeric ratings with promotions, which was designed explicitly to change the management culture of the Commission. Of all parts of the reform, this was by far the most controversial. This paper traces the long arc of reform, as the original reform was replaced with a second version that was even more rigid and complex, leading to a third reform, in 2012, which returned the Commission in large part to the status quo ante, abandoning numeric ratings and the formal link to promotions. I analyze the reasons for the reforms and the problems and unintended consequences of each. In conclusion, I link this saga of repeated reforms to the broader literature on the effectiveness of attempts to change organizational culture through formal structural reforms.
In response to the 1999 crisis caused by the mass resignation of the European Commission, the Commission introduced a series of administrative reforms based in large part on New Public Management models. A centerpiece of those reforms was a new staff appraisal process linking numeric ratings with promotions, which was designed explicitly to change the management culture of the Commission. Of all parts of the reform, this was by far the most controversial. This paper traces the long arc of reform, as the original reform was replaced with a second version that was even more rigid and complex, leading to a third reform, in 2012, which returned the Commission in large part to the status quo ante, abandoning numeric ratings and the formal link to promotions. I analyze the reasons for the reforms and the problems and unintended consequences of each. In conclusion, I link this saga of repeated reforms to the broader literature on the effectiveness of attempts to change organizational culture through formal structural reforms.
PITTSBURGH PAPERS ON THE EUROPEAN UNION The Challenges of Linking Pay and Promotion: Repeated Reforms of the European Commission Staff Appraisal Process Carolyn Ban University of Pittsburgh Abstract In response to the 1999 crisis caused by the mass resignation of the European Commission, the Commission introduced a series of administrative reforms based in large part on New Public Management models. A centerpiece of those reforms was a new staff appraisal process linking numeric ratings with promotions, which was designed explicitly to change the management culture of the Commission. Of all parts of the reform, this was by far the most controversial. This paper traces the long arc of reform, as the original reform was replaced with a second version that was even more rigid and complex, leading to a third reform, in 2012, which returned the Commission in large part to the status quo ante, abandoning numeric ratings and the formal link to promotions. It analyzes the reasons for the reforms and the problems and unintended consequences of each. In conclusion, it links this saga of repeated reforms to the broader literature on the effectiveness of attempts to change organizational culture through formal structural reforms. Keywords: European Commission; administrative reform; pay for performance; organizational structure About the Author Carolyn Ban is a Professor at the Graduate School of Public and International Affairs at the University of Pittsburgh. ISSN: 2164-6732 (Online) | https://doi.org/10.5195/ppeu.2017.30 This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License. This site is published by the University Library System of the University of Pittsburgh as part of its D-Scribe Digital Publishing Program, and is cosponsored by the University of Pittsburgh Press. Carolyn Ban The Challenges of Linking Pay and Promotion: Repeated Reforms of the European Commission Staff Appraisal Process Carolyn Ban Introduction Over a decade ago, the European Commission (EC) introduced a major series of administrative reforms generally referred to as the Kinnock reforms. Named after the then Vice President of the Commission and Commissioner from the UK Neil Kinnock (European Commission, 2000; Kinnock, 2004), this broad reform, undertaken in response to a major scandal, included a new process of formal staff appraisal linked to promotion, typically a central component of New Public Management: (NPR) reforms, designed to improve performance. Scholars have analyzed the passage of the reforms (Kassim, 2004a; Kassim, 2004b; Stevens and Stevens, 2006) and their implementation (Bauer, 2007; Levy, 2003; Schön-Quinlivan, 2011). Some have written about the initial reactions to the reforms and have posited that they might actually lead to reduced effectiveness in developing policies (Bauer, 2008; Levy, 2006). However, that was not the end of the story. In fact, the initial reform of 2003 (which I term version 1) was strongly disliked and was replaced in 2007 by version 2, which was actually even less successful and created many problems. These issues led to the administration introducing version 3 in 2011, which in many ways returned the Commission to the status quo ante with no formal linkage between appraisal results and promotion. There has been no scholarly discussion of this succession of attempts to get performance appraisal and promotion right. This study has two goals: first, to document this complex history, which is not broadly known outside of the EC, and second, to analyze each of the successive reforms, focusing on four main themes: Motivation: Were reforms designed appropriately to increase motivation? Is there any evidence that they did so? Or did they, instead, “crowd out” intrinsic motivation or even cause demotivation? Fairness and perceived fairness: Were appraisals under each system perceived as accurately reflecting staff performance and as fair? And was the link to promotion clear? Management reform: The initial reform was explicitly designed to change the role of midlevel managers (Heads of Unit or HoU’s) within the Commission, emphasizing their responsibilities as first-line supervisors, rather than primarily as policy entrepreneurs. Did the reforms succeed in changing the culture of management and the role of managers? Implementation: Did all of the actors down the chain of command buy into the reforms and implement them as intended? Or did they define or distort the procedures to protect the previous cultural assumptions or to reduce demands on themselves? Were the resulting systems actually carried out as the reformers expected? I begin with a brief introduction to the European Commission for readers unfamiliar with the organizational setting and with a short summary of the relevant theories and debates on these four themes. I then detail the goals and procedures for each version of the appraisal and promotion process and analyze the impacts of each in terms of the four themes. This is, in effect, a Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 2 The Challenges of Pay and Promotion series of linked case studies, in which the organizational context is held more or less constant (with some exceptions) so that we can see clearly the effects of each appraisal model. 1 Background The European Commission The European Union (EU) differs from other international organizations. A regional association with 28 (soon to be 27) members, it has considerable policy-making power, as member states give up considerable sovereignty and commit to adopting EU legislation within their countries. The European Commission (EC) is often described as the EU’s executive branch, but that does not capture its actual functions. New legislative proposals originate in the Commission and are then sent for passage to the European Parliament (directly elected by citizens) and the Council of Ministers (representing the member states). Thus, the EC is closer in function to a large think-tank, conducting the analysis, consultation, and drafting of proposals. It is organized into Directorates General (DGs) and services, based on function and similar to national governments’ ministries. In most policy areas, the EC has a limited implementation and oversight role, as the member states are responsible for harmonizing their laws with EU directives and implementing and enforcing them. In some policy areas, such as competition policy (similar to US antitrust policy), it is empowered to take direct action. The Commission staff, thus, is quite small compared to national administrations. At the time of this study, the staff had grown rapidly to a high of about 35,000 members, with 10 new member states (eight of which were from Central and Eastern Europe) joining in 2004 and two more (Romania and Bulgaria) entering in 2007. As a result, the EC recruited new staff from those countries, bringing in staff equal to about 20 percent of the staff preenlargement (Ban, 2013). The entry process is highly competitive and rigorous, so the majority are highly qualified (and sometimes overqualified), and pay and benefits are quite generous compared to national government employees. Norms in many DGs reinforce the sense of belonging to an elite group, with long working hours, broad individual responsibilities, and often high job stress. Four Themes: Theories and Debates Motivation: The core underlying assumption of linking performance appraisals to concrete rewards, whether pay or promotion, is simply that, if offered an attractive incentive, most people will work harder or do better work. In academic terms, this has been articulated as expectancy theory (Vroom, 1964; Porter and Lawler, 1968), which postulates that reward systems change behavior if staff perceive that the work goals can realistically be reached, that meeting the goals will actually lead to the promised reward, and that the reward (almost always financial) is an attractive one. Each link in that process is potentially problematic, and an extensive literature debating the effectiveness of such appraisal and reward systems is sharply split: private-sector research and texts (e.g., Mathis et al., 2014) are more positive than research or textbooks focusing on the public or nonprofit sectors (Ingraham, 1993; Kellough, 1993; Perry et al., 2009; Nigro and Kellough, 2014). In both, there is considerable discussion of when such systems are more or less effective and why bad incentive systems are so prevalent in organizations (Magee et al., 2011). Formally linking pay to performance is widespread in the private sector, and while some private-sector sources acknowledge challenges, the general tone in texts and research is quite positive and assumes that 1 See the methodological appendix for more information on research design and data collection. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 3 Carolyn Ban this is normal business practice. The assessment is reflected in many public-sector reforms that fall under the rubric of New Public Management, which generally advocate adoption of business models of management by the public sector (Pollitt and Bouckaert, 2011). The literature on the public and nonprofit sectors is heavily critical and questions the effectiveness of pay for performance, either for operational reasons, such as pay differentials too small to matter or, more broadly, because pay for performance fundamentally misjudges staff motivation to join the sector or to exert greater effort at work. Extensive research has demonstrated the importance of intrinsic motivation based on responding to internalized drives, such as a “deep interest and involvement in the work,… curiosity, enjoyment, or a personal sense of challenge” (Amabile, 1997: 44, cited in Denhardt et al., 2013: 73) and also of Public Service Motivation (PSM), rooted in a desire to serve the public or to make a difference on public policy (Perry and Wise, 1990; Perry and Hondeghem, 2009). Within the context of the European Commission, this is often articulated as a commitment to building Europe (Ban and Vandenabeele, 2009). Research in the public sector has often reported failure of merit pay to improve work performance and found that that by emphasizing extrinsic motivators, i.e., money or promotion, such systems may undercut or “crowd out” intrinsic motivation (Deci, 1971; Deci et al., 1999; Bellé 2015). A related negative effect is that systems linking pay or promotion to appraisal results are almost always zero-sum games. If employees buy into the extrinsic logic, the organization risks triggering an increase in competitive behavior and a reduction in collaboration, teamwork, and sharing of information (Nigro and Kellough, 2014). Given the sharp split between private- and public-sector assessments of merit pay systems, what can we predict about their acceptance in the European Commission, which has elements of each sector? EC pay levels are more in line with the private sector, and many recruits have had prior experience working in private-sector firms, some at quite high levels. There is, however, clear evidence of intrinsic motivation, including Public Service Motivation, among EC officials (Ban and Vandenabeele, 2009), and the nature of the work is closer to that of the public sector. Unfortunately, the Commission did not conduct a base-line study to assess levels of motivation of either kind prior to introducing reforms, but many interviewees remarked on their impact on motivation. Fairness and Perceived Fairness: The second major challenge in gaining staff acceptance of new performance appraisal or promotion systems is fairness, both actual fairness or accuracy of the appraisal and perceived fairness. These are obviously linked but can differ sharply. Appraisal accuracy is easiest to attain if the work is simple and can be directly measured or counted. Rewarding sales staff is usually quite straight-forward, but higher-level work is more difficult to observe, with projects that extend over time and that depend on the work of others. The resulting information asymmetry increases the potential for bias (Whitener et al, 1998), and there is a wide range of possible sources of conscious or unconscious bias, based on race, gender, age, sexual orientation, or party affiliation among others. Individual managers may apply assessment standards differently. Absent careful record keeping, managers are subject to recency bias, overweighting recent performance, or to central tendency error, avoiding making clear distinctions (see e.g., Milkovich and Wigdor, 1991; Kellough, 2012). There is, however, a real risk that, even if evaluations are accurate and fair, they will be perceived as unfair by employees who believe they deserved a higher rating. Most people are initially supportive of merit pay systems because they tend to overrate themselves and therefore Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 4 The Challenges of Pay and Promotion assume that they will benefit. In fact, most people think they are above average, a statistical impossibility, and many, especially the weakest but also interestingly the strongest (who underestimate their abilities), are actually unable to assess accurately their own achievements (Kruger and Dunning, 1999). Those who perceive they are underrated are likely to blame the rater for being biased unless confronted with clear and detailed evidence. As a result, “[e]fforts to link pay to the outcomes of such a process may invite more employee alienation than motivation” (Kellough (2012:182). Within the context of the EC, where most staff see themselves as high achievers, this research would predict that they would indeed wish to be rewarded for their performance because they expect to be rated highly. But they may also have an inflated sense of their own performance. Implementation Over the past 30 years, policy research has increasingly focused on the complexity of policy implementation (Pressman and Wildavsky, 1984), the advantages of top-down versus bottom-up implementation (Sabatier and Mazmanian, 1980; Sabatier, 1986), and the importance of reflecting policy learning and adaptation (Goggin et al., 1990), best studied through qualitative research (Hill and Hupe, 2009). Most empirical studies of merit pay, however, focus on outcomes (i.e., impact on employee performance, morale, and turnover) rather than on the implementation process. In the case of the European Commission’s implementation of successive reforms, an understanding of the actors involved is critical. They include the central HR office (at that time termed DG Admin), the layers of leadership within each DG, and the HR unit within each DG, which played an active role in introducing and monitoring each new appraisal system. Even if there is extensive training for key actors on the technical requirements of the new system, it is unrealistic to assume that all of the actors are on board and have accepted the logic of the new system. Rather, one can predict considerable resistance, at least initially. There is the risk that actors may have values or goals that diverge from the stated intent of a reform. Reforming management culture: Introduction of administrative reforms, particularly in management, faces a contradiction. Reformers see these new approaches as designed to change the culture and improve management, particularly in areas such as accountability, transparency, and staff motivation. But results of assessment of these reforms have indicated that they are more likely to succeed if they are supported by the existing culture (Merriman, 2011; Magee et al., 2013). Further, reforms of pay or promotion that stress high rewards for a subset of staff are likely to be resisted by unions or staff associations. These groups see their role as defending the majority of officials from being disadvantaged or treated unfairly, leading some scholars to question whether it is possible to introduce such systems in a unionized environment (Durham and Bartol, 2009; Schay and Fisher, 2013). All four factors may contribute to an evolutionary approach to implementation – an iterative process in which an organization learns through trial and error what will work within a specific organizational environment. As we shall see, that may be a useful way of conceptualizing the repeated cycles of reform at the European Commission. Reform in Action: The Status Quo Ante and Successive Reforms Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 5 Carolyn Ban We turn now to a brief description of the situation prior to introduction of the first reform and then present the three successive systems for appraisal and promotion, describing each approach and analyzing the results based on the four themes discussed above. We then draw more general conclusions about the implications of this case for other international organizations. The Formal Status Quo Ante: Weak Appraisal, Strong Seniority Given the weak formal systems of management prior to the Kinnock reforms, it is not surprising that the system in place was simple and limited in its utility. The form included: a description of duties carried out over the reporting period, a note on languages used followed by a self-assessment of language skills, and an account of publications and new knowledge gained during the reporting period. This [was] followed by an assessment of 14 aspects of performance under the three headings listed in…the Staff Regulations, namely ability, efficiency and conduct in the service. (Stevens and Stevens, 2001: 99). These criteria were displayed in a grid format, with the appraiser ticking a box by placing what were commonly referred to as “little crosses” to rate performance on each criterion on a five-point scale, with a small space for optional comments. There was no formal goal-setting or assessment of criteria, which was not surprising, since many staff members did not even have formal job descriptions, let alone formal goals specified at the beginning of the rating period. Several formal rules and informal customs limited the utility of the reports but also the potential damage that the process could cause. Appraisals were conducted only every two years, and, with the joint approval of the supervisor and staff member, the previous rating could simply be carried over, in a process called “reconduction,” resulting in full appraisals for some only every four years. Stevens and Stevens (2001: 101) report that “reporting officers hardly ever use anything but the top two marks (excellent or very good)…and the general assessment is usually completed in perfunctory and uncritical terms.” There was no formal self-assessment, and it is not clear how many supervisors actually met with their employees one-on-one to discuss the appraisal. Further, according to a manager in DG ADMIN (now DG HR), around 25 to 30 percent of evaluations were never completed, which he saw as “an unsatisfactory level of completion.” Given this low rate and the clear “leniency bias,” results were not terribly useful for guiding decisions on promotion, leading to reliance on more informal sources of information (Stevens and Stevens, 2001: 101). An earlier attempt to introduce a numeric system by awarding points based on the number of little crosses in each column was resisted by the unions, resulting in a bizarre situation. According to an HR manager: There was a system of 5, 3, 2, 1, 0 points. And the administration defined a ‘moyen cible” – a target average, more or less, and the administration published for each DG the averages given, for each DG and for each Directorate. …So there was a rather paradoxical situation, where the administration was trying to introduce the system of points, but… the unions…opposed this system, and so the administration said “we are introducing the points but we will neutralize them.” That means that they didn’t appear in the reports, but we know they exist. We can use them for analytic purposes, but officially you don’t have a report with 35 points, you have a report with 3 excellent and 2 superior. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 6 The Challenges of Pay and Promotion In short, the previous system had the big advantage of simplicity (at least until the attempt to calculate points), and it probably had a moderately positive effect on morale and reinforced individuals’ self-image as high performing because managers could be generous in their ratings. However, the old system was of limited utility in making decisions on promotion or in justifying them, particularly in the face of appeals. Reactions to the attempted reform prefigured the challenges of implementation faced in the future. Version 1: The Shock of Reform The new system of staff appraisal and promotion arrived quickly and faced major challenges. I begin with an overview of the formal procedures and then analyze the design and impact of the reform based upon the four themes presented above. Version 1: the formal policy It is important to reiterate that, while most merit pay systems link performance directly to pay, the Commission attempted to link appraisal to promotion, rather than directly to pay. This reflects the EC’s career system, in which the majority of officials enter at the bottom (for new professional officials entering after May 1, 2004, AD5) and work their way up a ladder (which was made four rungs longer in the Kinnock Reforms) to the very top, AD 14. Therefore, the most important extrinsic reward in the Commission is promotion. Of course, promotion includes a pay increase, but it is also a change in status. Given the longer ladder, only those promoted rapidly will ever reach the level that will qualify them for a head of unit position (at least AD 9) or higher. According to a senior staff person who took part in implementing the new system, “We wanted to go for a more managerial and more merit-based management…We introduced job descriptions, and we introduced objectives…. They were two of the pillars of the new CDR [Career Development Review – the official name of the new form].” The CDR was quantitative. Each staff member received a single summary number of merit points ranging in theory from one to twenty (a typical French test-scoring range). That range, however, quickly became truncated because of central guidance warning DGs to avoid using the highest scores except for extraordinary cases and requiring that the mean number of points overall in each DG be within one point of the overall average of the Commission. Since this meant that the ratings were in essence a zero-sum game, supervisors who would have liked to give higher ratings to some staff were inhibited by not wanting to give other staff below-average ratings. Thus, the design of the system succeeded in avoiding a leniency bias by virtually forcing a central tendency bias. In addition, each DG had a pool of priority points, awarded by the Director General (DG) in consultation with the senior staff and a representative of a staff association. This introduced greater flexibility and allowed DGs to speed promotions for high achievers or those who were within a few points of promotion. But it reduced transparency and reintroduced the potential for bias. And it too was hemmed in by rules limiting the number of priority points that could be used (Grøn 2011). An individual’s total points were saved in a “sac à dos” or rucksack, that is, they were collected over time and were not lost if one moved to a different DG or service. Staff whose points reached the threshold for promotion were automatically promoted. In theory, that could happen in two years for very high-fliers in the lower grades or quite slowly for those whose performance was relatively weak, but a poor performer could not be blocked from eventual promotion. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 7 Carolyn Ban Analysis of Version 1: Implementation and impact Scholars who studied the introduction of the Kinnock reforms bought into the standard values of NPR reforms and lauded the fact that the Commission would now be rewarding performance, ignoring the fact that they had done so, if less formally, all along. They certainly did not question that this was happening. But implementation was predictably challenging. The Commission devoted considerable effort to training both managers and officials on how the new system would work, especially on the evaluation process, strongly reinforcing the message that decisions would be made strictly on their performance. There was one immediate success: Everyone understood that the evaluations needed to be taken seriously, and virtually everyone was evaluated. There was considerable support for the process, including both self-evaluation by the official and a face-to-face dialogue between the official and the Head of Unit to discuss the annual assessment and to set annual goals, although many HoUs complained about the time required to complete the CDRs and conduct the meetings with staff members (Bauer, 2008). The link of the evaluation to the reward was, however, problematic. The system of points was bureaucratic and complex, and the addition of priority points reduced the perception of fairness. In a system where rewards are deferred, goal displacement was common. As one senior manager described it, “People focalize not on whether they are doing a good job or what they should be doing or how their career is going to develop over the next two, five, ten years, but whether they get 15 points or 15 and a half points. Disaster.” Implementation also faced the challenge of coordinating different actors with different values. While the policy makers in DG ADMIN (now DG HR) espoused support for rewarding performance, they were responsible for creating a highly bureaucratized system. As one director described it, “I considered it as a disaster. It was technocratic, based on a pseudo-scientific notion that you could evaluate everybody between a few points.” DG ADMIN leaders also focused on controlling grade inflation (and thus budget) by truncating the range of points. This fed the tendency of Heads of Unit to avoid making sharp distinctions among officials. The regression to the mean happened quickly, so that, according to a HoU for HR, 70 to 80 percent of evaluations were clustered between 14 and 16. Further, while the stated goal was to reward performance, what governed the system was actually a different norm: to maximize promotions. Almost all HR staff within the DGs and a number of managers reported that “of course” they were trying to promote as many people as possible. Priority points were often used not to reward top performers but to give officials close to promotion an extra push. One factor complicating the allocation of points and increasing the uncertainty of the system was that the thresholds (the numbers of points needed for promotion) varied by grade and were not set until near the end of the process, as they depended on the budget and the number of people within each grade who were close to promotion. Directors General clearly were invested in maintaining the flexibility and control of the process provided by the priority points, and they led the meetings of directors at which points were allocated across the DG. Central allocation of points obviously limited the evaluations that the Heads of Unit could write. In addition, staff associations met with Directors General, reviewed allocation of points, and advocated for those who they saw as being shortchanged. They saw their role as supporting “good soldiers,” the broad group in the middle, enforcing the commitment made to them Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 8 The Challenges of Pay and Promotion by Kinnock and included in the Staff Regulations, which lists the percentage of people in each grade who could expect to be promoted. Obviously, this reduced the number of points available for the high-fliers. While some HoUs took the whole process very seriously and devoted a great deal of time to it, almost all complained about how burdensome and “heavy” (a favorite word in the Commission) the process was. Initially their resentment was expressed quite loudly, as many felt this chore was competing for their time with what really counted: acting as policy entrepreneurs (Levy, 2006; Bauer, 2008, 2011), which was more interesting and prestigious. Further, their own incentive structures had rewarded their success as policy entrepreneurs (Ban, 2013). But over time, they did come, albeit sometimes grudgingly, to accept the necessity for evaluations. They also developed coping mechanisms, which for some meant delegating the lion’s share of the task to their deputies. In sum, the implementation was rocky, and the regression to the mean and rigidity of the system had a paradoxical effect. The importance of performance was actually reduced, slowing progress of super-stars up the grade ladder. Equity and motivation As we saw above, staff members are more likely to accept linking pay or promotion to performance appraisal results if they perceive that the system treats them fairly and that the rewards are meaningful. As expected, their initial enthusiasm reflected their perception that they would benefit from the system. Further, many, especially the newer hires, took the training seriously and really expected to be rewarded based on their actual performance. Within two years of the introduction of the CDR system, their disillusionment was quite deeply felt. They often blamed the zero-sum nature of the system and the fact that the points were established in advance of the formal evaluation for weakening the actual tie between their performance and the reward system. But what really irritated many of them was that the goal of rewarding performance was undermined. As one official described it, [In my DG] promotions for the whole DG are maximized, so this means they said ‘you get one more point if you are just about to be promoted, one less point if you have just been promoted, and the average otherwise.’ So there is absolutely no incentive to work more if you look at this. [Describes being on a major project and working long days under pressure]…So actually the two officials who have been working at this maybe we got marginally more, but definitely not worth the trouble. Several complained about the lack of transparency of the priority points. As one expressed it, “the system is open to abuse, it’s open to privilege and preference.” Also, its unpredictability undercut the confidence that they would get the rewards they felt they deserved. One person who was particularly incensed stated: What seems to be strange is this points system. … It’s a stupid system, I think, and it’s a discriminatory system, and it’s a fake system. It doesn’t encourage anybody I think…So it’s a 14 or 14.5, and then people…debate on this, but in the end you can change the threshold from 23 to 45 during one year. I’m just “What? What are you talking about? Go go, I don’t care.” Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 9 Carolyn Ban Nor was the reward so attractive as to motivate people to change their behavior. Several comments were similar to those from an official who said that in the training, “they told us that ‘If you work very hard, you get promoted one year earlier than the one who works on average. And the one who works on a very low level gets promoted one year later.’ So not much sense to try hard.” In sum, if the goal was to improve motivation or job satisfaction of officials, the reform was not very successful. Frequently, people described the system as being demotivating. One HR staff member in a DG told me that her DG did an annual staff satisfaction survey, but they always had to wait until several months after the CDR was completed, or else the results were heavily colored by the negative reactions to the evaluation process. Version 2: Trying again to get it right As we saw, version one created numerous problems, but the top managers focused most on the problem of centrality bias, i.e., the lack of variance in evaluations, which reduced the sense that performance was being rewarded and meant that even high-flyers were moving up the ladder slowly. When Siim Kallas was appointed Commissioner for Administrative Affairs, Audit, and Anti-Fraud and had an “away day” with the Directors General to ask for their input, they expressed their dissatisfaction with the staff appraisal system. According to several directors of HR in the DGs, pressure for change came from members of cabinets, that is, the small teams that supported each Commissioner, made up largely of officials seen as high flyers since serving on a cabinet is often the route to rapid promotion. But the result, which I dub version two, was an excellent example of the law of unintended consequences. Although significantly different, it suffered from the same ills as version 1, while introducing new problems that were even more serious. I briefly discuss the mechanics of the new system and look at the implementation challenges, effects on perceptions of equity, and impact on motivation. The mechanics of version 2: Forced distribution Since a central goal of this version was to reduce the centrality bias by forcing managers to make distinctions, the new system required managers to place each official into one of five categories. The top two, 1A and 1B, had quotas: A maximum of eight percent could receive a 1A rating, and 22 percent could get a 1B. Almost everyone else received a rating of 2, and levels 3 and 4 were almost empty. That meant that 70 percent of officials received what they perceived as a negative appraisal. Further, each category had attached to it a range of points. For 1A, the range was 11 or 12 points; for 1B, 8 to 10; and for 2, 5 to 7 points. In theory, the system was designed to get people away from focusing on points, but in reality, most people knew what points they had received. Analysis of Version 2: Implementation and impact Even more starkly than version 1, the greatest challenge with version 2 was the pressure to balance rewarding high-performing officials with ensuring that most officials would have a “normal career,” i.e., that they would eventually be promoted. The solution, which had massive unintended consequences, was not only forced distribution but also forced rotation. If the system were based purely on merit, there would probably be a rather high consistency in evaluations over time, as the Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 10 The Challenges of Pay and Promotion best performers often continue to excel. But since allocation of points was a zero-sum game, that would cause others to fall behind what was considered “normal” progression. Therefore, DG HR sent out guidance calling for at least 30 percent of officials to move up and another 30 percent to move down the categories. That meant people were told “Congratulations, you received a promotion this year, but next year don’t expect to get very many points.” The second source of problems, linked to the concept of a normal career, was the continued emphasis on ensuring as many promotions as possible through a process of “maximization,” in which people received extra points. As one HR staff person explained, A lot of people got at most 5, and then, with 'maximization' some people got a 9 because people felt sorry for them and wanted them to move up. And people would see this and say "I, who worked so hard, got only a 5, and this very person got a 9?" When we had to do it, we felt it was the fair thing to do if someone was 1 or 2 points from the threshold. We had to ensure the normal speed of promotion of "good soldiers.” According to several people, the result was “bizarre.” In fact, the results were exactly contrary to the stated goals of the reform. One HR staff person in a DG argued for changing it without success. She acknowledged that if “you just pick the person who is closest to the threshold [for promotion], irrespective of merit,…your bright people are going to pay the price, not just in year X, but…throughout the period.” The HR staffer maintained that this approach favored the average or slightly less-than-average performers, but “your best performers are going to get hit.” To make sure that the quotas were not exceeded, the assignment of people into the categories and allocation of points was even more tightly centralized than in version 1. The new version forced heads of unit to draft written appraisals that fit the points rather than reflecting the officials’ actual level of performance, thus undercutting honest dialogue between officials and their evaluators. This was also quite formally structured. As a head of unit described it: I felt inhibited – you were actually consulting the little guidebooks, “Okay Category 2. That means I cannot possibly say the person is doing anything outstanding because if it was outstanding it should be 1B” [It was] pathetic. The worst part was …that we still had to pretend that the categorization was based on performance. Other managers reported anger at being part of a “corrupt” system, which forced the “reverse engineering” of their appraisals. Version 2: Perceptions of equity and motivation As should be evident, neither the heads of unit nor the officials being assessed saw this system as evaluating work fairly or allocating points equitably. Individuals reported their distress at receiving a 5 or 6 rating, and managers expressed frustration at having to give such low ratings. As one HR head of unit told me, “one of my best ASTs (assistants)…got 11 last year and 5 this year. She’s even better this year than last year, but she had a promotion last year, so she’s tremendously demotivated.” The negative effects of this system were amplified because of a critical element of the organizational culture: the self-perception by the officials of the EU institutions that they constituted a professional elite, a group that had won their positions through the grueling competition process after years of training. As we saw above, previous research has found that pay for performance systems clash with most employees’ self-perception that they are above average. Of course, that is Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 11 Carolyn Ban a statistical impossibility, but when workers receive a lower rating than they think they deserve, they tend to blame the rater or otherwise fault the system. Thus, the system may motivate a few but at the cost of demotivating the majority (Meyer, 1975). That effect is greatly magnified in an environment where almost everyone has, throughout their education and early career, genuinely been above average. Even though logically most were average (or “good”) when compared to their peers, this clashed so sharply with their own sense of self-worth and of fair play that the psychological costs were very high. Reactions were dramatic and were demonstrated in both language and in action. A content analysis of the words used to describe the system found high consistency. There were no positive references. Negative descriptors included: subjective, pseudoscientific, ludicrous, absurd, crazy, and discriminatory. The system was criticized for hurting relations between people and harming teamwork. To sum up, as one person said, “It is about everything that is bad about management. It is the worst system I have ever come across.” Even more telling, the rate of appeals soared. Under version 1, initial high rates of appeals declined as people learned to live with the system, but under version 2, appeals remained very frequent. In one DG, I was told that over 20 percent had filed a formal appeal. In some cases, heads of unit who were forced to give what they saw as unfair ratings actually encouraged their subordinates to appeal. This level of appeals crated a crisis; it overwhelmed the process and was a major impetus for yet another attempt at getting it right. Version 3: A Better balance? The serious problems with version 2 led to wide-spread support for yet another change. According to one senior HR manager in a DG, version 2 had been pushed strongly by thenCommissioner Kallas and would probably be dropped as soon as he left that position. In 2010, Kallas moved from the position of Commissioner for Administrative Affairs, Audit, and Anti-Fraud (including HR) to become Commissioner for Transport. DG HR came under the political direction of Maroš Šefčovič, who was named European Commissioner and Vice-President for InterInstitutional Relations and Administration. In 2009, Irene Souka was named Director General for DG HR. Under their leadership, the change process moved quickly. According to an HR official within a DG, in 2011, even before the formal approval of the new system in the College of Commissioners DG HR was already introducing it with a heavy PR push across the Commission, so that the new system could be launched in time for 2012 staff appraisals. This final iteration in the series of reforms took the Commission back to where it started in several ways. The new system was qualitative, not quantitative, with no points and no rucksack. The automatic link between appraisal results and promotion was severed. This immediately solved one problem created by both of the previous versions – the high rate of appeals – as officials received appraisal results separately from the promotion decision and the basis of promotion decisions became considerably less transparent, making it more difficult to identify grounds for an appeal. There were, however, several ways in which version 3 differed from the status quo ante. First, the appraisal process did not return to a simple checking of boxes. The new system was designed to encourage dialogue between officials and their supervisors. The first step is selfappraisal: all officials complete a form requiring them to describe their accomplishments for the year in a series of six short essays. The instructions for the first one are below: _________________________________________________________________________ Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 12 The Challenges of Pay and Promotion Efficiency What were your main achievements of the past year? Explain the circumstances that contributed to these achievements (or made them more difficult) and the context (e.g. variety of tasks, changes during the year). Describe the impact of these achievements (How did your achievements contribute to the goals of the Commission?) If there were significant goals that were not achieved, what could have been done differently to achieve a different result (if anything)? Please cover the main achievements of your work for the Commission, including work you may have done which was not supervised by your Reporting officer. (Recommended maximum: 500 words) _____________________________________________________________________________ The other headings are: Ability Conduct Languages Responsibilities Learning (Optional) General comment As with versions 1 and 2, the supervisor then conducts his/her appraisal (which may consist of notes on the self-appraisal) and meets with the official. There is, however, one striking difference between version 3 and previous versions: The entire process is carried out by the heads of unit without review or input from their directors, and since heads of unit no longer must make their report fit the pre-assigned points, they are free to be more open and honest. As one head of unit reported, the current appraisals form “are better drafted, and you give a better feedback to people.” Another asserted that they were “more objective because they are more critical.” But of course, this system also gave supervisors the freedom to say more positive things as well, which could put pressure on the now loosely coupled promotion process. Directors interviewed were generally comfortable with the new system. One told me she had no problem not reviewing the forms, as heads of unit see who is doing well or not. She concluded, “Overall, I think it’s very fair to the head of unit to leave him or her to make that appraisal.” However, absent the quasi-automatic promotion process when staff had amassed the requisite number of points, the lack of points may actually have strengthened the role of Directors in the promotion process because Directors have to fight vigorously in favor of their staff in promotion negotiations at the DG level. Analysis of version 3 Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 13 Carolyn Ban Implementation and impact Implementation was a much easier process this time, since the system was simpler. As we have seen, the reform accomplished one key objective: the appeals rated dropped sharply. The broader impact issues are discussed below in the conclusions. Perceptions of equity While virtually all managers interviewed applauded the changes, officials were more mixed in their responses. The training they had received had stressed the values of transparency and predictability, provided by points, awarded publically, and placement of those points into a “rucksack” that moved with the individual. In the eyes of many, the new system lacked transparency. As one official put it: So basically at the end of the appraisal you have some sort of a report that you can verify yourself in the electronic system, and you can say, “Oh yes, I agree with what my superiors are saying.” But still I don’t know how close to the next promotion this is bringing me. Because they may be saying very nice things about me, but I don’t know what they are saying about my other colleagues. Maybe they are saying even nicer things. And eventually at the end of the three-year period, I don’t know if all those nice comments are enough to move me to the next level. So I feel this system is not transparent and not predictable. Still, in spite of some discomfort, most have accepted the latest changes, and the situation has finally stabilized. No one is predicting or advocating for any future changes. Conclusions Now that version 3 has been in place for five years, we can look back at this whole saga and assess what effect it has had on the Commission, specifically in terms of two central goals of the reform: to change the Commission’s management culture by redefining the role of Heads of Unit to strengthen their abilities and commitment to their roles as managers and to strengthen the relationship between officials’ performance and promotion, with the assumption that doing so would improve their motivation and job performance. Changing the role of the Head of Unit As we have seen, a central goal of the Kinnock Reforms was to move Heads of Unit from their historical role as policy entrepreneurs to greater emphasis on good management, including expanding their personnel management responsibilities, most importantly by requiring them to set clear objectives for officials and to conduct serious appraisals of their success in meeting those objectives. Version 3 reflects at least a moderate level of culture change among non-supervisory officials in that the training and expectations created by versions 1 and 2 shaped a perception that appraisal was important. Thus, officials expected and continue to expect that their supervisors will take it seriously. There were a few reports of heads of unit delegating the appraisals to their deputies, and a few who complained that the whole process was not worth the effort because the appraisal results no longer conferred points. But by and large, the change did preserve some of the benefits of versions 1 and 2, especially the self-appraisal and dialogue. As one person put it, “We have many fewer appeals, so there is less of a psychodrama in the organization.” The initial reactions of Heads of Unit to version 1 ranged from acceptance that appraisal was a normal management function to near-apoplexy and sharp anger at this imposition of a role that was seen as not only burdensome but demeaning. But over time, the anger cooled as managers Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 14 The Challenges of Pay and Promotion learned how to manage the process so that by version 2 their anger was redirected toward the absurdity of the system, as they argued for the flexibility to give honest and constructive feedback to their staff. As discussed below, the arrival of new managers also helped to change the culture. Strengthening the link between performance and promotion While there is ample evidence that the reform succeeded in changing management culture, at least moderately, the effects on officials being evaluated were far less clear. Linking performance to pay or to promotion is a complex process under any circumstances, and the formal systems introduced in versions 1 and 2 were more complex than most. Let us look first at whether either version strengthened the rewards for performance and then whether the reforms had an impact on motivation or job performance. The extent to which performance was accurately assessed and rewarded is inextricably tied to a central element of the traditional culture: the emphasis placed on seniority. While the reform rhetoric focused clearly on performance, the reformers were unable to eliminate the role seniority played. First, promotion largely by seniority is common in rank-in-person systems that assume career progression. Second, within the Commission, the culture strongly supported getting as many people promoted as possible, so that it was taken for granted that “of course” one would give more points to those close to promotion under both version 1 and 2, even if that meant giving mediocre staff higher scores than those who had outstanding years, which shocked those who had believed the rhetoric, and which clearly had a demotivating effect. Second, the EC staff are represented by several staff associations (or unions), and the support of at least some of them was critical for passage of the Kinnock Reforms. As is typical of unions, they defined their role as protecting the “good solider,” the solid staff member in the middle. Thus, the guarantee that most officials would be able to progress at a rate that would ensure a “normal career” was not stressed in the presentation of the new reforms, but it was written into the staff regulations. Throughout all three versions, the staff unions played an important role in the appraisal/promotion process, and in fact, union representatives played and continue to play a formal role in the decision process. As an informant from DG HR described it, It is helpful to have them in checking that rules are applied. To make sure that the DGs consider other candidates for promotion in a serious way, the unions meet with each DG before the promotion exercises. It definitely affects their behavior. One DG changed his mind and took a more inclusive approach. Most promotions are based on seniority with a few fast streamers. The stress on seniority and role of staff associations helped to undercut the stated reform goal of rewarding performance. Ironically, under version 1 it was probably harder than previously to move an outstanding staff member up quickly. Under version 2, even though some people did receive high scores and were promoted, the positive message was undercut at the same time. Some were explicitly told not to expect a good assessment in the following year, no matter what their performance was. In short, the Commission’s travails mirrored the experiences of other publicsector organizations, where similar systems failed because staff did not see the expected link between performance and reward (Perry et al., 2009). While my interview data were highly consistent about the lack of a perceived link between performance and promotion, it is difficult to move to the next logical step and to assess the impact of the system on morale, motivation, or job performance. There was never any attempt to Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 15 Carolyn Ban collect baseline data on any of these variables or to conduct any formal assessment of changes that could be attributed to the reforms. The absence of serious evaluation is very likely a signal that the goals of the reform were, in large part, symbolic, i.e., to rebuild confidence in the Commission and help it to recover from the damage to its reputation caused by scandal by adopting NPR-type reforms that paralleled those already adopted by the member states. In fact, what evidence there is points to a workforce that, while not insensitive to pay levels or to the benefits of rapid promotion, is nevertheless strongly motivated by values very close to what has been termed PSM (Public Service Motivation), which were largely unaffected by the repeated reforms of the appraisal/promotion systems. Formal reform and organizational culture: a longitudinal case study of repeated reforms Now that version 3 has been in place for several years, the net result is a system that gives top managers the flexibility to promote rising stars quickly and to hold back those who are mediocre or weak employees. Changing management culture is a slow process, reflecting both gradual acceptance by current managers and the arrival of new managers who have not bought into the traditional culture. In the case of the Commission, the simultaneous processes of reform and enlargement meant that there was an unusually large incoming group of new managers from outside, instead of promotion from within, and they brought a diversity of previous management experiences as well as increased demographic diversity (Ban, 2013). In the end, one can say that the role of managers and culture of management have changed, both because of “new blood” and because most had come to terms with the management functions they were required to carry out. In spite of the upheaval of frequent reforms, the role definition of heads of unit has changed as they have accepted a somewhat greater emphasis on management. Even under version 3, absent the formal link to promotion, appraisal is taken more seriously than before the reform process began. The appraisal format is more detailed, and the staff take seriously their own self-appraisal, thus, pushing managers to take their role seriously as well. Managers have come to terms with the increased emphasis on people management, and there is no sign (in spite of initial anxiety) of a weakening effectiveness of their policy role. In other words, these reforms can in the long run be seen as having had a weak impact, mostly positive, but certainly less wide-reaching than the dramatic change in culture promised by Kinnock. Taking the whole cycle of repeated reforms as an extended case study in the relationship between formal personnel system reforms and administrative culture, the circular relationship between formal administrative reforms and management culture is clear. While we do see moderate reform in management culture, the repeated reforms indicate the persistence of deep-seated cultural values that were strong enough to force changes in new administrative systems that clashed too sharply with those values. The current challenges to the organizational culture come less from formal system reform and more from budgetary pressures that are cutting both the number of officials and their benefits, forcing DGs to spin off many functions to semi-independent agencies, and perhaps, over the long run, reducing the attractiveness of employment within the European Commission (Georgakakis, 2014). Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 16 The Challenges of Pay and Promotion Methodological Appendix This appendix provides a brief overview of the research methodology used for the larger study of which this is a part. Research design: The data on officials’ reactions to the series of reforms reported here are drawn from an indepth study of the impact of enlargement of the European Union on the European Commission. That study analyzed the arrival of thousands of new staff into the Commission as a result of the admission of 10 new member states in 2014 and two more in 2017. This research relied primarily on in-depth semi-structured interviews, the majority of which (92) were conducted in 2006-2007, during a sabbatical year spent in Brussels. Given the slow recruitment process, a total of 56 additional interviews, some of which were follow-ups, were conducted through 2012, for a total of 148. The sample was roughly equally divided between those entering from the new member states and those from “old” member states. The interviews contained a series of questions on the Kinnock reforms, introduced in parallel with the enlargement (Ban, 2013), which enabled me to follow over time the introduction of repeated reforms of staff appraisal and promotion. The results of the study of enlargement (Ban, 2013) included a brief discussion of the reactions to the Kinnock Reform. This paper delves more deeply into the issues surrounding the series of reforms of staff appraisal. The research approach for the study as a whole was quasi-anthropological. While I could not embed myself in the EC as the authors of earlier anthropological studies of the culture of the Commission were able to do (Abélès et al., 1993; Shore, 2000; 2007), I could come close through a qualitative approach using in-depth semi-structured interviews. That meant it was necessary to rely on qualitative analytic methods, as the questions posed varied from one interview to the next depending on the grade level and personal life and work experience, and therefore, responses could not be merged into a data file in which all interviewees were responding to the same stimulus (Hannah and Lautsch, 2011). Rather than a positivist approach with formal hypothesis testing, this is an inductive study of an ongoing change process within a complex environment, which draws comparisons across three cases of reform within a single organization. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 17 Carolyn Ban Works Cited Abélès, Marc, Irène Bellier, and Maryon McDonald. 1993. “Approche anthropologique de la Commission européenne.” Report prepared for the European Commission. Amabile, T. 1997. “Motivating Creativity in Organizations: On Doing What You Love and Loving What you Do.” California Management Review. 40(1): 39-58. Ban, Carolyn. 2010. “Reforming the Staffing Process in the European Union Institutions: Moving the Sacred Cow out of the Road.” International Review of Administrative Sciences 76(1):524. Ban, Carolyn. 2013. Management and Culture in an Enlarged European Commission: From Diversity To Unity?. Houndsmills: Palgrave Macmillan. Ban, Carolyn and Wouter Vandenabeele. 2009. “Motivation and Values of Staff of the European Commission,” Paper presented at the meeting of the European Union Studies Association, Los Angeles, April. Bauer, Michael. 2007. “The Politics of Reforming the European Commission Administration.” in Management Reforms in International Organizations, edited by Michael Bauer and Christoph Knill. Baden-Baden: Nomos. Bauer, Michael W. 2008. “Diffuse Anxieties, Deprived Entrepreneurs: Commission Reform and Middle Management.” Journal of European Public Policy 15(5):691-707. Bellé, Nicola. 2015. “Performance-Related Pay and the Crowding Out of Motivation in the Public Sector: a Randomized Field Experiment.” Public Administration Review. 75(2): 230-241. Deci, Edward. 1971. “Effects of Externally Mediated Rewards On Intrinsic Motivation.” Journal of Personality and Social Psychology. 18 (1): 105-115. Deci, Edward, Richard Koestner, and Richard Ryan. 1999. “A Meta-Analytic Review of Experiments Examining the Effects of Extrinsic Rewards and Intrinsic Motivation.” Psychological Bulletin, 125(6): 627-668. Durham, Cathy and Kathryn Bartol, 2009. “Pay for Performance.” In Edwin Locke, ed. Handbook of Principles of Organizational Behavior. 2nd ed. Chichester; John Wiley and Sons. European Commission, 2000. Reforming the Commission: a White Paper, Part 1, COM (2000)200 final/2, April 5. Denhardt, Robert, Janet Denhardt, and Maria Aristigueta. 2013. Managing Human Behavior in Public and Nonprofit Organizations. 3rd ed. Los Angeles, Sage Publications. Georgakakis, Didier. 2014. “À la Fois Pilote et Victime de l’Austérité? Les Transformations de Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 18 The Challenges of Pay and Promotion l’Administration de l’Union européenne sous l’Effet de la Crise économique et financière,” Revue française d’Administration publique 151-152 : 805-824. Goggin, Malcolm, Ann Bowman, James Lester, and Laurence O’Toole, Jr. 1990. Implementation Theory and Practice: Toward a Third Generatioin. Glenview, IL: Scott, Foresman/Little Brown. Grøn, Caroline Howard. 2011. “Having it Both Ways? Balancing Knights and Knaves When Motivating European Commission officials.” Review of Public Personnel Administration. 31(4): 369-385. Hannah, David and Brenda Lautsch. 2011. “Counting in Qualitative Research: Why to Conduct it, When to Avoid it, and When to Closet it.” Journal of Management Inquiry. 20(1): 14-22. Hill, Michael and Peter Hupe. 2009. Implementing Public Policy. 2nd ed. London: Sage Publications Ltd. Ingraham, Patricia W. 1993. “Of Pigs in Pokes and Policy Diffusion: Another Look at Pay-forPerformance.” Public Administration Review, 53(4): 348-356. Kassim, Hussein. 2004a. “A Historic Accomplishment: the Prodi Commission and Administrative Reform.” in The Changing European Commission, edited by Dionyssis Dimitrakopoulos. Manchester, UK: Manchester University Press. Kassim, Hussein. 2004b. “The Kinnock Reforms in Perspective: Why Reforming the Commission is an Heroic, But Thankless Task.” Public Policy and Administration 19(3):25-41. Kellough, J. Edward. 2012. “Managing Human Resources to Improve Organizational Productivity: the Role of Performance Evaluation.” in Norma M. Riccucci, ed. Public Personnel Management: Current Concerns, Future Challenges. Boston: Longman. Kellough, J. Edward, and Haoran Lu. 1993. “The Paradox of Merit Pay in the Public Sector: Persistence of a Problematic Procedure.” Review of Public Personnel Administration 13(2):45-64. Kinnock, Neil. 2004. “Words of thanks and short presentation on 'The Reform of the European Commission'.” Eipascope 1:6-9. Kruger, Justin and David Dunning. 1999. “Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments.” Journal of Personality and Social Psychology. 77(6): 1121-1134. Levy, Roger. 2003. “Critical Factors in Public Management Reform: the Case of the European Commission.” International Review of Administrative Sciences 69:553-66. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 19 Carolyn Ban Levy, Roger. 2006. “European Commission Overload and the Pathology of Management Reform: Garbage Cans, Rationality and Risk Aversion.” Public Administration 84(2):423-39. Magee, Joe, Gavin Kilduff, and Chip Heath. 2011. “On the Folly of Principal’s Power: Managerial Psychology as a Cause of Bad Incentives. SciVerse ScienceDirect; Research in Organizational Behavior 31:25-41. Mathis, Robert, John Jackson, and Sean Valentine. 2014. Human Resource Management. Stamford, CT: Cengage Learning. Mazmanian, Daniel and Paul Sabatier. 1983. Implementation and Public Policy. Glenview, IL.: Scott, Foresman and Company. Merriman, Kimberly. 2010. “Lost in Translation: Cultural Interpretations of Merit Pay,” Compensation and Benefits Review. 42(5): 403-410. Milkovich, George and Alexandra Wigdor, eds., 1991. Pay for Performance: Evaluating Performance Appraisal and Merit Pay. Washington, DC: National Press. Milkovich, George and Jerry Newman. 2008. Compensation. 9th ed. New York: McGrawHill/Irwin. Nigro, Lloyd and J. Edward Kellough. 2014. The New Public Personnel Administration 7th ed. Boston: Wadsworth. Perry, James L. and Lois Recascino Wise. 1990. “The Motivational Bases of Public Service.” Public Administration Review 50(3): 367–73. Perry, James, Trent Engbers, So Yun Jun. 2009. “Back to the Future? Performance-Related Pay, Empirical Research, and the Perils of Persistence.” Public Administration Review, 69 (1): 39-51. Perry, James and Annie Hondeghem, eds. 2008. Motivation in Public Management: the Call to Service. Oxford; Oxford University Press. Pollitt, Christopher, and Geert Bouckaert. 2011. Public Management Reform: a Comparative Analysis. 3rd ed. Oxford: Oxford University Press. Porter, Lyman, and Edward Lawler. 1968. Managerial Attitudes and Performance. Homewood, IL: Irwin. Pressman, Jeffrey and Aaron Wildavsky. 1984. Implementation: How Great Expectations in Washington are Dashed in Oakland. 3rd edition. Berkeley: University of California Press. Sabatier, Paul. 1986. “Top-down and Bottom-Up Approaches to Implementation Research: A Critical Analysis and Suggested Synthesis.” Journal of Public Policy 6: 21-48. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 20 The Challenges of Pay and Promotion Schay, Brigitte and Sidney Fisher. 2013 “The Challenge of Making Performance-Based Pay Systems Work in the Public Sector,” Public Personnel Management. 42 (3): 359-384. Schön-Quinlivan, Emmanuelle. 2011. Reforming the European Commission. Houndsmills: Palgrave Macmillan. Shore, Cris. 2000. Building Europe: the Cultural Politics of European Integration. London: Routledge. Shore, Cris. 2007. "”European Integration in Anthropological Perspective: Studying the 'Culture' of the EU Civil Service.” in Observing Government Elites, edited by R. Rhodes, P. t'Hart, and M Noordegraf. Basingstoke: Palgrave MacMillan Stevens, Anne, and Handley Stevens. 2001. Brussels Bureaucrats? the Administration of the European Union. Houndsmills: Palgrave. Stevens, Handley, and Anne Stevens. 2006. “The Internal Reform of the Commission.” in David Spence and Geoffrey Edwards, eds. The European Commission. London: John Harper. Vroom, Victor. 1964. Work and Motivation. New York: John Wiley. Whitener, Ellen, Susan Brodt, Audrey Korsgaard, and Jon Werner. 1998. “Managers as Initiators of Trust: An Exchange Relationship Framework for Understanding Managerial Trustworthy Behavior.” The Academy of Management Review, 23(3): 513-530. Pittsburgh Papers on the European Union pgheupapers.pitt.edu | Vol. 4, No. 1 – October 2017 | https://doi.org/10.5195/ppeu.2017.30 21