Toward “Flexible Uniformity”? Civil Service Reform, “Big Government Conservatism,” and the Promise of the Intelligence Community Model

As the Obama administration pieces together its own civil service reform program, it may find solutions to key reform challenges in an oft-overlooked Bush administration human resource management initiative in the national security arena. While press and scholarly attention focused largely on the administration’s reform efforts at the Departments of Homeland Security and Defense, discussed at length in the article by Kellough, Nigro, and Brewer in this symposium, the development of a common personnel framework across the U.S. Intelligence Community went relatively unnoticed. The author argues that human resource management changes made pursuant to the Intelligence Reform and Terrorism Prevention Act of 2004 provide a potential model for the Obama administration as it addresses three key reform challenges that have long plagued policymakers: replacing the General Schedule with a modernized approach to compensation and classification, achieving a balance between uniformity at the executive branch level and flexibility at the agency level, and reconfiguring the Senior Executive Service.

As the editors note in the introduction to this symposium, a key component of President Bush's "big government conservative" agenda for the United States came packaged as a more assertive international presence. This came in the wake of 9/11, the Iraq invasion, and the subsequent war on terrorism. And at the heart of these efforts was the need for a robust intelligence effort to attenuate the stovepiping, lack of coordination, and fail ures to communicate among national intelligence agencies that the 9/11 Commission identified in its postmortem of the causes of the tragic events of that day (Zegart, 2009). 1 In short order, a balkanized human resource management (HRM) system also was identified as an obstacle to integration within the Intelligence Community (IC) and an effort was launched in 2006 to create a common set of personnel rules across the com munity. This initiative received much less attention, however, than did the HRM reform efforts at the newly created Department of Homeland Security (DHS) and at the Department of Defense (DoD) that Ed Kellough, Lloyd Nigro, and Gene Brewer discuss in their article in this symposium. With the broad scope of change incorporated in both the MaxHR system at DHS and the National Security Personnel System (NSPS) at DoD, the numbers of federal employees affected, and the public contro versy that surrounded both, MaxHR and NSPS came to represent the Bush administra tion's signature approach to civil service reform.
Thus, when in October 2009 President Obama signed the National Defense Auth orization Act for Fiscal Year 2010, thereby repealing NSPS, and with the preceding demise of the MaxHR system at DHS, the Bush administration's civil service reform legacy appeared to be one of "ashes." 2 However, just as David Rosenbloom suggests in his article in this symposium that President Bush's Supreme Court nominees and their subsequent rulings will constitute an HRM legacy for the Bush administration, I argue that the reform initiative within the IC may yet provide a more positive legacy of the Bush years. Specifically, the promise of the Bush administration's HRM effort to fashion big government conservatism in the national security arena lies in its potential for extending to the executive branch the IC's federated approach to reconciling three key challenges confronting wouldbe reformers of the civil service system: (a) replacing the General Schedule (GS) with new rules for compensating and classifying federal employ ees, (b) balancing the need for the standardization of pay and classification policies across the government and a simultaneous need for the customization of those same policies at the agency level, and (c) reconfiguring the Senior Executive Service (SES) to fulfill the origi nal vision of a corps of generalists whose careers would span agency lines.
The article begins with an overview of the primary HRM challenges facing any new administration in Washington. It then reviews the logic and tactics of the Bush admin istration's efforts to address these issues and summarizes why, according to conventional wisdom, it failed. The article concludes by reviewing how and why specific efforts taken in the national security arena-in particular, the common personnel framework developed within the IC subsequent to the passage of the Intelligence Reform and Terrorism Prevention Act-provide a model for addressing the three key reform chal lenges: replacing the GS, achieving a balance between uniformity and flexibility, and reconfiguring the SES.

Civil Service Reform Challenges
A central challenge in reforming the federal civil service is that of deciding how pay increases for federal employees should be apportioned, whether on the basis of lon gevity, performance, or the market. A number of key oversight groups have argued that individual performance be given greater weight in pay setting and that the tradi tional GS should therefore be abandoned in favor of broadbanding. A second chal lenge, related to the first, is that of the "balkanization" of the federal personnel system. This occurred as individual agencies opted out of various Title 5 rules, includ ing those governing compensation and classification, which have resulted in a patch work of personnel systems across the government. A third challenge is to foster greater cooperation and collaboration among members of the SES across agency lines. The following dissects these three challenges and reviews and critiques several com mon remedies that reform advocates have offered over the years.

The Obsolescence of the GS
The GS has long been criticized as an obsolete approach to classification and pay. Problems with the GS most often cited by critics include: (a) the classification stan dards at the heart of the system are outdated; (b) the 15grade system is overly com plex, allowing technicians a disproportionate impact on personnel decisions; (c) the system is relatively insensitive to the market; (d) the rigidities inherent in the system are an impediment to change; and (e) the system leads to insufficient distinctions in the pay levels of high and low performers.
In a 2002 report, the Office of Personnel Management (OPM) identified the GS as a "system whose time has come, and gone" (p. i). In the report, OPM noted that the GS system is anchored in the Classification Act of 1949 and that at the time the law was passed, "Over 70 percent of Federal whitecollar jobs consisted of clerical work . . . " (p. i). OPM added that although the GS worked well in a workforce that was predomi nantly clerical, it is not well suited for a workforce in which "knowledge work" domi nates and in which jobs are less well defined and distinguished (p. i).
A second argument for replacing the GS is that the excessive number of rules relat ing to classification and paysetting leaves key management decisions in the hands of staff personnel. A 1993 report on Reinventing Human Resource Management issued by the National Performance Review concluded that the GS results in "fragmented accountability": "The GS classification system is difficult to understand and to use. This prevents managers-who actually best know the work being classified and its value to the organization-from assuming the primary role in classifying jobs" (para. 12).
A third area of concern is that the rulebound and, hence, rigid nature of the GS system becomes an impediment as organizations attempt structural change. Broadbanding is more flexible in this regard. In its 1995 report, Modernizing Federal Classification: Operational Broadbanding Systems Alternatives, the National Academy of Public Administration noted that, "Broader bands allow managers to shift their workforce to new roles more easily. In the current rightsizing and downsizing mode, such flexibility would be of great benefit" (p. 1). Broadbanding as an approach to classification and compensation was first introduced as part of the "China Lake" demonstration project in 1980. 3 With broadbanding, multiple grades are combined into a single band, thus expanding the range of salaries associated with any one position. Thus, whereas within the GS a grade has a 30% range (i.e., a GS grade's maximum rate equals 130% of its minimum rate), bands often are as wide as 50%. Broadbanding obviates the need to make fine distinctions in levels of job difficulty and responsibility, and as a result, managers can play a larger role in classification decisions. Broadbanding also pro vides recruitment advantages, as qualified recruits can be offered salaries anywhere within the specified range rather than, as generally occurs within the GS, only at step one of the grade to which the position is assigned.
A fourth criticism that has been made of the GS is that it is insufficiently sensitive to the market. This is in part due to its preeminent focus on "internal equity"-that is, on ensuring that pay levels are the same for jobs with equivalent levels of difficulty and responsibility. The grade system is the primary mechanism for achieving internal equity. However, as OPM has observed, "Labor market shortages and excesses are described and analyzed in terms of occupations, skills, specialties, and locations, not grade level . . . " (2002, p. 14). Needed, according to OPM, is a pay system that pro vides a better balance between internal equity, external equity (i.e., relative to the market), and "contribution" equity (relating to individual performance).
Greater weight on contribution equity, of course, implies that more consideration be given to individual performance in pay setting. In its 2002 report, OPM describes the GS system as "performance insensitive": The way the Government delivers pay increases-and the relative value of those pay increases-reinforces the message that performance is secondary at best. In any given year, federal employees receive more pay increases for remaining on the rolls than for meeting or exceeding performance expectations. (p. 12) According to OPM, part of the problem with the step system that characterizes the GS is that the supervisor's choice is "binary"-that is, to either grant a step or not grant a step. In this system, virtually everyone gets their scheduled step increases. Most broadbanding systems, in contrast, do not include steps, and pay increases can be scaled according to level of performance. Moreover, although broadbanding does not necessarily imply pay for performance (PFP), most broadbanding systems do incorporate a closer link between pay and performance than exists with the GS. Early broadband systems, including that at China Lake, provided that pay increases be scaled according to level of performance. China Lake's appraisal system recognized five levels of per formance ranging from "performance that is demonstrably exceptional" to "substan tially below fully successful." Employees assigned the lowest rating were excluded from the governmentwide "comparability" increase as well as any individual incre ment, whereas those with the highest rating could receive a base salary increase of up to 6% plus a bonus of 6%.
Whether PFP as a generic concept is well suited to the public sector has been a subject of much controversy. A series of studies have raised questions about the viabil ity of PFP in the public sector (e.g., Kellough & Nigro, 2002;Kellough & Selden, 1997;Pearce & Perry, 1983;Perry, Engbers, & Jun, 2009). For example, in their eval uation of the federal Merit Pay System, Pearce and Perry conclude that: . . . merit pay, a traditional motivational technique in the private sector, has encountered difficulties that are peculiar to its public sector context. It is doubt ful that any merit pay system in the private sector has had to function in the face of an inherently ambiguous performance environment, tight budgetary restraints, freedom of information about individuals' salaries, diffuse authority for imple mentation, a major managerial succession, and significant changes in organiza tional goals. Yet, these types of factors are continuing features of the public sector context for merit pay. Although merit pay is desirable in principle, its effectiveness may be severely constrained or negated within the environment of the federal sector. (1983, p. 324) Notwithstanding Pearce and Perry's (1983) hypothesis nor Perry et al.'s (2009) suggestion that PFP deficiencies in the public sector are institutionally rooted, positive outcomes have been identified at multiple federal personnel demonstration projects that have incorporated both PFP and broadbanding. These include projects at China Lake, the National Institute for Standards and Technology (NIST), the Department of Commerce, and AcqDemo at DoD. For example, • At China Lake, the percentage of employees agreeing with the statement, "Pay raises here depend on how well you perform," increased from 47% in Year 1 to 60% in Year 10. In the control group, the corresponding percent ages were 46% in Year 1 to 40% in Year 10. Employee support for the project grew from 29% at the project's inception to 70% by Year 14 (Schay, 1996). • At the Department of Commerce, the percentage of employees agreeing with the statement, "Pay raises here depend on how well you perform," were as follows: the baseline went from 36% to 54% in Year 7. The control group percentages went from a baseline of 34% to 35% in Year 7. • At China Lake, the percentage of respondents agreeing with the statement, "High performers tend to stay with this organization," increased from 31% in Year 2 to 42% in Year 10, while declining from 29% to 24% in the control group. Similar patterns were found at DoD's AcqDemo project and at the NIST (Thompson & Seidner, 2008).

The Problem of Balkanization
As agencies have attempted to evade the constraints of the GS, alternative approaches to compensation and classification, most of which feature broadbanding, have been implemented. The broader problem thereby created has been the "balkanization" of the federal personnel system. In place of the "onesizefitsall" approach represented by the GS are multiple sets of personnel rules that have created governance concerns. John Berry, the Obama administration's initial director of OPM, commented on this phenomenon in a November 2009 speech. He noted that, "A significant and growing number of our employees are not in the GS pay system" (Berry, 2009, para. 25). Berry also has called the federal pay system "balkanized to the point of a risk of failure" (Rosenberg, 2009, para. 1). The proportion of the federal workforce exempted from portions of the Title 5 rules increased under both Presidents Clinton and Bush. During the 1995 to 2001 period, for example, Congress granted special flexibilities to the Internal Revenue Service, the Federal Aviation Administration, the Office of Student Financial Assistance in the Department of Education, and the Patent and Trademark Office in the Department of Commerce (Thompson, 2001). Thus, today, OPM lists over 30 agencies or subagency units totaling over 360,000 employees in its 2008 report on "alternative pay systems" (OPM, 2008a).
Although balkanization is consistent with the tenets of strategic human capital management (SHCM) to the extent that it allows agencies to tailor personnel rules to agency strategy, culture, and technology, there are problems with this approach from a governance perspective. For example, the principle of internal equity is compro mised as employees with similar qualifications, experience, and performance in different dep artments receive different levels of compensation. Also, the government is increas ingly divided into "have" and "havenot" organizations as departments and agencies able to secure hiring and pay flexibilities gain a competitive advantage over those left in Title 5. There also are practical problems, including that (a) transfers between dep artments and agencies become more complicated due to heterogeneous hiring stan dards and pay systems, (b) oversight is more difficult in a system with multiple sets of rules, and (c) the government ends up paying more than it otherwise would for employees with scarce skill sets due to competition between departments and agencies (Thompson, 2001).
As such, a 2004 Government Accountability Office (GAO) report endorsed the creation of a governmentwide "framework" within which customization could occur. The collective opinion of participants convened in a forum on this topic was summa rized in the report as follows: There was widespread recognition that a "one size fits all" approach to human capital management is not appropriate for the challenges and demands govern ment faces. However, there was equally broad agreement that there should be a governmentwide framework to guide human capital reform built on a set of beliefs that entail fundamental principles and boundaries that include criteria and processes that establish the checks and limitations when agencies seek and implement their authorities. (GAO, 2004, p. 2;also see OPM, 2004, p. i) The SES: Specialists or Generalists?
The SES, as created under the Civil Service Reform Act (CSRA) of 1978, was to serve as a corps of generalist executives whose careers would traverse agency lines and who thus would promote interagency collaboration and cooperation (Huddleston, 1992). Instead, the vast majority of SES members spend their entire careers in a single agency. In a 2009 report, the Partnership for Public Service made the following comment: Most career senior executives remain in the same agency and do not move through the government to share their expertise or provide an enterprisewide, collaborative approach as envisioned by the 1978 law. In fact, only 2.3 percent or fewer members of the SES left their jobs annually between 2004 and 2008 for another SES position in a different agency. (p. ii) Concerned that the "SES as a whole . . . is not providing a corporate or enterprise wide view of the federal government," the partnership recommended that the SES be divided into a "'National SES Corps' of mobile managers who will have rotational assignments and clear expectations that they will work in multiple agencies . . . " and an "agencybased SES" corps (p. ii).

The Bush Administration's HRM Reform Agenda
With these challenges, proposed reforms, and critiques as the context of federal per sonnel policy on taking office the Bush administration's HRM reform program relied heavily on a set of ideas associated with the concept of SHCM. This, too, was a con cept that had gained currency within the federal government over the past 10 to 15 years. Currency within the federal government over the past 10 to 15 years. Fundamental ideas that underlie this doctrine include: (a) personnel "practices and policies differ widely across organizations and across employee groups within orga nizations," (b) performance is enhanced to the extent that personnel practices and policies "fit" an organization, and (c) line managers are permitted to exercise greater discretion with regard to HRM achieving that fit (Milkovich, 1988, p. 263).
Within the federal government, the GAO has taken the lead in promoting SHCM concepts. In 2001, GAO added "Strategic Human Capital Management" to its list of "high risk" management challenges, and in 2002, it identified the following key tenets of SHCM: • Agency leaders view people as an important enabler of agency performance.
• Human capital professionals partner with agency leaders and line managers in developing strategic and program plans. • The agency's human capital approaches demonstrably support organizational performance objectives. • Decisions involving human capital management and its link to agency results are routinely informed by complete, valid, and reliable data. • Agency strategies for investing in human capital are fully integrated with needs identified through its strategic and annual planning. • The agency tailors its human capital strategies to meet its specific mission needs. • Managers, teams, and employees at all levels are given the authority they need to accomplish programmatic goals. • The organizational culture is resultsoriented and externally focused. (2002, pp. 1013) The Bush administration was selective in its application of SHCM concepts. The section of The President's Management Agenda (OMB, 2002) on "Strategic Management of Human Capital" focused on just three reform elements: • The need for agencies to link "human capital strategies" to "organizational mission, vision, core values, goals, and objectives" (p. 14) and to engage in workforce planning to "correct skills imbalances, and provide other tools to recruit, retain, and reward a highquality workforce" (p. 13). • The abandonment of a "onesizefitsall approach" (p. 11) to HRM. This refers to the need to allow agencies to deviate from the Title 5 rules for the purpose of tailoring personnel practices to agency mission, culture, and technology. • Support for "performanceoriented compensation" (p. 13) whereby the pay of federal workers would be linked to an assessment of their performance.
A second document in which key elements of the Bush HRM strategy were articu lated was a policy paper issued in January 2001 by the Heritage Foundation entitled, Taking Charge of Federal Personnel (Nesterczuk, Devine, & Moffit, 2001). Although not a formal expression of policy, the substance of the paper's recommendations figured centrally in the administration's HRM strategy. Moreover, the report's chief author was subsequently appointed to a position in the OPM overseeing the design of the NSPS.
In the report, Nesterczuk et al. (2001) were critical of the federal employee unions for their role in obstructing change to the HRM status quo. They criticized President Clinton for promoting labor-management partnerships that, in their view, "elevated fed eral unions to equality with management" (p. ii) and that enabled the federal employee unions to "become a counterweight to the political management appointed by the President" (p. 4). Accordingly, Nesterczuk et al. recommended that, the "new President will need to revoke President Clinton's executive order [on partnership] and demon strate from the outset that his approach to reforming the federal bureaucracy will emphasize political responsibility and accountability to the taxpayers"(2001, p. 4). President Bush followed their recommendation by issuing Executive Order 13203 in February 2001, formally revoking Executive Order 12871, which had created a National Partnership Council and which had mandated the creation of partnership councils across the government (Clinton, 1993).
The principles articulated in the President's Management Agenda (OMB, 2002) and in Taking Charge of Federal Personnel (Nesterczuk et al., 2001) took concrete form in three major and illfated initiatives by the Bush administration. In the after math of 9/11 came the new personnel systems designed first for the DHS-the MaxHR system-and then for DoD-the NSPS. Indeed, in no other arena of public policy was the approach to big government conservatism plied with more vigor, scope, or (ulti mately) frustration by the Bush administration as the realpolitik of personnel manage ment in the federal government overtook the bestlaid plans of its proponents.
As Ed Kellough and his colleagues recount in detail in their article in this sympo sium, MaxHR was promoted as advancing a "flexible" and "contemporary" HRM system at DHS that adhered to longstanding provisions of civil service law relating to merit principles, prohibited personnel practices, and veterans' preference. DHS, how ever, was allowed a substantially free hand with regard to matters such as performance appraisals, classifications, pay rates and systems, labor-management and employee relationships, adverse actions, and appeals (GAO, 2003). In short order, the labormanagement relationships section of the legislation became the focus of a legislative battle when Senate Democrats attempted to include a provision limiting the presi dent's authority to waive collective bargaining rights for DHS employees. The battle was resolved in the president's favor subsequent to the loss of two Democratic Senate seats in the 2002 elections (Brook & King, 2007).
Likewise, Congress authorized DoD to create a new personnel system for civilian employees pursuant to the provisions of the National Defense Authorization Act of 2003. The flexibilities allowed were similar to those for DHS except that DoD gained flexibility not provided to DHS in matters of hiring and reductioninforce. Similar to the strategy pursued with regard to MaxHR, the federal employee unions filed suit in federal court arguing that the labor-management relationships provisions of the pro posed regulations would violate the statutorily guaranteed collective bargaining rights of employees. DoD eventually won in court, but it was a hollow victory. In December 2007, Congress passed and the president signed the Fiscal Year 2008 Defense Authorization Bill, which restored traditional collective bargaining rights for DoD employees (Ballenstedt, 2007). After the election of President Obama, the unions pressed their political advantage, and in October 2009, the President signed the National Defense Authorization Act for Fiscal Year 2010, officially repealing NSPS and providing for the gradual transition of employees back to the GS by January 1, 2012.

Finally, the Bush administration's efforts to repeal President Clinton's Executive
Order 12871 mandating the creation of labor-management "partnership" councils throughout the government met a similar fate. The administration also had taken a number of actions that had the effect of limiting the collective bargaining rights of fed eral employees (Thompson, 2007). In January 2002, Bush withdrew the collective bargaining rights of 500 employees in the Department of Justice, primarily in the offices of the U.S. Attorneys and in the Criminal Division (Slater, 2004;Tobias, 2004). In January 2003, the collective bargaining rights of the employees of the National Imagery and Mapping Agency were withdrawn on national security grounds, and the head of the Transportation Security Administration (TSA), James Loy, issued an order stating that TSA employees, "in light of their critical national security responsibilities, shall not . . . be entitled to engage in collective bargaining or be represented for the purpose of engaging in such bargaining by any representative or organization," on the grounds that "[m]andatory collective bargaining is not compatible with the flexibility required to wage the war on terrorism" (Slater, 2004, p. 3). In September 2009, legisla tion passed the House Oversight and Government Reform Committee to restore full collective bargaining rights to TSA. On December 9, 2009, President Obama signed an executive order reestablishing labor-management partnerships with the executive branch (Parker, 2009a(Parker, , 2009b.

Toward Flexible Uniformity? Civil Service Reform, Joint HRM Capabilities, and the IC
With the repeal of NSPS, the failure of implementation of MaxHR, and successful actions to counter the labor-management relationships initiatives, the Bush adminis tration's HRM reform efforts were widely perceived as substantially in shambles as Barack Obama assumed the presidency. However, with much less attention than was accorded either MaxHR or NSPS, an initiative by the Bush administration to create a common HRM policy framework for elements within the IC provides a potential model for the Obama administration as it confronts the challenge of civil service reform.
Although no comprehensive civil service reform proposal has been forthcoming to date, the Obama administration has expressed an intent to reform several elements of the current system. One priority is to modernize the government's core pay and clas sification system-the GS. In a November 2009 speech, the director of the OPM, John Berry, referenced, "an historic opportunity for comprehensive reform of our civil ser vice system" (para. 4). In referring to the GS, Berry stated, "Five decades after the last major attempt at pay reform, the cracks are showing," adding, "We could limp along for a few more years in the current GS system or we can seize this moment to build something new" (2009, para. 25).
It is with regard to the challenge of finding the "right balance between flexibility and uniformity" (OPM 2004, p. i) that the IC's HRM initiative becomes relevant. Recall that in reviewing the intelligence failures underlying the 9/11 attacks, the 9/11 Commission (National Commission on Terrorist Attacks Upon the United States, 2004, p. 400) rec ommended the creation of a new position of Director of National Intelligence (DNI) with responsibility for the overall intelligence budget, with the authority to hire and fire senior managers, and with the "ability to set standards for the information infrastructure and personnel" (p. 410). Among other things, the Intelligence Reform and Terrorism Prevention Act of 2004 instructed the DNI to create common personnel policies and programs for the IC. The intent was that such a framework would foster collaboration across agency lines, for example, by facilitating the transfer of personnel between agen cies. Historically, members of the community have had substantial autonomy in HRM matters. This has resulted in a patchwork of different systems that have impeded the transfer of personnel between agencies and that have contributed to parochialism on the part of its members. Consistent with the general limits placed on the scope of the DNI's authority, however, the legislation did not authorize the DNI simply to impose new HRM policies on the various agencies. Rather, the DNI was to "prescribe" such policies, leaving the exact scope of his authority in this regard somewhat ambiguous.
The Office of the Director of National Intelligence (ODNI) took the lead in the effort to develop a common HRM framework across the IC. As ODNI did not have the authority simply to impose a system of its own design, a long series of negotiations took place among the 17 intelligence "elements." Although ODNI's efforts met with some resistance, IC members eventually were able to agree on a set of policies in the form of Intelligence Community Directives, or "ICDs." These included ICD 651, Performance Management System Requirements for the Intelligence Community Workforce; ICD 650, National Intelligence Civilian Compensation Program: Guiding Principles and Framework; and ICD 653, PaySetting and Administration Policies for the Intelligence Community Workforce. The key challenge in developing policy in each of these areas was to find a suitable balance between the desires for a standardized policy across the community and for flexibility in adapting those policies to the needs of individual agencies.

Performance Management
With regard to performance management, the 17 elements of the IC agreed that: • All IC employees would be evaluated on the basis of both performance "objectives" and performance "elements." Performance objectives relate to each individual and are supposed to be specific and measurable, whereas performance elements are generic and apply to everyone. Each are weighted equally in the overall evaluation. • Six performance elements-accountability for results, communication, criti cal thinking, engagement and collaboration, personal leadership and integ rity, and technical expertise-are standard across the community. • All IC (civilian) employees are assessed according to a common set of per formance elements to contribute to a sense on the part of employees that they are being treated fairly compared with their brethren in other components.
• Each performance element also represents specific values and behaviors and as such reinforces the performanceoriented and collaborative culture that the DNI is attempting to promote. • Although, pursuant to ICD 651, all IC agencies must have six performance ele ments in their performance appraisals, the ICD also provides that agencies can supplement these six elements. For example, the Central Intelligence Agency (CIA) will add "continuous learning" as a seventh performance element.
The transition to the new performance management requirements was fairly straightforward for agencies such as the CIA, the National Security Agency, and the Defense Intelligence Agency, as their missions are related to the collection and analy sis of intelligence. However, others of the IC elements are embedded in agencies or departments whose missions extend beyond just intelligence. Included in this category are the Federal Bureau of Investigation, the Drug Enforcement Administration, and the Office of Intelligence and Analysis (OIA) in the Treasury Department. Integrating the requirements imposed by ICD 651 with the systems already in place within each organization has posed a challenge. For example, whereas ICD 651 identifies six core competencies, the "Headquarters Departmental Offices" (DO) unit at Treasury, includ ing OIA, has only four. Treasury entered into a memorandum of understanding (MOU) with ODNI agreeing that OIA employees would be appraised by the three DO compe tencies that correspond to IC competencies supplemented by the other three IC com petencies. There were similar MOUs between ODNI and the other Title 5 agencies, such as the Department of Energy and the Department of State. ICD 651 further defines five levels at which IC employees are to be rated as follows: unacceptable, minimally successful, successful, excellent, and outstanding. Rating Level 2, "mini mally successful," is optional. Agencies are allowed to supplement the generic ratings standards included in the directive.

The National Intelligence Civilian Compensation Program
The centerpiece of the IC's National Intelligence Civilian Compensation Program is a paybanding/PFP system modeled after one that has been in place at the National GeospatialIntelligence Agency (NGA) since 1999. Gaining agreement on a common paybanding system was a significant milestone for the IC. The result of the IC's deliberations on these matters is depicted graphically in Figure 1. The band structure is based on an "occupational taxonomy" that was developed to ensure consistency and comparability in job categories across the IC. The taxonomy includes seven mission categories and three work categories (career groups), as well as six work "levels." Also, similar to previous efforts, the IC assigns each employee to a "pay pool." Each pay pool manager is allocated a specific amount for performancebased pay increases annually based on the amount spent historically on withingrade pay increases, quality step performance awards, and promotions. Pay pool allocations have typically totaled approximately 2.4% of payroll. Distinctive to the IC's system, however, is the use of a mathematical formula to translate each individual's summary performance rating into a pay increase.
Also distinctive are extra "steps" added to each band to accommodate the concerns of several agencies that employees at the top of their grade should have to compete for performancebased pay increases. The IC's payband system gives agencies discretion to (a) assign each position to a job category that determines the band to which the posi tion is assigned, (b) determine the starting and end points of its own bands as long as those bands fall within ranges shown, and (c) determine where the "control point" falls within each band. The control point represents the mean salary of positions with simi lar duties and responsibilities in the private sector.
In general, the salaries of employees below the control point rise more rapidly to the midpoint than those of employees that are above the control or midpoint. In other words, pay progression slows the higher one is in the band. A problem with the new system from the standpoint of community integration, however, is that those IC com ponents located in agencies with personnel procedures that are governed by Title 5such as the OIA in the Department of Treasury, the Office of Intelligence and Coun terintelligence at the Department of Energy, and the Bureau of Intelligence and Research at the Department of State-have not yet received legislative authority to implement pay banding and, hence, will continue to operate under the GS rules. Finally, the IC's payband structure also is the first "federated" payband structure within the federal government. Federation in this context implies a balance between elements Figure 1. The intelligence community's pay-band structure Source: Thompson and Seidner (2009). that are standard across the IC and those in which agencies have discretion. It is an example of the type of balance "between flexibility and uniformity" referred to by OPM (2004) as key to personnel system reform.
Although the IC's system addresses only issues of performance management, clas sification, and pay, these elements, along with hiring, have been at the center of agency requests for exemptions from Title 5 rules. That the system was designed and adopted by consensus ensures that both community and agency needs are addressed. The IC's system is sufficiently generic that, with minor modifications, it could be implemented across the entire executive branch. The system therefore provides a potential model for the Obama administration as it embarks on a search for a new governmentwide pay system. Table 1 presents a summary of the IC's federated approach to HRM showing those policy elements that are standard across the community and those in which agen cies have discretion.

Reforming the SES
A final element of the IC reforms that is promising for any effort at civil service reform during the Obama administration is the Joint Duty program. The 9/11 Commission recommended the creation of a Joint Duty program as a means to encourage collabo ration and to break down cultural barriers between agencies within the IC. The IC's program is modeled after that introduced within the military pursuant to the GoldwaterNichols Act of 1986. Under the terms of this program, anyone aspiring to membership in the SES within the IC must have served a 12month tour of duty in an agency other than their home agency.
As noted earlier, although the SES was designed by the CSRA partly to serve as a corps of generalist executives who would promote interagency collaboration and cooperation by working in various agencies during their careers, most have not done so. An SESwide Joint Duty program along the lines of what the IC has created holds promise for a return to CSRA's original vision. The purpose would be to broaden the perspective of those who reach the top levels of the career service and to foster col laboration among agencies in addressing issues that cross agency lines. Consistent with the IC's model, aspirants to the SES would be required to take assignments in other agencies as a condition of promotion.
An executive order issued by President Bush in May 2007 provides a precedent for such an approach. Executive Order 13434, titled "National Security Professional Deve lopment," would extend the joint duty concept to senior civilians within the national security community. The executive order directs the development of a National Strategy for the Development of Security Professionals and specifies that the plan provide for "interagency and intergovernmental assignments" (Bush, 2007). In November 2008, OPM issued policy guidance encouraging agencies to implement a "qualification requirement" for national security professionals in SES positions relating to "inter agency" experience (OPM, 2008b). According to what performance elements will employees be evaluated?
What mechanism will be used to deter supervisors from granting too many high ratings?

Policy
The evaluations of all employees are based 50% on results and 50% on behaviors. There are five rating levels.
All employees assessed according to six common performance elements.
Supervisors and managers to be rated on how well they manage the performance of their employees.
Areas of agency discretion Agencies decide the results measures. Agencies decide whether to use the "minimally successful" rating level.
Elements may establish additional performance elements.
Agencies do the actual ratings of the supervisors. (continued)

Pay-band structure
Policy question How many career groups will there be? How many pay bands are there for each career group?
How wide are the pay bands?
How does an employee get promoted from one band to the next?

Policy
There will be three career groups. Each agency maps its employees to the bands. There will be three pay bands for the supervision/ management and technician/administrative support groups and four pay bands for the profession group. The pay bands will extend from the equivalent GS steps shown in Figure 1.
No policy.

Areas of agency discretion
None None Agencies determine their own bandwidths as long as they fall within the community-wide parameters.
This is an agency determination.

Pay setting Policy question
What guarantees are there that employees will receive at least the amount of the annual comparability increase? Do low-performing employees still receive the annual comparability increase?
What is the minimum pay increase for an employee being promoted from one band to another?
How big are the performance pay pools?
Policy Employees with annual rating of "successful" or higher receive pay increases not less than the adjustment to the rate range.
Employees with a rating of "unacceptable" receive no pay increase.

Conclusion: The Bush Legacy, the Obama Administration, and the Next Wave of Reform?
This article has argued that with the repeal of NSPS and with the federal employee unions in temporary ascendancy in Washington, the Bush HRM legacy may reside in significant part with the federated personnel system developed by the IC. Whether this element of the Bush legacy bears fruit during the Obama administration remains unclear. As noted, OPM chief John Berry (2009) has adopted a proreform disposition, identifying in a 2009 speech "an historic opportunity for comprehensive reform of our civil service system" (para. 4). When Berry's predisposition is combined with President Obama's stance in sup port of PFP for teachers-and the dysfunctional aspects of the current system-the possibility of a civil service reform initiative cannot be discounted. One option would be for the Obama administration to couple pay and classification reform with the grant ing to federal employee unions of the right to bargain over pay. An agreement along these lines under consideration by President Clinton was scuttled when the Republicans took control of Congress in 1995 (Barr, 1995). The advantages to the Obama adminis tration would be both the union support that would be achieved and the replacement of an aging and obsolete system with one that would arguably facilitate the hiring and retention of high performers, provide managers with greater flexibility in pay and clas sification matters, and simplify classification processes.
As with its immediate predecessors, if the Obama administration decides to tackle civil service reform, it will confront the challenge of presenting a regime that achieves the "right balance between flexibility and uniformity" as identified by OPM in 2004. Just as the GS is regarded as overly rigid and centralized, the current trend toward balkanization reveals the shortcomings of excessive decentralization. The ideal sys tem would feature a "federated" element with the granting to agencies of greater flex ibility in matters of pay and classification than is provided under the GS while preserving governmentwide rules and procedures. As the only example of federation now extant, the IC's system provides a natural model. Also, in providing an overarching frame work that individual agencies can tailor to their needs, the broadbanding component of that system serves as an example of how federation would work.
Not unlike President Bush's big government conservative agenda, President Obama's agenda is a bold one, albeit one informed by a decidedly different philosophical bent concerning the role of the state. However, just as was understood by George W. Bush, big agendas of whatever philosophical disposition require big thinking about how best to pursue a difficult balance between the advantages of HRM flexibility and HRM unity as that agenda sprawls across federal agencies. The Obama administration need not look far to see the rudiments of such a system in aspects of the IC reforms pursued by its immediate predecessor.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this article.

Funding
The author(s) received financial support from the IBM Center for the Business of Government for this research.