绩效考核的困境外文翻译及原文

原文一：

The Dilemma of Performance Appraisal

By Peter Prowse and Julie Prowse

It will outline the development of individual performance before linking to performance management in organizations.The outcomes of techniques to increase organizational commitment, increase job satisfaction will be critically evaluated. It will further examine the transatlantic debates between literature on efficiency and effectiveness in the North American and the United Kingdom) evidence to evaluate the HRM development and contribution of performance appraisal to individual and organizational performance.

Appraisal potentially is a key tool in making the most of an organisation’s human resources. The use of appraisal is widespread estimated that 80–90%of organizations in the USA and UK were using appraisal and an increase from 69 to 87% of organisations between 1998 and 2004 reported a formal performance management system (Armstrong and Baron, 1998:200).There has been little evidence of the evaluation of the effectiveness of appraisal but more on the development in its use. Between 1998 and 2004 a sample from the Chartered Institute of Personnel and Development (CIPD, 2007) of 562 firms found 506 were using performance appraisal in UK.

What is also vital to emphasise is the rising use of performance appraisal feedback beyond performance for professionals and managers to nearly 95% of workplaces in the 2004 WERS survey (seeTable 13.1).Clearly the use of Appraisals has been the development and extension of appraisals to cover a large proportion of the UK workforce and the coverage of non managerial occupations and the extended use in private and public sectors.

Critiques of appraisal have continued as appraisal shave increased in use and scope across sectors and occupations. The dominant critique is the management framework using appraisal as an orthodox technique that seeks to remedy the

weakness and propose of appraisals as a system to develop performance.

This “orthodox” approach argues there are conflicting purposes of appraisal (Strebler et al, 2001). Appraisal can motivate staff by clarifying objectives and setting clear future objectives with provision for training and development needs to establish the performance objective. These conflicts with assessing past performance and distribution of rewards based on past performance (Bach, 2005:301).

Employees are reluctant to confide any limitations and concerns on their current performance as this could impact on their merit related reward or promotion opportunities(Newton and Findley, 1996:43).This conflicts with performance as a continuum as appraisers are challenged with differing roles as both monitors and judges of performance but an understanding counsell or which Randell(1994)argues few manager shave not received the raining to perform.Appraisal Manager’s reluctance to criticise also stems from classic evidence fromMcGregor that managers are reluctant to make an egative judgement on an individual’s performance a sit could be demotivating,leadto accusationsoftheirown supportand contributiontoindividual poor performance and to also avoid interpersonal conflict (McGregor, 1957).

One consequence of this avoidance of conflict is to rate all criterion as central and avoid any conflict known as the central tendency.In a study of senior managers by Long neckeretal.(1987),they found organisational politics influenced ratings of 60 senior executives.The findings were that politics involved deliberate attempts by individuals to enhance or protect self-interests when conflicting courses of action are possible and that ratings and decisions were affected by potential sources of bias or inaccuracy in their appraisal ratings (Longeneckeret al., 1987).

There are methods of further bias beyond Longenecker’s evidence. The political judgements and they have been distorted further by overrating some clear competencies in performance rather than being critical across all rated competencies known as the halo effect and if some competencies arelower they may prejudice the judgment acrossthe positive reviews known as the horns effect (ACAS, 1996).

One consequence of this avoidance of conflict is to rate all criterion as centralThereare methods Some ratings may only cinclude recent events and these

are known as the recency effects. In this case only recent events are noted compared to managers gathering and using data throughout the appraisal period .A particular concern is the equity of appraisal for ratings which may be distorted by gender ,ethnicity and the ratings of appraisers themselves .A range of studies in both the US and UK have highlighted subjectivity in terms of gender (Alimo-Metcalf, 1991;White, 1999) and ethnicity of the appraise and appraiser(Geddes and Konrad, 2003). Suggestions and solutions on resolving bias will be reviewed later.

The second analysis is the radical critique of appraisal. This is the more critical management literature that argues that appraisal and performance management are about management control(Newton and Findley, 1996;Townley, 1993). It argues that tighter management control over employee behaviour can be achieved by the extension of appraisal to manual workers, professional as means to control. This develops the literature of Foucault using power and surveillance. This literature uses cases in examples of public service control on professionals such a teachers (Healy, 1997) and University professionals(Townley, 1990).

This evidence argues the increased control of public services using appraisal as a method of control and that the outcome of managerial objectives ignores the developmental role of appraisal and ratings are awarded for people who accept and embrace the culture and organizational values . However, this literature ignores the employee resistance and the use of professional unions to challenge the attempts to exert control over professionals and staff in the appraisal process (Bach, 2005:306).

Thisevidence argues One of the different issues of removing bias was the use of the test metaphor (Folgeretal.,1992).This was based on the assumption that appraisal ratings were a technical question of assessing “true” performance and there needed to be increased reliability and validity of appraisal as an instrument to develop motivation and performance. The sources of rater bias and errors can be resolved by improved organisational justice and increasing reliability of appraiser’s judgement.

However there were problems such as an assumption that you can state job requirements clearly and the organization is “rational” with objectives that reflect

values and that the judgment by appraisers’ are value free from political agendas and personal objectives. Secondly there is the second issue of subjectivity if appraisal ratings where decisions on appraisal are rated by a “political metaphor”(Hartle, 1995).

This “political view” argues that a appraisal is often done badly because there is a lack of training for appraisers and appraisers may see the appraisal as a waste of time. This becomes a process which managers have to perform and not as a potential to improve employee performance .Organisations in this context are “political” and the appraisers seek to maintain performance from subordinates and view appraises as internal customers to satisfy. This means managers use appraisal to avoid interpersonal conflict and develop strategies for their own personal advancement and seek a quiet life by avoiding censure from higher managers.

This perception means managers also see appraisee seeks good rating and genuine feedback and career development by seeking evidence of combining employee promotion and pay rise.This means appraisal ratings become political judgements and seek to avoid interpersonal conflicts. The approaches of the “test” and “political” metaphors of appraisal are inaccurate and lack objectivity and judgeme nt of employee performance is inaccurate and accuracy is avoided.The issue is how can organisations resolve this lack of objectivity?

Grint(1993)argues that the solutions to objectivity lies in part with McGregor’s (1957) classic critique by retraining an d removal of “top down” ratings by managers and replacement with multiple rater evaluation which removes bias and the objectivity by upward performance appraisal. The validity of upward appraisal means there moval of subjective appraisal ratings.This approach is also suggested to remove gender bias in appraisal ratings against women in appraisals (Fletcher, 1999). The solution of multiple reporting(internal colleagues, customers and recipients of services) will reduce subjectivity and inequity of appraisal ratings. This argument develops further by the rise in the need to evaluate project teams and increasing levels of teamwork to include peer assessment. The solutions also in theory mean increased closer contact with individual manager and appraises and increasing services linked to customer facing evaluations.

However, negative feedback still demotivates and plenty of feedback and explanation by manager who collates feedback rather than judges performance andfail to summarise evaluations.There are however still problems with accuracy of appraisal objectivity asWalker and Smither (1999)5year studyof 252 managers over 5 year period still identified issues with subjective ratings in 360 degree appraisals.There are still issues on the subjectivity of appraisals beyond the areas of lack of training.

The contribution of appraisal is strongly related to employee attitudes and strong relationships with job satisfaction(Fletcher and Williams, 1996). The evidence on appraisal still remains positive in terms of reinvigo rating social relationships at work (Townley,1993)and the widespread adoption in large public services in the UK such as the national health Service (NHS)is the valuable contribution to line managers discussion with staff on their past performance, discussing personal development plans and training and development as positive issues.One further concern is the openness of appraisal related to employee reward which we now discuss.

Appraisal and performance management have been inextricably linked to employee reward since the development of strategic human resource management in the 1980s. The early literature on appraisal linked appraisal with employee control (Randell, 1994;Grint, 1993;Townley, 1993, 1999) and discussed the use of performance related reward to appraisals. However therecent literature has substituted the chapter titles employee “appraisal” with “performance management”(Bach, 2005; Storey, 2007) and moved the focus on performance and performance pay and the limits of employee appraisal. The appraisal and performance pay link has developed into debates to three key issues:

The first issue is has performance pay related to appraisal grown in use?

The second issue is what type of performance do we reward?

and the final issue is who judges management standards?

The first discussion on influences of growth of performance pay schemes is the assumption that increasing linkage between individual effort and financial reward increases performance levels. This linkage between effort and financial reward

increasing levels of performance has proved an increasing trend in the public and private sector (Bevan and Thompson, 1992;Armstrong and Baron, 1998). The drive to increase public sector performance effort and setting of targets may even be inconsistent in the experiences of some organizational settings aimed at achieving long-term targets(Kessler and Purcell, 1992;Marsden, 2007). The development of merit based pay based on performance assessed by a manager is rising in the UK Marsden (2007)reported that the: Use of performance appraisals as a basis for merit pay are used in65 percent of public sector and 69 percent of the private sector employees where appraisal covered all nonmanagerial staff(p.109).

Merit pay has also grown in use as in 1998 20% of workplaces used performance related schemes compared to 32% in the same organizations 2004 (Kersley et al., 2006:191). The achievements of satisfactory ratings or above satisfactory performance averages were used as evidence to reward individual performance ratings in the UK Civil Service (Marsden, 2007).Table 13.2 outlines the extent of merit pay in 2004.

The second issue is what forms of performance is rewarded. The use of past appraisal ratings as evidence of achieving merit-related payments linked to achieving higher performance was the predominant factor developed in the public services. The evidence on Setting performance targets have been as Kessler (2000:280) reported “inconsistent within organizations and problematic for certain professional or le ss skilled occupations where goals have not been easily formulated”. There has been inconclusive evidence from organizations on the impact of performance pay and its effectiveness in improving performance. Evidence from a number of individual performance pay schemes report organizations suspending or reviewing them on the grounds that individual performance reward has produced no effect in performance or even demotivates staff(Kessler, 2000:281).

More in-depth studies setting performance goals followed by appraisal on how well they were resulted in loss of motivation whilst maintaining productivity and achieved managers using imposing increased performance standards (Marsden and Richardson, 1994). As Randell(1994) had highlighted earlier, the potential objectivity

and self-criticism in appraisal reviews become areas that appraisees refuse to acknowledge as weaknesses with appraisers if this leads to a reduction in their merit pay.

Objectivity and self reflection for development becomes a weakness that appraises fail to acknowledge as a developmental issue if it reduces their chances of a reduced evaluation that will reduce their merit reward. The review of civil service merit pay (Makinson, 2000)reported from 4 major UK Civil Service Agencies and the National Health Service concluded that existing forms of performance pay and performance management had failed to motivate many staff.

The conclusions were that employees found individual performance pay divisive and led to reduced willingness to co-operate with management ,citing managerial favorites and manipulation of appraisal scores to lower ratings to save paying rewards to staff (Marsden and French, 1998).

This has clear implications on the relationship between line managers and appraises and the demotivational consequences and reduced commitment provide clear evidence of the danger to linking individual performance appraisal to reward in the public services. Employees focus on the issues that gain key performance focus by focusing on specific objectives related to key performance indicators rather than all personal objectives. A study of banking performance pay by Lewis(1998)highlighted imposed targets which were unattainable with a range of 20 performance targets with narrow short term financial orientatated goals. The narrow focus on key targets and neglect of other performance aspects leads to tasks not being delivered.

This final issue of judging management standards has already highlighted issues of inequity and bias based on gender (Beyer, 1990; Chen and DiTomasio, 1996; Fletcher, 1999). The suggested solutions to resolved Iscrimination have been proposed as enhanced interpersonal skills training are increased equitable use of 360 degree appraisal as a method to evaluate feedback from colleagues as this reduces the use of the “political metaphor”(Randell, 1994;Fletcher, 1999).

On measures linking performance to improvement require a wider approach to enhanced work design and motivation to develop and enhance employee job

satisfaction and the design of linkages between effort and performance are significant in the private sector and feedback and awareness in the public sector (Fletcher and Williams, 1996:176). Where rises be in pay were determined by achieving critical rated appraisal objectives, employees are less self critical and open to any developmental needs in a performance review.

There are key issues that require resolution and a great deal depends on the extent to which you have a good relationship with your line manager . Barlow(1989)argued `if you get off badly with your first two managers ,you may just as well forget it (p. 515).

The evidence on the continued practice of appraisals is that they are still institutionally elaborated systems of management appraisal and development is significant rhetoric in the apparatus of bureaucratic control by managers (Barlow, 1989). In reality the companies create, review, change and even abolish appraisals if they fail to develop and enhance organisational performance(Kessler, 2000).

Despite all the criticism and evidence the critics have failed to suggest an alternative for a process that can provide feedback, develop motivation, identify training and potential and evidence that can justify potential career development and justify reward(Hartle, 1997).

Source: Journal of Accounting Research, Jun2009

译文一：

绩效考核的困境

作者：彼得·布朗斯，茱莉亚·布朗斯

绩效管理发展中的个人业绩与企业组织相联系，以提高组织承诺的成果的实现性，审慎评估以增加工作满意度。它将进一步研究文献的跨领域的效率和有效性议题在北美和英国的证据，以评估个人和组织绩效的人力资源开发和考核贡献。

评估是充分利用企业组织人力资源的一个关键工具。绩效考核的使用分布广泛，根据1998年至2004年一个正式的绩效管理系统的报告(Armstrong and Baron, 1998:200)，估计80-90％的美国组织正在使用，而英国使用比例从69%增加至87％。有很少关于在发展考核的证据，而有更多的使用效益。 1998年至2004年从英国特许人事样本562，发现在英国506人使用表现评估，这是还必须强调的是越来越多地使用的表现。

在2004年WERS调查中，在几乎95％的工作场所强调的关键是基于专业人员和管理人员的绩效的绩效考核反馈。显然，考核已经是不断发展和拓展，已经覆盖很大比例的英国劳动人口和非职业管理以及在私营和公共部门推广使用。

考核评比继续作为绩效波动被更多的部门和行业使用。占主导地位的批判是作为一个正统的技术的管理框架，旨在纠正弱点及评估作为一个发展绩效的系统。

这种“正统”的做法认为有考核冲突目的(Strebler et al, 2001）。考核可以通过明确的目标激励工作人员，制定明确的培训和发展的需要，提供未来的目标来确定绩效目标。这些与过去的绩效考考核和奖励与分配的冲突是基于过去的表现(Bach, 2005:301)。

员工不愿意吐露任何限制和关注他们目前绩效因为这可能对绩效的影响，与相关奖励或晋升机会（Newton and Findley, 1996:43）。这些绩效冲突对于考核人来说是连续作为管理者和判定者的业绩不同角色的挑战，而是理解辅导员的兰德尔1994）认为少数管理人员没有得到执行。McGregor认为绩效考核经理不愿意考核是有典型的证据的，管理者不愿意就个人的表现和培训服务进行判断可能

使士气低落，导致自己的支持和贡献个人表现欠佳，并同时避免人际冲突（McGregor, 1957)。

避免冲突一个后果是，以速度为中心的所有标准并避免为中心的高级管理人员倾向的冲突。由longneckeretal（1987年）对于高级管理者的研究，他们发现组织政策的影响比率60名高级管理人员，调查结果显示，通过加强个人或保护自身利益来刻意的参与政策行动时，当可能的冲突情况，在他们的评估等级比率和是由偏见或受影响的潜在来源决定的（Longeneckeret，1987）。

这有Longenecker’s的证据进一步证明偏见的方法。他们的政治判断和已扭曲不自量力，一些表现能力进一步明确而不是在关键的环节和已知的额定能力，如果一些较低的判决他们可能影响作为角色整个正面影响（ACAS，1996年）。

有些评级可能只包括最近发生的事件，这些都是近期的已知影响。在这种情况下，经理搜集和使用的只有在整个考核期间最近的比较事件。特别关注对绩效数据的公平性的考核，这可能通过性别，种族和考核者的阶层自身而扭曲。一些对于美国和英国的研究凸显了对于性别（Alimo-Metcalf, 1991;White, 1999）考核和考核者的级别的主观性（Geddes andKonrad, 2003）。建议和解决偏见的方案稍后将检讨。

第二个分析是对考核激进的批判。这是更多重要管理文献中认为，评估和表现管理都是关于管理控制(Newton and Findley, 1996;Townley, 1993)。It argues它认为加强对员工行为的管理控制，可实现的对人工考核的拓展，是控制的专业手段。这发展了Foucault利用职权和监视的文献。这些文献使用像教师这样的专业人员的公共服务控制的例子(Healy, 1997)和大学专业人员的(Townley, 1990)。

这方面的证据认为使用考核来加强对公共服务的控制，作为一种控制方法并认为，管理客观性结果忽视了考核和等级评定的发展作用，均获得人们接受和支持的文化和组织价值。然而，这种文献忽略员工阻力与利用工会来挑战试图对专业人员和工作人员在评估过程施加控制。(Bach, 2005:306)

消除偏见的不同议题之一是使用隐喻的测试（Folgeretal，1992年）。这是基于假设考核比率一个评估“真实”绩效的技术性问题，这需要增加考核的可靠性和有效性作为一种发展动力和绩效的工具。评价者偏见和错误的来源可以通过

改进组织公平，增加考核者鉴定可靠性解决的。

但有一个假设的问题，如你可以清晰描述工作要求，该组织是通过考核人从价值和政治议程自由裁量权的价值观和判定反映“理性”的目标。其次存在第二个问题是主观性，如果由考核决定的的绩效比率是由一个“政治隐喻” 决定的(Hartle, 1995)。

这种“政治观点”认为，考核做不好是因为绩效评估者缺少训练，他们把绩效考核看成是一种浪费时间。这成为管理者必须执行，而不是作为一个潜在的改进员工绩效的一个过程。本文中的组织是“政治”的，评估师设法维持下属的绩效并视考核为内部顾客满意。这意味着管理人员使用考核，以避免人际冲突和发展战略的个人地位，并寻求一个安静的生活，避免较高的管理人员的谴责。

这种看法是指管理者也看到考核者寻找好考核比率和反馈以及职业发展，通过寻求员工促进和支付手段相结合的证据.这意味着考核比率成为政治上的判断并设法避免人际冲突。在绩效的“测试”和“政治”的比喻是不准确的考核，缺乏判断雇员绩效客观性，精度偏差。这问题是组织如何能解决这种缺乏客观性？

格林特（1993）认为，在对客观的解决方案的一部分，是麦格雷戈的（1957年）由再培训的经典批判和清除“向下”的管理人员评分封顶和更换多个考核者考核以消除偏见和由绩效评估的客观性基础上。向上有效性的考核意味着主观性的考核比率，这种方法还消除建议考核评分，以消除对妇女的考核(Fletcher, 1999) 多元化报表解决方案（内部同事，客户和收件人服务）将减少评分的主观性和考核比率的不公平。这争论的进一步发展所需要评估的项目团队和越来越多的增加团队合作水平，包括同事评估。在该解决方案的理论上，平均增加经理和加强与个别考核和联系以增加与客户服务所面临的考核。

但是，负反馈仍然是伤害性很大的。通过整理反馈意见，而不是考核绩效表现的经理大量工作。然而，我们总结反馈和解释考核。这与考核的准确性仍然存在客观问题，沃克和史密瑟（1999）用360考评法五年研究了252位经理五年期间，这主观考核问题仍然在评估主体的问题超出缺乏培训等领域。

在考核的贡献是密切相关员工的态度和强烈与工作满意度(Fletcher andWilliams, 1996)。该证据正面考核仍然在振兴工作中的社会关系方面（汤利，1993年）和大型公共服务的广泛采用，英国等作为国家卫生服务体系（NHS）是

各级管理人员的宝贵贡献讨论有关其员工过去的表现，讨论个人发展计划和培训，并积极议题，进一步关注发展是开放性与员工的报酬，我们现在讨论的考核。

20世纪80年代，由于战略人力资源管理的发展，考核和绩效管理已与雇员奖励密不可分。关于考核的早期文献中就将考核与员工控制(Randell, 1994;Grint, 1993;Townley, 1993, 1999)相联系，并讨论了有关考核奖励绩效的使用问题.然而有一部分文献已经取代了章节标题雇员“考核”与“绩效管理”(Bach, 2005; Storey, 2007)，焦点转移到绩效和绩效激励与员工考核的限制上来。绩效与绩效工资的联系已发展为对三个关键问题的讨论：第一问题讨论绩效工资是否已经与考核发展在使用中同步进行？

第二个问题是什么类型的绩效，我们应该奖励？

最后的问题是谁来判断工作标准？

关于绩效工资方案不断增长的影响是第一个讨论的问题，它是在增加个人的努力和物质回报之间的联系以提高绩效水平。这种个人努力和物质回报联动以提高绩效水平，证明了在公共和私营部门的发展趋势(Bevan and Thompson, 1992;Armstrong and Baron, 1998)。这样促使增加公共部门的绩效努力和目标设置可能甚至在实现旨在实现长期目标的某些组织设置中会不一致 (Kessler and Purcell, 1992;Marsden, 2007），通过经理的基于绩效评估的工资的优点的发展是由英国的马斯登（2007）报告中提出：考绩用作绩效工资的基础是在覆盖了非管理层员工的公共部门中的使用比率为65%，私营部门的使用比率为69%（109页）。绩效工资的使用也在增加，在1998年组织使用绩效工资的比率为20％，而在相同的组织2004年这一比率达到32% (Kersley et al., 2006:191)。令人满意的比率或以上的绩效超过令人满意的成果平均数的将作为证据，以奖励英国公务员中个人业绩评级(Marsden, 2007)。

第二个问题是什么样的绩效形式应该获得奖励。对于过去绩效比率的使用是为实现与实现更高的奖赏支付的证据在公共服务中式鉴定评级绩效的主要因素。The evidence on Setting performance targets have been as Kessler (2000:280)关于设置业绩目标已被证明凯斯勒（2000:280）的报告“不一致的内部组织和目标模糊的组织中的某些专业问题或技术水平较低的地方”。这里有从组织对绩效工资的影响pay and its effectiveness in improving performance.和提高表

现的效率的不确定的证据。从大量的个人绩效薪酬计划报告的证据中，机构暂停或审查他们对个别表现却没有获得任何报酬的影响，甚至伤害工作人员（Kessler, 2000:281)。

建立绩效目标更深入的研究,紧随其后的是取决于他们的评估,是如何导致失去动力,同时保持生产率和实现管理者使用性能提高的绩效标准(Marsden and Richardson, 1994)。正如As兰德尔（1994）早前曾强调，潜在的客观性和自我批评在考核区域中，成为评核者拒绝承认评估师的弱点如果从而导致了他们的绩效工资的减少。

客观性和自我反思的发展成为一个弱点，那就是考核失败作为一个发展问题，如果它减少了降低评估的机会，这将减少他们的绩效奖励。审视公务员的绩效工资(Makinson, 2000) 来自于英国4大公务员服务机构和公务员国家卫生服务的结论，绩效工资和存在形式绩效管理没能激励许多工作人员

得出的结论是，雇员发现个人绩效工资分裂并且导致降低管理合作的意愿，管理偏好的引用，对于绩效分数降低比率以减少员工的绩效工资的问题的修复(Marsden and French, 1998)。

这有明确含义关于一线管理人员和考核之间的关系，被动的结果并减少承诺提供明确的证据将个人绩效考核与在公共部门的奖励联系起来。雇员注重这样的问题，以获取关键绩效以集中与关键因素有关的而不是所有的个人性格相联系的具体目标为重点相关的关键绩效指标。 Lewis（1998）的一个银行业绩效工资的研究强调，实施目标是一个在20个绩效目标范围内遥不可及的狭隘的短期业绩目标。这种狭隘的对重点对象关注而忽略其他绩效方面将导致工作无法实现。

最后一个问题是这种判断管理标准已经突出基于性别的不公平和偏见的问题(Beyer, 1990; Chen and DiTomasio, 1996; Fletcher, 1999)。建议的解决歧视的方案为加强人际技巧训练来增加运用360度考核的公平性并作为考核评估的方法从同事的反馈，因为这可以减少在“政治隐喻”使用(Randell, 1994;Fletcher, 1999)。

连接绩效与工作改进的措施连接绩效和连接需要更广泛的方法，加强工作设计和发展动力以及提高员工工作满意度和努力和绩效之间的联系的设计，是重要的私营部门和反馈和在公共部门的宣传(Fletcher and Williams, 1996:176)。

凡在工资上升由实现的关键考核目标决定的，员工缺乏自我批判和开放在任何绩效考核的发展需要中。

需要讨论和解决的关键问题是取决于你与您的顶头上司是否有良好的关系。Barlow(1989)如果你说你跟的两位经理不好，你可能只是把这事给忘它（第515页）。There are key issues that require resolution and a great deal depends on the关于考核继续实践的证据，他们还阐述了管理体制和发展的系统考核在战略决策控制上的应用(Barlow, 1989). In reality the companies create, review, change and even abolish (Barlow, 1989)。在现实中，如果他们没有评估发展和提高组织绩效(Kessler, 2000)，公司将会创建，审查，变更，甚至取消考核。尽管所有的批评和批评者的证据都没有显示一个过程，可以提供反馈，发展动力，确定培训和潜在的理由和证据，可以潜在的职业发展并断定奖励（Hartle, 1997)。

资料来源：大学会计研究所。2009年6月。

原文二：

Performance Evaluations and Efficient Sorting

By: RAY, KOROK.

Much of the production in firms takes place over time. This paper seeks to understand the value of interim performance information on long projects. In particular, the model explores the sorting effects of performance evaluations. Conducting an interim performance evaluation increases efficiency by providing the option to end projects with low early returns. The main result: It is efficient to allocate more resources towards the end of a project. This result holds under a variety of scenarios: when the worker has unknown ability, when the outside options vary with output, and even under an agency framework with a risk-averse agent.

The production of goods and services takes time. It takes time to build cars, write software, develop drugs, formulate strategy, market products, audit clients, and value companies. In fact, many firms today organize their production in the form of projects that run for weeks, months, or even years at a time before the firm can sell the finished product in a market. For example, software development runs through design, implementation, and testing stages before final release. This paper examines the problem of performance evaluation before a project is finished. The main idea is that performance evaluations generate efficient sorting. They provide the firm with the option of ending projects with low early returns. A performance evaluation conducted at the interim stage sorts employees into two groups: stay or quit. It is efficient to quit if the early returns are weak, and to stay otherwise. If the agent’s early output is low, it is in the interest of both the firm and the agent to discontinue work and collect their respective outside options. In this sense, the performance evaluation assigns an agent to his most efficient use. The role of the evaluation, therefore, is to perform this sorting.

Interim performance evaluations change the efficient allocation of re-sources over time in surprising ways. The main result shows that it is efficient to assign more resources to the later stages of the project. The evaluation creates the possibility of

termination after the early stage, which reduces the marginal return from first-stage effort, and so the agent shades his effort downward in the early stage. Once the agent advances, the possibility of termination vanishes and the marginal return to effort rises, so he works harder. In sum, it is efficient to stay if the early returns are high, and for those that stay, it is efficient to work harder.

The usual analysis of performance evaluations takes place in agency mod- The usual analysis of performance evaluations takes place in agency models that span an enormous literature in accounting, economics, and finance (see Baiman [1982, 1990], Indjejikian [1999], Lambert [2001], Prendergast[1999] for reviews). Agency theory has enjoyed wide theoretical attention over the last 20 years, and has dramatically advanced our understanding of performance measurement. Unfortunately, its empirical support remains mixed, as Prendergast [2002] and Lazear [2003] document. The results of most agency models are highly sensitive to their details—the information structure, the timing of the game, the restrictions on the contracts, etc. So when the details of the game change, even slightly, so do the predictions of the model.

This paper takes a different approach. The focus here is not on incentives, but on sorting. I propose a simple and robust efficiency model that does not rely on a complex contracting game within the firm. The aim is to understand the value of interim performance information to all parties, not the opportunistic use of information by one side. The analysis here shifts attention away from distortions from first-best to improvements in first-best.

Performance evaluations don’t just slice up a fi xed pie, but make the pie itself bigger. Exploring the sorting effects of interim performance evaluations in a first-best world abstracts from conflicts of interests between parties and makes more apparent the economic forces driving the effort allocation decision. Nonetheless, I show that the main result of the model is still robust within an agency context. In particular, when a risk-neutral principal cannot observe effort and contracts with a risk-averse agent, the agent still shifts his effort into the later stages. The difference in risk preferences between the two parties causes the agent to distort his effort choice within each period, but does not qualitatively change his effort decision between periods.

The sorting effect of the interim evaluation o nly impacts the agent’s dynamic decision problem, and does not interact with the difference in preferences between the principal and agent. Therefore, the result still holds within an agency context, even though it is not necessary to operate in a second-best world to generate the sorting effect. The model consists of an agent working for a firm on a project over two stages. His effort levels across stages are perfect substitutes, so the production technology by itself does not skew his effort allocation. Production is indivisible: The firm cannot capture the full value of output until the end of the project. This is the sense in which production takes time. In particular, incomplete projects fetch a low market price, normalized to zero. The firm has invested so much firm-specific capital and labor into the project that it becomes too costly for other firms to understand and finish it. Finally, firms and workers have outside options, which include future profits from new projects for the firm and future wages from new jobs for the employee.

The existing body of theoretical literature on intertemporal allocation The existing body of theoretical literature on intertemporal allocation of effort in multistage projects falls into three categories: first-best analyses, models based on general agency considerations under discounting and/or risk aversion, and career-concerns models. The first strand of this literature, dating back to the early 1970s, is similar to my paper in that it examines efficient intertemporal effort allocation in research and development (R&D) projects, in the absence of any incentive problems. Examples include Kamien and Schwartz [1971], Luss [1975], and Grossman and Shapiro [1986]. Kamien and Schwartz [1971] study the time pattern of optimal R&D expenditures when the total cumulative effort E required tocomplete the project is not known. They find that per-period investment (effort) is increasing in time if the hazard rate associated with the cumulative distributive function of E is nondecreasing. Luss [1975], on the other hand, examines a setup in which a project is completed in known finite time, but its returns depend on effort exerted each period, and the effectiveness of effort decreases over time. He finds that it is optimal to exert the highest effort in the first period, medium-level constant effort in the following intermediate periods, and the lowest effort in the last period. Grossman and Shapiro

[1986] argue that when the total amount of effort required to complete a project is known, optimal effort increases over time. This result occurs due to discounting: The marginal product of effort falls at the discount rate. However, when time to completion is not known, effort is higher when the expected value of the project is higher, which does not necessarily imply a time-monotonic path of effort. This result is shown to be in line with Kamien and Schwartz’s [1971] result: If the hazard rate is increasing over time, so is the expected project value. Finally, it is interesting to note that in the discrete two-period case, their model predicts that second stage effort will be lower than first stage effort, provided that the second stage occurs. This is due to a disappointment effect: The occurrence of the second stage is unambiguously bad news; it shows that the project will take two periods rather than one.

The second strand of literature addresses optimal effort choice in anagency setting. A seminal paper in this line of work is Lambert [1984], who shows that when the project has time separable, mutually independent returns each subperiod, a risk-averse manager will choose effort to “smooth” income to keep it close to its ex ante expectation. In particular, second-stage effort is decreasing in observed first-stage output (but it can be higher or lower than first-stage effort). The author also shows that this result is an artifact of the incentive problem: Under the first-best, each period’s effort is independent of the previous periods’ output. Given that so much of today’s production takes plac e in long-term projects, it is natural to explore the role and consequences of performance evaluations on projects before they finish. By their very definition, projects take time, and firms cannot collect full value from them until they are completed. So interim performance evaluations give firms the option to end projects (or fire employees on existing projects) if the early returns are low. Performance evaluations efficiently sort projects and employees into those that quit and those that stay. A surprising consequence of this sorting is its effect on dynamic resource allocation: It is efficient to allocate more effort towards the later stages.

The logic of the main result runs as follows. Because production is indivisible, firms and agents cannot sell the project after the first stage, and instead use

early-stage output to calculate future project value and ultimately the value of continuing. If the continuation value is sufficiently low, both parties prefer to capture their outside options. This is the role of interim performance evaluations. The possibility of termination after the first stage reduces the expected project return in stage 1, and hence lowers the marginal return to effort. Thus the agent invests more of his resources (effort) later in the project. This is the main consequence of interim performance evaluations. The paper shows that the main result is robust under a variety of permutations. First, the result generalizes beyond the simple two-stage model to the finite stage game, proving that restriction to two stages is without loss of generality. Second, even if the evaluation returns detailed performance information, so that the agent can condition his second-stage effort on first-stage output, the agent efficiently works harder on average in stage 2 than in stage 1. Third, if the outside options vary with output such that the marginal increase in outside options does not exceed the marginal cost of effort, the agent exerts more effort on average in stage 2 than in stage 1. And most importantly, this result holds within an agency framework. With a risk-neutral principal and a risk-averse agent, the agent still shades his effort downward in the early stage. The differential risk preferences do not change the allocation of effort between stages even though they do change the allocation of effort within each stage.

The purpose here is to show that production that takes place over time gives firms the option of ending projects before they finish. Exercising this option and collecting intermediate performance information skews the efficient allocation of resources in surprising ways. The prior agency literature on career concerns gives a mixed verdict on optimal effort allocation; some papers predict increasing effort over time while others predict decreasing effort over time. The aim of my analysis is to argue that sorting delivers a strong prediction that resource allocation increases over time. I present the majority of the analysis in a first-best world to show the main forces at work and highlight the key economic intuition.

Source: Journal of Accounting Research, Sep2007, V ol. 45 Issue 4