Performance Management

Chapter 13
Ethics, Law and Governance

Bachelors Program in Human Resource Management
Performance Management   •   Bachelor

Ethics, Law and Governance

Introduction and context

Performance management is often presented as a cycle of goal setting, coaching, appraisal, development and recognition. In practice, it is also a system for distributing opportunities, pay, promotion, corrective action and, sometimes, continued employment. Because these decisions affect people’s livelihoods and dignity, performance management is not merely an administrative or measurement activity. It is an ethical, legal and governance challenge.

A technically sophisticated appraisal system can still be unfair if its standards are unclear, its ratings are inconsistent, or employees cannot challenge decisions. Conversely, a legally cautious system may fail ethically if it treats people as data points, ignores disability or cultural differences, or discourages employees from speaking about concerns. Effective governance connects organizational purpose with accountable decision-making. It establishes who may decide, what evidence may be used, how decisions are documented, how conflicts are managed and how errors are corrected.

Employment law differs across jurisdictions. This chapter therefore presents general principles rather than legal advice. HR professionals must consult applicable legislation, collective agreements, regulatory guidance, contracts and organizational policies in the relevant jurisdiction.

Learning objectives

By the end of this chapter, students should be able to:

  1. Explain the ethical foundations of performance management, including procedural and distributive justice.
  2. Identify legal and governance risks in goal setting, appraisal, feedback, reward and discipline.
  3. Design transparent, inclusive and accessible performance processes.
  4. Apply principles of employee voice, documentation, data protection and appeals.
  5. Evaluate the use of technology and artificial intelligence in performance decisions.
  6. Recommend controls that improve accountability and compliance.

1. The ethical foundations of performance management

1.1 Performance management as a relationship of power

Managers normally control, or strongly influence, ratings, development opportunities, bonuses and recommendations for promotion. This creates an asymmetry of power. Employees may therefore agree with an assessment publicly while disagreeing privately, especially where disagreement may affect their future. Ethical performance management recognizes this asymmetry and creates safeguards against arbitrary or retaliatory decision-making.

Four ethical principles are particularly important:

  • Respect for persons: Employees should be treated as adults whose dignity, privacy and voice matter.
  • Non-maleficence: Processes should avoid preventable harm, humiliation, discrimination and unnecessary disclosure of personal information.
  • Beneficence: Feedback and development should help employees improve and succeed, rather than merely justify a predetermined outcome.
  • Justice: Similar cases should be treated consistently, while relevant differences—such as disability, caring responsibilities or unequal access to resources—should be considered fairly.

Ethical judgment is not identical to treating everyone in exactly the same way. Identical treatment can reproduce disadvantage when employees have unequal working conditions or require reasonable accommodation.

1.2 Procedural and distributive justice

Procedural justice concerns whether the process used to reach a decision is fair. Typical indicators include:

  • clear standards communicated in advance;
  • opportunity to provide evidence and respond;
  • consistency across comparable cases;
  • decision-makers who are impartial and appropriately trained;
  • reasons that explain the decision;
  • a genuine route of review or appeal; and
  • correction of demonstrable errors.

Distributive justice concerns whether outcomes are fairly allocated. Relevant outcomes include ratings, pay, bonuses, promotion, development access, workload and disciplinary consequences. A fair procedure does not guarantee that every employee receives the same outcome. It requires that differences be based on relevant evidence rather than favoritism, stereotypes or irrelevant personal characteristics.

Interactional justice concerns the quality of interpersonal treatment. A manager may follow a formal process but still act unfairly by humiliating an employee, withholding information or communicating aggressively. Respectful explanations, active listening and confidentiality are therefore substantive elements of fairness.

A useful diagnostic question is: Could the organization explain this decision clearly to the employee, an auditor and an independent reviewer without relying on confidential gossip or undocumented impressions? If not, the process is vulnerable.

2. Legal architecture and compliance risks

The precise legal rules vary by country, but performance systems commonly intersect with the following areas.

2.1 Equality and anti-discrimination law

Employment law commonly prohibits discrimination in recruitment, terms and conditions, promotion, discipline and termination. Protected characteristics may include race, ethnicity, sex, gender identity, pregnancy, age, disability, religion, national origin and other categories defined by law. A performance process can create unlawful discrimination even when no manager intends to discriminate. Examples include:

  • a “culture fit” criterion that rewards conformity to a dominant group;
  • penalizing an employee for disability-related absence without considering accommodation;
  • rating communication style differently according to gender or cultural expectations;
  • using customer preference as a justification for discriminatory treatment; and
  • applying an attendance standard that disproportionately excludes a protected group without adequate justification.

HR should distinguish direct discrimination, where a person receives less favorable treatment because of a protected characteristic, from indirect or disparate-impact risk, where a seemingly neutral rule disadvantages a protected group and cannot be objectively justified or reasonably adjusted.

2.2 Disability, health and reasonable accommodation

Performance standards should describe essential outcomes and legitimate job requirements, not assume one method of working. Reasonable accommodation may include assistive technology, modified schedules, accessible formats, additional breaks, altered communication methods or adjustment of non-essential duties. Accommodation is not an unfair advantage; it is a mechanism for enabling equitable participation.

Managers should not require unnecessary medical details. HR should establish a confidential process through which employees can request support, and performance documentation should distinguish capability, conduct, attendance and health information rather than blending them into a vague negative judgment.

2.3 Wage, working-time and contractual obligations

Performance-related pay must comply with wage, equal-pay, working-time and contractual requirements. Employers should clarify whether an incentive is discretionary or contractual, how eligibility is determined and how leave, part-time work, protected absence or organizational restructuring affect calculations. A manager should not use a discretionary bonus as a hidden penalty for protected activity or lawful absence.

2.4 Employment status, collective representation and consultation

Contractors, temporary workers, interns and employees may have different legal rights, but misclassification can create significant liability. Where a collective agreement or works council applies, performance criteria, monitoring and disciplinary procedures may require consultation or adherence to negotiated rules. Employees may also have legally protected rights to representation during formal disciplinary or grievance processes.

2.5 Privacy and data protection

Performance records are personal data in many jurisdictions. Data protection principles commonly require that information be processed lawfully, fairly and transparently; collected for specified purposes; limited to what is necessary; kept accurate; retained only as long as justified; secured against unauthorized access; and made available to individuals through applicable access rights.

A sound performance-data policy should answer:

  1. What data are collected?
  2. Why are they collected?
  3. What is the lawful or organizational basis for processing?
  4. Who can access them?
  5. How long are they retained?
  6. How can errors be corrected?
  7. Are automated tools involved?
  8. How can an employee challenge a decision?

Information collected for one purpose should not silently be repurposed for another. For example, building-access data collected for security should not automatically become a productivity score without necessity, transparency and legal review.

3. Governance: turning values into controls

Governance is the system by which authority, responsibility, oversight and accountability are arranged. It is more than a policy document. It includes decision rights, controls, records, review mechanisms and organizational behavior.

3.1 A governance framework

A practical performance-governance framework has six connected elements:

  1. Purpose and principles: State what performance management is intended to achieve and the principles of fairness, inclusion, privacy and development.
  2. Roles and authority: Define the responsibilities of employees, line managers, HR, senior leaders, data protection officers, employee representatives and appeal panels.
  3. Standards and evidence: Specify how goals are set, what evidence is relevant and how contextual factors are considered.
  4. Process controls: Require review points, calibration, conflict-of-interest checks and approval for high-impact decisions.
  5. Records and transparency: Maintain accurate records, communicate decisions and protect confidentiality.
  6. Assurance and improvement: Audit outcomes, investigate complaints, monitor patterns and revise the system.

A policy without ownership is ineffective. For example, if everyone is responsible for fairness, no one may be accountable when a biased pattern appears. Governance should identify a named process owner and an escalation route.

3.2 Separation of roles and conflicts of interest

The manager who observes performance may be appropriate to provide an initial assessment, but should not always be the sole and final decision-maker. Independent review is particularly important where there is a conflict of interest, a close personal relationship, a complaint against the manager or a potential termination. Separation does not require bureaucracy in every routine conversation; it requires proportionate controls for higher-risk decisions.

3.3 Calibration and consistency

Calibration meetings can reduce rating differences between managers, but they also create risks. Group discussion may amplify stereotypes, expose confidential information or force artificial ranking. Good calibration uses defined criteria, evidence and anonymized or carefully controlled comparisons. It should not turn a developmental assessment into a competition for a fixed number of high ratings unless such a distribution is demonstrably justified.

4. Designing a fair performance process

4.1 Goal setting

Goals should be relevant to the role, sufficiently specific to guide action and realistic given available resources. They should identify the expected result, time period, quality requirements and dependencies. Employees should have an opportunity to discuss goals rather than merely receive them.

A goal is not fair when the employee lacks the authority, information or resources necessary to achieve it. Goals should be revisited when priorities, staffing, technology or market conditions materially change. Otherwise, a process may punish employees for organizational decisions beyond their control.

4.2 Evidence and documentation

Evidence should be job-related, timely, sufficiently reliable and proportionate. Useful evidence may include work outputs, service-quality indicators, project records, documented behaviors and feedback from relevant stakeholders. A single unverified complaint or a manager’s general impression should rarely determine a high-impact decision.

Documentation should record:

  • the agreed expectation;
  • relevant evidence and its source;
  • dates and context;
  • feedback provided;
  • the employee’s response;
  • support or resources offered;
  • agreed next steps; and
  • review dates.

Documentation is not a weapon for building a case retrospectively. It should be created throughout the cycle and shared appropriately. Over-documentation can also be harmful if it captures irrelevant personal information or creates surveillance rather than useful evidence.

4.3 Feedback and difficult conversations

Effective feedback describes behavior or results, explains impact and identifies a path forward. It avoids labels such as “lazy,” “not committed” or “bad attitude,” which are ambiguous and vulnerable to bias. Managers should ask for the employee’s perspective, listen for barriers and distinguish lack of capability from lack of clarity, resources or opportunity.

Where performance is below standard, a fair improvement plan should specify the gap, expected standard, support, timescale, review dates and consequences of insufficient improvement. Expectations must be achievable and consistently applied. Formal action should not be a surprise created by an annual appraisal.

4.4 Inclusion and accessibility

Inclusive design anticipates human variation. Performance systems should provide accessible documents, interpreters or communication alternatives where needed, and reasonable flexibility in meeting arrangements. Remote and hybrid employees should not be rated lower merely because they are less visible. Managers should evaluate outcomes and relevant behaviors rather than proximity, responsiveness at all hours or participation in informal networks.

Organizations should analyze whether certain groups receive fewer stretch assignments, lower-quality feedback or less access to development. These are process issues, not simply individual employee outcomes.

5. Employee voice, appeals and remedies

Employee voice includes routine participation, consultation, feedback about the process, representation and formal challenge. Voice improves information quality because employees often know about resource constraints, inconsistent standards or bias that managers cannot see.

A credible appeal process should:

  1. explain what decisions may be appealed;
  2. set a reasonable time limit without making access impossible;
  3. identify acceptable grounds, such as factual error, procedural irregularity, inconsistency, discrimination or failure to consider relevant evidence;
  4. assign an impartial reviewer;
  5. allow the employee to submit evidence and receive appropriate representation;
  6. provide a reasoned outcome; and
  7. protect the employee from retaliation.

An appeal is not necessarily a complete rehearing. It may review whether the original process was reasonable and evidence-based. However, where new evidence or a serious conflict of interest exists, a fresh assessment may be necessary.

Organizations should monitor appeal outcomes. A high rate of successful appeals may indicate weak first-stage decisions; a zero-success rate may indicate either exceptional consistency or a process that discourages legitimate challenges.

6. Technology, analytics and artificial intelligence

Digital systems can improve record keeping, remind managers about review dates and identify disparities. They can also intensify surveillance and reproduce historical discrimination.

Risks include:

  • inaccurate or incomplete data;
  • proxy discrimination, where variables correlate with protected characteristics;
  • opaque scoring that employees cannot understand;
  • automation bias, where managers accept a system’s recommendation uncritically;
  • excessive monitoring of keystrokes, location or facial expressions;
  • security breaches; and
  • function creep, in which data are reused beyond the original purpose.

Before deploying an algorithmic performance tool, an organization should conduct a documented impact assessment. It should test validity for the actual job, examine outcomes across relevant groups, provide human oversight, preserve an explanation and appeal route, restrict access, and suspend the tool when material harms are detected. A vendor’s claim that a system is “objective” is not evidence of fairness.

AI may assist with summarizing documented feedback or identifying missing review steps, but high-impact decisions should not be delegated without accountable human judgment. Managers remain responsible for verifying outputs and considering context. Employees should be informed when automated tools materially influence evaluation.

7. A practical decision framework

When designing or reviewing a performance decision, HR professionals can apply the following sequence:

Define: What decision is being made, and what legitimate organizational purpose does it serve?

Test relevance: Is each criterion genuinely related to the role and outcome?

Check evidence: Is the evidence accurate, sufficient, current and open to response?

Check equality: Could the criterion disadvantage a group? Have accommodation and contextual factors been considered?

Check process: Was the standard communicated, applied consistently and decided by an authorized person?

Check privacy: Is the data collection necessary, secure and retained appropriately?

Check voice: Can the employee explain, challenge and appeal the decision?

Record and review: Are reasons documented, and will outcomes be audited for patterns?

This framework does not eliminate judgment. It makes judgment visible, disciplined and reviewable.

8. Illustrative case: the “visibility” problem in a hybrid organization

Background

Northbridge Services, a fictional customer-support organization, introduced hybrid work. At year-end, several managers rated office-based employees more highly than remote employees, using phrases such as “more committed” and “easier to rely on.” Remote employees reported that they handled comparable workloads and received fewer informal opportunities to demonstrate their contribution.

Problem

The organization faced ethical and governance concerns: ambiguous criteria, proximity bias, inconsistent evidence and possible disadvantage for employees with disabilities or caring responsibilities. The ratings affected bonuses and promotion nominations.

Analysis

HR reviewed goal records, workload data, feedback timing and ratings by work location. The review did not assume that a difference proved discrimination; it examined whether the difference could be explained by job-related factors. It found that managers had received no guidance on evaluating hybrid work and that informal office observations were frequently recorded as evidence.

Approach and solution

Northbridge paused final promotion decisions, allowed employees to submit evidence and required managers to reassess ratings against agreed outcomes. It replaced “visibility” with defined criteria such as service quality, completed commitments, collaboration and documented customer outcomes. Managers received training on bias and accessible hybrid supervision. A calibration panel reviewed high-impact decisions, and the policy prohibited monitoring keystrokes as a proxy for performance. Employees received an appeal route and a summary of the evidence used.

Outcome

Some ratings changed, several development opportunities were redistributed and the organization identified a need to improve goal setting at the beginning of each cycle. The process did not guarantee identical ratings by location; it made differences explainable through job-related evidence.

Learning points

  1. Vague criteria invite bias.
  2. Equal treatment may require different methods of access and evidence.
  3. Data analysis should diagnose patterns, not replace judgment.
  4. Procedural safeguards are especially important when outcomes affect pay or progression.
  5. Governance must address ordinary managerial habits, not only deliberate misconduct.

9. Challenges and limitations

No performance system can measure contribution perfectly. Some work is collaborative, long-term or difficult to quantify. Excessive measurement can encourage gaming, reduce intrinsic motivation and disadvantage employees who perform essential but less visible work. Rules can also become rigid and prevent managers from responding to genuine context.

Fairness itself may involve competing claims. A uniform deadline may support consistency, while an adjusted deadline may be necessary for accommodation. A manager may need to protect confidentiality while giving enough explanation for an employee to challenge a decision. Governance therefore requires proportionality: controls should be stronger where the decision is more consequential, data more sensitive or risk of bias greater.

Legal compliance is a minimum standard, not proof of ethical quality. An organization may comply with formal requirements while creating a culture in which employees fear speaking up. Conversely, good intentions do not excuse failure to follow legal duties. Ethical judgment and legal compliance should reinforce each other.

10. Best-practice checklist

An accountable performance system should:

  • use clear, role-related and communicated standards;
  • review goals when circumstances materially change;
  • train managers in evidence, bias, feedback, accessibility and documentation;
  • distinguish performance, conduct, attendance, health and capability issues;
  • provide reasonable accommodation and accessible processes;
  • limit data collection to legitimate needs;
  • protect confidentiality and define retention periods;
  • use independent review for high-impact or conflicted decisions;
  • provide employee voice, representation and appeal;
  • audit outcomes for group disparities and unintended effects;
  • document reasons rather than conclusions alone; and
  • revise the system in response to evidence and employee experience.

Chapter Summary

Performance management distributes important opportunities and consequences, making ethics, law and governance central rather than peripheral concerns. Procedural justice requires clear standards, relevant evidence, participation, consistency, impartiality, explanation and review. Distributive justice focuses on whether outcomes are allocated on legitimate grounds, while interactional justice emphasizes dignity and respectful treatment.

Legal risks commonly arise from discrimination, failure to accommodate disability, unequal pay, contractual breaches, privacy violations, inadequate consultation and improper use of monitoring or automated tools. Governance converts principles into accountability through defined roles, decision rights, documentation, calibration, oversight, appeals and audits.

A fair system evaluates job-related contribution while considering context and access to resources. It protects employee voice, uses data proportionately, and treats AI as an aid requiring validation and human accountability. The central professional obligation is to create decisions that are explainable, evidence-based, inclusive, legally compliant and open to correction.

Key Terms

Procedural justice: Fairness of the process used to reach a decision.

Distributive justice: Fairness of the allocation of outcomes and resources.

Interactional justice: Fairness and respect experienced during interpersonal treatment.

Reasonable accommodation: A suitable adjustment enabling a person to participate in work without imposing disproportionate hardship.

Direct discrimination: Less favorable treatment because of a protected characteristic.

Indirect discrimination: A neutral rule that disproportionately disadvantages a protected group without adequate justification.

Performance evidence: Relevant information used to assess contribution or behavior.

Calibration: A structured review intended to improve consistency between assessments.

Employee voice: Opportunities for employees to express views, provide evidence and influence or challenge decisions.

Appeal: A formal request to review a decision or process.

Governance: Structures and controls through which authority and accountability are exercised.

Conflict of interest: A personal or organizational interest that may compromise impartial judgment.

Data minimization: Collecting and using no more personal data than necessary.

Purpose limitation: Using data only for specified and legitimate purposes.

Automation bias: The tendency to trust automated recommendations excessively.

Algorithmic impact assessment: A structured evaluation of an automated system’s risks and effects.

Retaliation: Adverse treatment because a person raised a concern or exercised a lawful right.

Audit trail: A record showing how a decision was made and by whom.

Accessibility: Designing information and processes so people with differing abilities can use them effectively.

Proportionality: Matching the intensity of controls or consequences to the level of risk and impact.

Review Questions

Short Answer Questions

  1. Distinguish procedural justice from distributive justice.
  2. Why should performance goals be revisited when organizational circumstances change?
  3. State three principles that commonly apply to the processing of performance data.
  4. What is the purpose of an appeal process?
  5. Give two risks associated with algorithmic performance assessment.

Long Answer Questions

  1. Explain how ethical principles and employment law jointly shape performance-management design.
  2. Discuss how documentation can improve accountability while also creating privacy risks.
  3. Evaluate the role of calibration meetings in promoting consistency and their possible disadvantages.
  4. Explain how organizations can design an inclusive and accessible performance process.
  5. Discuss why legal compliance is necessary but insufficient for ethical performance management.

Analytical / Application Questions

  1. A manager rates a remote employee lower because the employee is “not visible enough.” Apply the decision framework in this chapter to assess and redesign the decision.
  2. An organization wants to use keystroke monitoring to identify low performers. Advise HR on relevance, privacy, validity, employee voice and governance safeguards.
  3. A disabled employee misses a target because an agreed workplace adjustment was delayed. Analyze how the organization should review the rating, document the circumstances and prevent recurrence.

Multiple Choice Questions with Answers

Question 1

Which form of justice concerns whether the process used to reach a performance decision is fair?

A. Distributive justice

B. Procedural justice

C. Market justice

D. Corrective justice

Correct Answer: B. Procedural justice

Question 2

Which is the strongest example of relevant performance evidence?

A. A manager’s belief that an employee lacks enthusiasm

B. An employee’s personality type

C. Documented completion of agreed project outcomes and quality standards

D. The number of hours an employee appears online

Correct Answer: C. Documented completion of agreed project outcomes and quality standards

Question 3

What is the main purpose of reasonable accommodation?

A. To remove all performance standards for a disabled employee

B. To enable equitable participation by adjusting non-essential barriers

C. To guarantee promotion regardless of performance

D. To collect more medical information

Correct Answer: B. To enable equitable participation by adjusting non-essential barriers

Question 4

Which practice best reflects data minimization?

A. Collecting all available employee data in case it becomes useful

B. Keeping performance records indefinitely

C. Collecting only data necessary for a defined performance purpose

D. Sharing appraisal data with all employees

Correct Answer: C. Collecting only data necessary for a defined performance purpose

Question 5

What is a key danger of vague criteria such as “culture fit”?

A. They always improve reliability

B. They can permit stereotypes and inconsistent judgments

C. They eliminate manager discretion

D. They prevent any employee feedback

Correct Answer: B. They can permit stereotypes and inconsistent judgments

Question 6

Which feature most strengthens an appeal process?

A. Review by the same person who made the original decision

B. No time limit or stated grounds for appeal

C. Impartial review with an opportunity to submit evidence

D. Automatic rejection of new information

Correct Answer: C. Impartial review with an opportunity to submit evidence

Question 7

What does automation bias mean in performance management?

A. Employees refuse to use digital systems

B. Managers place excessive trust in automated recommendations

C. Algorithms are always discriminatory

D. Organizations stop collecting performance data

Correct Answer: B. Managers place excessive trust in automated recommendations

Question 8

Which action best supports procedural justice in a high-impact rating decision?

A. Keeping the criteria confidential

B. Relying on one unverified complaint

C. Communicating standards, allowing response and providing reasons

D. Using a forced distribution without explanation

Correct Answer: C. Communicating standards, allowing response and providing reasons

Question 9

Why should performance outcomes be audited by relevant employee groups?

A. To prove that every group receives identical ratings

B. To identify possible patterns of disadvantage or inconsistent application

C. To replace individual assessment entirely

D. To remove the need for manager training

Correct Answer: B. To identify possible patterns of disadvantage or inconsistent application

Question 10

In the illustrative hybrid-work case, what was the most appropriate replacement for “visibility” as a criterion?

A. Time spent in the office

B. Frequency of informal conversations with managers

C. Defined outcomes, service quality and documented collaboration

D. Number of messages sent outside working hours

Correct Answer: C. Defined outcomes, service quality and documented collaboration

Performance Management
Chapter 13 of 14