Title: What Determines Performance
1What Determines Performance?
- Previous training (Human Capital)
- Innate ability (psychologists G)
- Effort
2Evaluation
- Evaluations serve several purposes.
- Measuring performance to use in implementing
incentive compensation (feedback and raises) - Measuring the workers general ability to forecast
her performance on another job. (Screening and
job allocation) - Evaluation is costly. This means that the firm
must choose the optimal degree of evaluation.
3How much and when?
- Evaluations should occur less often
- If workers have highly specific training. If is
unlikely that the firm will need to meet outside
offers quickly. - If the firm can screen workers easily before the
job is taken. - If the worker has long tenure with the firm.
General ability is known. Senior workers should
be evaluated on achievement not ability.
4How do we measure performance?
- Ideal Performance measure Value of person to the
organization and NOTHING ELSE - controllables effects of efforts on firm
value qi - improves alignment between employee firm
- uncontrollables effects of other factors
- measurement error, effects of other employees on
firm value, or effects of external events on firm
value - exposes employee to risk
- example stock price as a performance measure for
a CEO or a janitor
5How do we measure performance?
- Effect on organization bottom line (broad)
- Performance on tasks (narrow)
- Quantity of work
- Quality of work
- Objective evaluation (Quantitative) or Subjective
evaluation (Qualitative)
6Scope of a Performance Measure
Narrow Broad (fewer factors) (more
factors) High Distortion Low Distortion Low
Risk High Risk
- Narrower measures are those that tend to focus
on fewer aspects of performance, or on inputs - Narrower measures tend to reduce risk, but also
to distort incentives more - the greater the distortions, the weaker should
the incentives be - measures may be broader/ narrower or narrower on
several dimensions - broader job designs tend to imply broader
performance measures
7Scope Distortions
- Measures trade off risk v. distortions are
narrower or broader along several dimensions - 1. how many tasks to include?
- e.g., quantity v. quality
- 2. should intangibles or difficult-to-measure
effects be included? - e.g., opportunity costs
- 3. how large a unit should be measured?
- e.g., individual v. team v. unit v. division v.
corporate - 4. what time horizon?
- e.g., sales v. customer retention / growth
- Accounting s often have some of all 4 problems
8Broad vs Narrow Measures
- Narrow Measures
- Easier to measure
- Filter out some uncontrollable effects on
performance (example accounting earnings vs
stock price) therefore reduce employee risk - Distort incentives by focusing only on the narrow
view (short term for executives specific task at
expense of others for lower level)
9Broad vs Narrow Measures
- Broad Measures
- Allow tailored responses to specific issues
- Give correct incentives
- Line of sight issue is very important
- Often highly risky for employee due to
uncontrollables
10Likely Problems with Performance Measures
- Hard to measure qualities (intangibles)
- Opportunity cost of doing things
differently-value potentially unknown - Differing group sizes and qualities
- Individual performance often neglects cooperation
- Team performance can run into 1/n problem
- Small groups may differ in average quality
- Time period for performance often too
short-encourages short run actions that are bad
in the long run - Potential for manipulation (gaming)
- Distorts incentives (wrong weights)
11Possible Scope of Performance Measure for
Division Manager Constraints get less Decisions
are more
12Examples
- Balanced Scorecard
- emphasizes financial plus other measures to
reduce distortions - customer perspective quality, service, delivery
- internal perspective employee satisfaction,
unit cost, yield - innovation learning time to market, new
products, sales from new products - EVA (Economic Value Added)
- emphasizes inclusion of opportunity costs to
reduce distortions - residual income after cost of capital
- designed to support decentralization
- careful performance meas., reallocation of
decision rights, stronger incentives - goal ownership incentives for divisional
managers
13Summary
- Properties of performance measures to think about
- how aligned with firm value (how distorted)?
- can distortions be reduced by adding an incentive
on a second measure? - how risky for the employee?
- should the firm incur greater costs to measure
more accurately? - how manipulable?
- could this degrade the value of the measure?
14Let them do their work
- Constraints should be as few as possible with
good management performance measures should be
as broad as possible
15Benefits of Subjective Evaluations
- Refine quantitative performance
- Improve risk taking by introducing flexibility
for uncontrollable events - Change incentive system on-the-fly
- Expand communication
- Improve training
16Subjective Evaluation
- Judgment is an alternative to quantitative
- Typical problems
- inflation compression
- reluctance for negative feedback
- de-motivating effects?
- low trust of evaluator
- favoritism
- problems are more significant when promotion
stakes are higher - ½-life of appraisal systems ? 5 yrs
17Example TopGrading
- Pioneered at GE, tried by many other firms
- some form of forced curve
- GE top 20 middle 70 bottom 10
- those w/ consistently low ratings typically
forced out - Where does it work best?
- larger workgroups
- teamwork is less important
- culture is more aggressive / Darwinian
- extensive attention is paid to coaching
development - employees are given clear expectations about
weeding out - potential legal liabilities are monitored
carefully - great care is taken to make sure evaluations are
based on performance
18Who should do the evaluation?
- To minimize politicking, the person doing the
evaluation should not be accessible by the
worker. - This introduces a tradeoff since outside persons
have less information about the worker. - One compromise is to review from both above and
below (360 degree evaluation)
19Subjective Evaluations
- Supervisors do not like to do evaluations
- the supervisor absorbs all the grief from a poor
evaluation - everyone will be rated above average
- Negative feedback is not easy to give in a
positive way - Creates agency problem with supervisors
evaluations tend to have small variance around
above average - Incentives must be given for correct evaluations
- Evaluations need to move from past focus to
future focus - Forced distribution is an extreme form of
incentive
20Relative or Absolute Ratings?
- Absolute - Encourages cooperation between
workers. Encourages politicking. Introduces
tension between workers and management because
workers want low standard and management wants
high. - Relative - Cooperation between workers hurt.
Workers may collude to lower effort. Increases
tension among workers. Reduces opportunities for
supervisor bias and therefore leads to less
politicking.
21Dirty Little Secret in Business
- Lack of candor Very few workers get real
evaluations that allow them to see what they need
to improve. - Candor yields speed
- Candor reduces cost
- Why isnt candor used?
- Socialized to make nice
- More popular
- Easier not to talk about it
- Real reason self-interest rather than interest
in other person
22Candor/Differentiation
- Very difficult to instill Took GE nearly 20
years - GE 20/70/10
- Criticisms
- Unfair
- Undermines team
- Mean/immoral
- Only possible in the US
- Biggest Problem Top of 70 not much different
from Top 20. Less problem of bottom 10 not much
different from bottom of 70
23Ways to Mitigate Problems w/ Appraisals
- Reduce the stakes
- pay compression, weaker incentive intensities
- Evaluator Incentives
- pay for performance
- constraints monitoring
- multiple evaluators review of evaluations
grievance procedures - Culture of constructive feedback
- Avoid conflicts between evaluation coaching
goals - Other approaches 360, MBO
24The Appraisal Process
- Strive for day-to-day coaching, rather than a
comprehensive annual appraisal - reduces defensiveness to criticism
- more closely tied to events
- Focus on goal setting rather than criticism
- constructive
- mutual goal setting
- but fight the tendency to avoid tough feedback
- Consider separate appraisals for separate
purposes - evaluation for salary promotion
- coaching development
25Properties of A Successful System
- Successful Measurement System Provides
- Feedback that tells you whether you are - or
aren't - doing the right things, and doing them
well - Motivation that will spur you on to improve your
performance - Incentive for you and your team to 'do it better'
- But for all this to happen you must make sure
that you measure - the right things
- at the right time, and
- in the right place
26Receiving an Appraisal
- Ask for feedback, regularly
- Avoid defensiveness / check emotions
- take responsibility for mistakes
- dont politicize / lobby
- Focus on constructive suggestions
- set goals
- discuss priorities
- Identify what you need
- new skills
- information
- resources
27How much and when?
- Evaluations should occur less often
- There is little downside risk (or vice versa when
there is large upside gain) - When there is a high probability that the worker
will succeed at the job. - Workers are less likely to have outside offers
- At the end of careers