Week 5 - Incentive systems

Question 1

3 purposes/effects of incentives
1. SIGNALLING

Answer 1

1. to share- or stakeholders the importance of performance dimensions/targets
   eg. BP linked bonuses to safety performance after the oil spill to signal that they are taking safety measures more seriously
2. to signal their actions to COMPETITORS
   eg. revenue-based pay shows that CEOs are focusing on growing their sales and firm
   » then other companies won’t dare to compete and take their market share, knowing that the large competitor has an aggressive strategy and will retaliate

Question 2

An empirical analysis of continuing improvements following the implementation of a performance-based compensation plan (Banker, Lee, Potter and Srinivasan, 2000)

Takeaways relating to SELECTION & EFFORT EFFECTS of incentives

Answer 2

1. This paper shows that SELECTION effect is STRONGER in this (retail) company, accounted for the bulk of increase in productivity & hence also extremely important
   - although results might not be generalisable to other companies
   - usually effort effect is stronger
2. Selection effect implies…
   - sales of OLDSTAY > OLDQUIT and NEWQUIT
   - sales of NEWSTAY > OLDQUIT and NEWQUIT
3. Effort effect implies…
   - sales of all STAY increases over time

Question 3

3 purposes/effects of incentives
2. SELECTION

Answer 3

1. attracting & retaining valuable and/or HIGH-PERFORMING employees
   - naturally, employees who do like the incentive mechanisms are likely to stay at the firm (& those who don’t are likely to LEAVE)
2. In settings where the org. cannot use neither results control nor action control, PERSONNEL CONTROL becomes very important
   eg. new employees being screened on “fit”
3. also hiring employees with specific preferences
   eg. co.s with social goals paying BELOW-MARKET PAY to attract employees who are VALUE-CONGRUENT and not financially driven

Question 4

Why do some firms still give share-based compensation/STOCK OPTIONS to lower-level employees? (Oyer, 2004)

Although it seems to contradict what we previously discussed about the inability of lower-level employees to influence share price

Answer 4

SELECTION & RETENTION of the right type of employees!
- when the economy is booming, employees are more likely to stay within the org. instead of going to competitor who wants to pay more
- b/c the stock will also be worth more when the economy is growing
- so if pay is not adjusted, employees might leave

Question 5

3 purposes/effects of incentives
3. MOTIVATIONAL

+ reference to x4 academic articles

Answer 5

1. increase productivity/EFFORT
   - addressing lack of motivation
2. PRIORITISE tasks in multitask settings
   - addressing lack of direction

+ evidence that incentives motivate managers to engage in FRAUD/MISREPORTING too
» b/c we are putting more incentives on the areas that employees have MORE IMPACT
eg. last week’s target ratcheting & effort reduction

Reference to the x4 academic articles
- The more closely tied CEO compensation is to STOCK PRICE, the higher the likelihood of MISREPORTING/earnings mgmt
- also misconduct at non-executive levels (KPMG survey)

Question 6

2 important questions for incentive design
1. When should we use STRONGER INCENTIVES?

Answer 6

Given that there are no tradeoffs (eg. quality vs quantity), how much incentive to give depends only on 2 factors {both in denominator}:
1. r is the agent’s RISK AVERSION
- shifting some of org’s risk to employees
- more risk averse, lower incentive b/c we already know they won’t accept the more incentivised contract
- more risk neutral, higher incentive

2. sigma is the NOISE in the PERFORMANCE DIMENSION (ie. performance measure is affected by outside factors instead of employees)
   - high noise, lower incentive
   - don’t want to pay employees for randomness

Question 7

2 important questions for incentive design
2. When should we use MULTIPLE MEASURES?

Relative weights: put more emphasis on the performance measure that will pick up MORE VARIATION, not on the measure that will be more affected by noise

Answer 7

1. INFORMATIVENESS of the 2nd performance measure
   eg. customer satisfaction measure to supplement sales measure
2. If the 2nd measure tell us something NEW about the action we want to incentivise (1st PM is insufficient)
   eg. increasing customer satisfaction predicts higher sales in the FUTURE, compared to only measuring current sales
3. SENSITIVITY of 2nd PM
   - does the measure pick up MANAGER’S ACTIONS?
4. PRECISION of 2nd PM
   - is the measure affected by outside factors?

Question 8

Bank of the Desert (A) case
1. To what extent are the performance measures PROFIT DRIVERS?
2. How to check if performance measures are equally important? If not, what would be the “optimal” incentive weight for each performance measure?

Answer 8

Check if there is a RELATIONSHIP between the measures & profit.
1. Look at p-values < 0.05 for evidence of a STATISTICAL relation with (profit)
- if yes, the performance measures are informative & could potentially be included in an incentive-compensation contract
2. Coefficients are not useful if variables are measured in different UNITS, making the measures not comparable!
- Hence, no evidence that a measure is more important than another; should first STANDARDISE the variables to make the COEFFICIENTS COMPARABLE
*t-statistics and p-values remain the same b/c underlying data didn’t change

2. Compare the COEFFICIENTS after ensuring that the measures are in the SAME UNIT
   eg. cross-sell is associated with £xx increase in profit vs product sales is associated with £xxx increase in profit
   - To determine their RELATIVE IMPORTANCE and thus the OPTIMAL WEIGHTS, simply divide the standardised coefficients