ECON5323 Organisational Economics
Managing Careers Part I
, Recap
• In the “moral hazard” section, the main question was how to choose a
compensation scheme to attain the pie-maximising amount of effort from
an employee. Ingredients of most models were:
1. Some degree to which an action by the agent is not observable (hence
potential for moral hazard problem)
2. Legally binding contracts possible on whatever was observable (hence no
commitment problem)
• These models are most appropriate when thinking about how to design
contracts and compensation schemes for short-term projects
• This section: careers and incentives within long-term employer-employee
relationships
, Education and the inability to commit
• So far, we have considered workers to be homogenous - effort is equally
costly for all workers, no workers are more or less talented/skilled/lazy
than other workers
• One relevant aspect of a long-term employer-employee relationship is that
there may be gains on both sides from the worker engaging in training to
boost his productivity
• for short-term contracts or less skilled positions, this may be less relevant
• But a problem with long-term relationships is that not everything is
contracted
• Suppose a worker in a firm can engage in useful training, but the firm
cannot commit to any specific reward for the worker on completion of this
training
• We want to see how the choice of how much training to get depends on
• how specific to this employer the training is
• who’s paying for it
, Basic model: non-specific training, employer pays
!
• The level of training chosen is 𝑥 , at a cost of " 𝑥 " to the employer, called
Firm A
• this is a monetary cost, not a “psychological” one like the cost of effort
• When the worker is trained at level 𝑥 , he produces a profit of 𝑥 for Firm A
• However, since the training is completely non-specific, once the worker has
been trained, he can take his increased productivity elsewhere (e.g. to
Firm B)
Managing Careers Part I
, Recap
• In the “moral hazard” section, the main question was how to choose a
compensation scheme to attain the pie-maximising amount of effort from
an employee. Ingredients of most models were:
1. Some degree to which an action by the agent is not observable (hence
potential for moral hazard problem)
2. Legally binding contracts possible on whatever was observable (hence no
commitment problem)
• These models are most appropriate when thinking about how to design
contracts and compensation schemes for short-term projects
• This section: careers and incentives within long-term employer-employee
relationships
, Education and the inability to commit
• So far, we have considered workers to be homogenous - effort is equally
costly for all workers, no workers are more or less talented/skilled/lazy
than other workers
• One relevant aspect of a long-term employer-employee relationship is that
there may be gains on both sides from the worker engaging in training to
boost his productivity
• for short-term contracts or less skilled positions, this may be less relevant
• But a problem with long-term relationships is that not everything is
contracted
• Suppose a worker in a firm can engage in useful training, but the firm
cannot commit to any specific reward for the worker on completion of this
training
• We want to see how the choice of how much training to get depends on
• how specific to this employer the training is
• who’s paying for it
, Basic model: non-specific training, employer pays
!
• The level of training chosen is 𝑥 , at a cost of " 𝑥 " to the employer, called
Firm A
• this is a monetary cost, not a “psychological” one like the cost of effort
• When the worker is trained at level 𝑥 , he produces a profit of 𝑥 for Firm A
• However, since the training is completely non-specific, once the worker has
been trained, he can take his increased productivity elsewhere (e.g. to
Firm B)
