Incentives
An incentive is something that motivates an actor to do something. Two actors are said to have aligned incentives if they are most benefitted by the same action or outcome.
The agency problem
The agency problem or principal–agent problem occurs when one actor trusts another actor to take actions which are in their best interest. Humans tend to defect fairly reliably, and become increasingly likely to do so as the reward for doing so increases.
The solution to the agency problem is incentive alignment. This can take many forms.
- Punishment - if it is recognised post-hoc that you did not fulfil your duty, something will be taken away from you.
- Ideology - everyone has an incentive to act in accordance with their own set of morals. If you teach people (usually children) competently that failing to uphold a particular duty is inherently wrong, they’ll continue to do so because it feels bad to do otherwise.
- Shared outcome - make someone involved in the result of their actions. For example, by giving salespeople a commission, you incentivise them to create more sales. By paying people to report tax evasion, you get more reports of it occurring.
- Path of least resistance - make it convenient to do the right thing. Music streaming mostly killed music piracy because it was easier. Frequent waste disposal points reduces littering.
Incentive alignment becomes increasingly difficult as the number of actors in a system increases. It become especially difficult once you have enough multiple layers of management, each with their own layered incentives. Making all of those incentives meet at the point of actual value creation is incredibly difficult.
I think most political ideologies should live and die on their ability to solve the agent-actor problem and other incentive challenges.