A principal is an owner. An agent works for the owner, so you can think of an agent as an employee. The difference between a founder and an employee is the difference between a principal and an agent.
I can summarize the principal-agent problem with a famous quote attributed to Napoleon or Julius Caesar:
“If you want it done, Go. If not, Send.”
Which is to say: If you want to do something right, do it yourself; because other people just don’t care enough.
A principal’s incentives are different than an agent’s incentives
Now, the principal-agent problem pops up everywhere. In microeconomics, they try to characterize it this way: The principal’s incentives are different than the agent’s incentives, so the owner of the business wants what is best for the business and will make the most money. The agent generally wants whatever will look good to the principal, or might make them the most friends in the neighborhood or in the business, or might make them personally the most money.
You see this a lot with hired-gun CEOs running public companies, where the ownership of the public company is distributed so widely that there’s no principal remaining. Nobody owns more than 1% of the company. The CEO takes charge, stuffs the board with their buddies and then starts issuing themselves low-priced stock options, or doing a lot of stock buybacks because their compensation is based directly tied to the stock price.
If you can work on incentives, don’t work on anything else
Agents have a way of hacking systems. This is what makes incentive design so difficult. As Charlie Munger says,
“Never, ever, think about something else when you should be thinking about the power of incentives.”
Almost all human behavior can be explained by incentives. The study of signaling is seeing what people do despite what they say. People are much more honest with their actions than they are with their words. You have to get the incentives right to get people to behave correctly. It’s a very difficult problem because people aren’t coin-operated. The good ones are not just looking for money—they’re also looking for status and meaning in what they do.
As a business owner you are always going to be dealing with the principal-agent problem. You’re always going to be trying to figure out: How do I make this person think like me? How do I incent them? How do I give them founder mentality?
Only founders can fully appreciate the importance of founder mentality and just how difficult and gnarly the principal-agent problem is.
When you do deals, it’s better to have the same incentives
If you are a principal, you want to spend a lot of your time thinking about this problem. You want to be generous with your top lieutenants—in terms of ownership and incentives—even if they don’t necessarily realize it; because over time they will and you want them to be aligned with you.
When you do business deals, it’s better to have an aligned partnership where you both have the same incentives than a partnership where you have the advantage in the deal. Because eventually the other person will figure it and the partnership will fall apart. Either way, it’s not going to be one of those things that you can invest in and enjoy the benefits of compound interest over decades.
If you’re an employee, your most important job is to think like a principal
Finally, if you’re in a role where you’re an agent—you’re an employee—then your most important job is to think like a principal. The more you can think like a principal, the better off you’re going to be long-term. Train yourself how to think like a principal, and eventually you will become a principal. If you align yourself with a good principal, they will promote you or empower you or give you accountability or leverage that may be way out of proportion to your relatively menial role.
I’m always impressed by founders who promote young people through the ranks and allow them to skip multiple levels despite their lack of experience. Invariably it happens because this agent—who’s way deep down in the organization—thinks like a principal.
If you can hack your way through the principal-agent problem, you’ll probably solve half of what it takes to run a company.