Government Efficiency: An imagined dialogue
This is an excerpt from the Substack of Don Moynihan, a professor of Public Policy, at the University of Michigan.
What would it be like to ask DOGE leaders about their management strategies?
We have some ideas about how to make government work better.
Cool, what’s your background?
We have created some of the best companies in the world. And we understand tech innovation. Well, some of us do. . . .
Uh, okay, what are your ideas?
First, everyone should write down what they did every week, and send it to someone they don’t know who doesn’t supervise them.
Hmm…and then what?
Well, we will use the information to evaluate priorities.
So you understand the policy area and value of what is being done?
No, AI will help us.
Have you built and tested the AI?
No, that comes later.
Any other management tricks.
Yes, we put employees in a constant state of fear. We fire them and rehire them and refire them again. Make them return to office even as we are trying to sell their office space. Tell them they can resign, but then put the person processing the resignations on administrative leave. . . .
They must love that.
Haha, yes. It’s driving a lot of people who love working for government to leave. We forced some to stay up for 36 hours straight to help fire their colleagues. . . .
We are also cutting unnecessary spending.
Wow, seems like a big deal. How do you know it’s unnecessary? . . .
Some guy on X said it was. And none of it is really necessary if you think about it. Which we don’t.
But don’t some of these services exist because people value them?
We don’t value them. But never mind. In the future, AI will do all of this work. You don’t really need humans with specialist expert knowledge and skills when you have AI.
But you haven’t built the AI before firing all the staff.
No. Why do you ask? Look, Twitter went down occasionally, but its still going. . . .
Seems like a government would be different. Like mistakes could be a big deal. . . .Like, if the tax agency can’t collect money, don’t you lose hundreds of billions, maybe even trillions.
The real issue is that you are saving millions.
We have some ideas about how to make government work better.
Cool, what’s your background?
We have created some of the best companies in the world. And we understand tech innovation. Well, some of us do. . . .
Uh, okay, what are your ideas?
First, everyone should write down what they did every week, and send it to someone they don’t know who doesn’t supervise them.
Hmm…and then what?
Well, we will use the information to evaluate priorities.
So you understand the policy area and value of what is being done?
No, AI will help us.
Have you built and tested the AI?
No, that comes later.
Any other management tricks.
Yes, we put employees in a constant state of fear. We fire them and rehire them and refire them again. Make them return to office even as we are trying to sell their office space. Tell them they can resign, but then put the person processing the resignations on administrative leave. . . .
They must love that.
Haha, yes. It’s driving a lot of people who love working for government to leave. We forced some to stay up for 36 hours straight to help fire their colleagues. . . .
We are also cutting unnecessary spending.
Wow, seems like a big deal. How do you know it’s unnecessary? . . .
Some guy on X said it was. And none of it is really necessary if you think about it. Which we don’t.
But don’t some of these services exist because people value them?
We don’t value them. But never mind. In the future, AI will do all of this work. You don’t really need humans with specialist expert knowledge and skills when you have AI.
But you haven’t built the AI before firing all the staff.
No. Why do you ask? Look, Twitter went down occasionally, but its still going. . . .
Seems like a government would be different. Like mistakes could be a big deal. . . .Like, if the tax agency can’t collect money, don’t you lose hundreds of billions, maybe even trillions.
The real issue is that you are saving millions.