Some Economics Insights for Musk's DOGE

Firms seek to maximize their profits. Consumers seek to maximize their utility. What do governments maximize? If in 2025 , the U.S Government fails to “maximize” this criteria, will Musk’s DOGE pinpoint this inefficiency and enact a solution that gets us “closer to the maximum”? Does economic logic offer any lessons here that will help Musk’s Team to do their work? How do we spot government inefficiency? How do we measure it? How do we infer the counter-factual of how more efficient the government would be under a different set of rules?

I’m thinking about these issues after reading this piece. A former member of the Trump Administration’s NOAA has written a New York Times piece making the case that this Agency’s staff and budget not be severely cut by the nascent Musk DOGE.

The author is arguing that NOAA provides key public goods that help us to be aware (and adapt) to emerging weather challenges. Musk’s DOGE is unlikely to end NOAA. Instead, the question going forward is; “What is the optimal size and agenda for NOAA?”.

Economists emphasize the importance of marginal analysis. For each person hired by Government such as at the Department of Education or NOAA, what is the marginal benefit to society of this person’s output? The marginal cost is clear. The worker must be hired and paid and benefits such as health insurance must be paid for.

How do economists measure the marginal benefits of an extra government employee?

To answer this question, let’s answer a simpler question. What are the benefits to a pizza restaurant from hiring an extra worker?

Intro Econ 101 teaches us that we first ask; how much more pizza will the pizza place produce because it hired one more worker. Call this; Delta Pizza. This function is unlikely to be a constant because of diminishing returns in the underlying production function. If a pizza place has 1 oven and 900 cooks, hiring a 901st cook will just congest the kitchen!

Now, define the price of a slice of pizza as $P. So, P could equal $3. In this case, the marginal benefit to the firm of hiring this worker is P*Delta Pizza. The units are dollars! The firm will compare this to the cost of hiring this worker.

How Does Each Government Agency Calculate its Own “P*Delta Pizza”?

This is the key issue that DOGE will confront. Government services often do not have a market price so what marginal value of output should government use? Also, what is “Delta Pizza”. What are the units of output that government creates? Perhaps you will answer by saying; “information” or “national security”? These words are words that you can’t define. Write down your production function such that hiring more workers yields a measure of output that can be measured (Delta Pizza) and that can be priced using a market to determine its marginal value to society ($P). If you can’t do this, then Musk’s team can trim and trim and you won’t know if this is good or bad.

Now, in some cases we have privatized the government service and this allows researchers to see the counter-factual of what a private for profit firm would do if it chose how much of the good to produce and how to produce it.

Musk’s team should read our 2017 paper.

Jerch, Rhiannon, Matthew E. Kahn, and Shanjun Li. "The efficiency of local government: The role of privatization and public sector unions." Journal of Public Economics 154 (2017): 95-121.

Now, I recognize that a justification for government is that government provides public goods and mitigates negative externalities. The DOGE will have to address how productive is each agency in producing these services? What will be lost at the margin if these agencies slim down in terms of output (Delta Pizza) and lost value to society? For example if by shrinking the CDC, does the probability of the next COVID crisis increase by 1%? What do we lose if this occurs?

How will Musk’s team measure this lost output? How much does our national security suffer if we reduce the military’s budget by X%? What do we lose when government shrinks in size?

Or would you optimistically argue that we will gain as government agencies lose access to their “soft budget constraints”? Do you believe that when agencies face a hard budget constraint that they think harder about their priorities and focus on core issues rather than branching out into stuff that pleases the current President?

Another point; Each Cabinet Agency’s leader has private information about his agency’s budget fat. Under what incentive mechanism would these agency leaders truthfully reveal this information to Musk? What does the field of mechanism design in economics teach us here? Will Musk embrace economic logic for helping his agency to overcome the private information problem?

A final point; Suppose that DOGE focuses on shrinking the IRS. If the probability of a tax audit declines, Becker’s theory of cheating predicts that more Americans will report less income to the IRS and the revenue of the U.S government will shrink. Musk’s actions would “starve the beast” (David Stockman’s words). Would this be good or bad?

The probability of a tax audit is similar to the “Delta Pizza” mentioned above. If the IRS has one less auditor than the probability of a tax audit declines. In 2025, What is the optimal size of the U.S Federal Government? How many people should work at each agency? What should these people be doing all day long? How do we know if they are productive at their job? In the private sector, competition spurs efficiency and a focus on the consumer’s desires. In the public sector, what incentives do government employees have to be curious about what the consumers want? Will DOGE change who chooses to work for the Federal and State Governments and how they are paid and how they are evaluated? Going forward, will more MBA graduates work for government in slimmer, high powered incentive regime?