Here’s a fun game. Name the largest welfare program in the United States.

If you said Social Security, you’re thinking too literally. If you said Medicare, you’re warm but still inside the box. The correct answer—and I’m being only slightly provocative here—is the United States Department of Defense.

I don’t mean this in the “war is a waste of money” sense. I mean it structurally. The DoD is an $850 billion annual jobs program¹ that provides stable employment, comprehensive healthcare, housing subsidies, educational benefits, and a pension to approximately 2.9 million people²—many of whom would otherwise be competing in labor markets that have, in many regions, simply stopped producing good jobs.

This last point matters. Since roughly 1980, economic growth has concentrated in a handful of metro areas while vast stretches of the country experienced stagnation or decline. Labor force participation for prime-age men without college degrees fell from 97% in 1964 to 83% by 2015.³ Deaths of despair—suicide, overdose, alcoholic liver disease—more than tripled between 1992 and 2017 among middle-aged white Americans without a bachelor’s degree.⁴ The political system has not figured out how to address this directly. “Jobs programs for economically distressed regions” polls terribly, even among people who would benefit.

And so we’ve developed alternative delivery mechanisms.

The Packaging Is the Product

Americans, broadly speaking, don’t like welfare. What’s interesting is how much welfare Americans receive while believing they don’t receive any, because the delivery mechanism has been carefully designed to not pattern-match to “welfare.”

The political scientist Suzanne Mettler calls this the “submerged state”—government benefits delivered through the tax code, private employers, or complicated institutional arrangements that obscure their origin. But I think there’s something even more interesting going on: we’ve built entire industries that exist primarily to employ people, where the employment itself is the product, not whatever the industry nominally produces.

Consider what a welfare program needs to do:

1. Transfer resources to people who might otherwise lack them
2. Do so in a way that preserves dignity and social standing
3. Maintain political support from constituencies that might otherwise oppose redistribution

The genius of America’s hidden welfare state is that it accomplishes all three by disguising transfers as “jobs”—which are the one form of getting money that almost everyone agrees you deserve.

This comes at a cost. Hidden welfare is deeply inefficient welfare—often 2-3x more expensive than direct transfers would be to deliver the same income support. We’re paying a substantial premium for the ideological laundering. Whether that premium is worth it is a genuine question, but we should at least be clear that we’re paying it.

A Field Guide to American Workfare

Let me walk you through the main programs. I’ll try to be fair, which means I’ll probably annoy everyone.

The Military-Industrial Employment Complex

The facts first: Military recruitment correlates strongly with local economic distress. Base locations function as de facto regional development policy. Defense contracts are routinely justified on jobs grounds by legislators who don’t even pretend the strategic rationale is primary—see the saga of Congress ordering tanks the Army explicitly says it doesn’t want.

The implicit benefits package is comprehensive: healthcare for you and your family, housing allowance, educational benefits via the GI Bill, a pension if you stay twenty years. All of it earned through service, which is to say, wrapped in an honor narrative that makes it psychologically and politically impossible to categorize as welfare.

The counterargument writes itself: “These aren’t handouts, they’re compensation for difficult and dangerous work that serves the nation.” This is true! And it’s also true that we’ve chosen to structure national defense in a way that maximizes jobs-per-dollar rather than defense-per-dollar, and that this choice has obvious political economy explanations. We are, in effect, paying a premium for the welfare-laundering service.

I’m not saying this is wrong. I’m saying we should be honest about the function.

The Healthcare Administrative Complex

This one is wild.

The United States spends over $1,000 per person on healthcare administrative costs—approximately five times more than the average of other wealthy countries.⁵ We’re talking millions of jobs: billing specialists, prior authorization nurses, claims adjusters, denial management consultants, revenue cycle analysts, and my personal favorite job title, “insurance navigator.”

These are real jobs requiring real skills. Billing and coding specialists go through certification programs. They support families. They have health insurance through their employers (there’s a nice irony).

They also exist because we’ve designed a system of baroque complexity that requires an army of people to intermediate between patients, providers, and dozens of competing payers with different rules, formularies, and authorization requirements. The system’s inefficiency is the employment. Administrative complexity accounts for roughly 30% of excess U.S. healthcare spending compared to peer countries—the single largest component.⁶

A serious single-payer proposal would need to deal with this. The efficiency savings that advocates cite are substantially composed of these salaries. You’re not eliminating waste; you’re firing people whose livelihoods depend on it. This is one reason why single-payer polling collapses when you mention “eliminating private insurance.”

The Agricultural Support System

Farm subsidies are officially about food security and preserving family farms. The reality is that payments increasingly flow to large agribusiness, but the entire ecosystem—equipment dealers, seed and fertilizer suppliers, rural banks, ag extension services—employs vastly more people than the number of family farmers anyone is pretending to protect.

The ideological wrapper is beautiful: self-reliance, land stewardship, feeding America. Never mind that much of what we subsidize is corn for ethanol and export commodities. The farmers work hard. They embody cultural values. They are absolutely not on welfare.

The Carceral-Administrative Complex

Prisons employ approximately 400,000 people directly as correctional officers and jailers,⁷ with the broader ecosystem—courts, probation officers, bail bondsmen, private prison staff, prison vendors—adding substantially more.

Here’s the thing you might not know: prisons are deliberately sited in economically distressed rural areas as a development strategy. Towns actively compete for prisons. When reform advocates propose reducing incarceration, legislators object that this would devastate their districts. The prison guard union in California (CCPOA) has been one of the most effective opponents of criminal justice reform for exactly this reason.

The ideological framework (“tough on crime,” “public safety”) provides cover for what is, in effect, a rural jobs program funded by urban taxpayers to warehouse urban residents. This is… a lot.

The Tax Preparation Industry

This is the cleanest example because it’s almost pure rent-seeking.

The IRS could do your taxes for you. Many countries do this—they send you a form that says “here’s what we think you owe based on the W-2s and 1099s we already received, sign here if you agree.” The IRS developed this capability. Congress killed it. Intuit and H&R Block have spent decades lobbying to maintain tax complexity and block free filing.

The result is an industry that employs roughly 300,000 people, mostly seasonally, to solve a problem that shouldn’t exist. The workers are real. The skills are real. The social function is fake.

Real Estate Transaction Costs

American homebuyers pay 5-6% in transaction costs—mostly agent commissions, but also title insurance, closing attorneys, and various fees. This is roughly 2-3x the rate in most peer countries.

The industry employs approximately 1.5 million people as agents alone, most of them part-time or marginal performers in a tournament structure where the median income is quite low.

Are these jobs or is this a toll booth? Both. The “jobs” defense of high commissions isn’t crazy—you really would eliminate many people’s livelihoods by moving to a lower-cost model. But we should acknowledge that we’re paying a premium to keep them employed, and that premium falls on every home transaction in America.

Why This Equilibrium Is Stable

Each of these systems shares a common structure:

Concentrated beneficiaries, diffuse costs. Defense jobs are concentrated in specific districts; the cost is spread across all taxpayers. Healthcare administrators have an industry association; the premium is buried in everyone’s insurance.

Ideological insulation. The framing—patriotism, markets, public safety, small business, homeownership—makes beneficiaries resistant to seeing themselves as welfare recipients and the public resistant to seeing the spending as welfare.

Real human stakes. The jobs are real. The skills are real. The people are real. Any reform has to answer “what happens to them?” and the honest answer is usually “something bad, at least transitionally.”

Missing counterfactual. We don’t know what these people would do in a more efficient system. They might flourish! They might not. The uncertainty makes “preserve the status quo” feel like the safe choice.

The Reform Dilemma

Here’s where I’ll probably lose whatever coalition I’ve assembled by annoying everyone equally.

If you’re on the political right, you might say: “Great, let’s cut these inefficient programs and let the market sort it out.” But the market sorting it out looks like regional economic crises, and the revealed preference of conservative legislators is to protect defense bases and farm subsidies with great ferocity. The rhetoric is free markets; the practice is jobs preservation.

If you’re on the political left, you might say: “Let’s make the welfare explicit. Replace hidden transfers with visible ones—UBI, direct payments, actual welfare.” But you’ll immediately discover that Americans hate visible welfare, even Americans who benefit from hidden welfare. The dignity of a “job” is not replaceable with the dignity of a “transfer,” at least not in our current cultural configuration.

The true political economy insight is that efficiency and legitimacy are in tension in ways that don’t fit neatly into either ideological framework. We’ve arrived at systems that are inefficient because the inefficiency is the political mechanism that allows the redistribution to occur.

Where Does This Leave Us?

A few observations, in lieu of a conclusion:

1. Be suspicious of “efficiency” arguments that ignore transition costs. When someone says we can save $X by eliminating system Y, ask who currently gets that $X as salaries. This doesn’t mean inefficient systems are good. It means the savings are someone’s income.

2. But be honest about the markup we’re paying. Hidden welfare costs 2-3x what explicit welfare would cost to deliver the same income support. We’ve decided the packaging matters more than the contents. That’s a choice, and we should own it.

3. Identity matters as much as income. A billing specialist with a credential, a career ladder, and an employer is socially and psychologically different from a transfer recipient with the same income. We can argue about whether this should be true, but it is true, and it shapes what’s politically feasible.

4. The hidden welfare state is bipartisan. Rural welfare is wrapped in conservative values (farming, military, law enforcement). Suburban welfare is wrapped in market values (healthcare “industry,” real estate “profession”). Urban welfare is the only kind we actually call welfare, which is one reason it’s politically vulnerable.

5. Honesty about welfare is probably not a winning political strategy. No legislator will campaign on “I support inefficient job creation as implicit redistribution.” The whole thing works because we don’t acknowledge it.

I don’t have a clean recommendation here. The system exists because it solves multiple problems simultaneously: it redistributes, it employs, it maintains dignity, and it maintains the ideological fiction that distinguishes deserving from undeserving recipients of public money. It does all of this expensively. Whether a better system is possible probably depends on solving the identity problem first, and I don’t see much progress on that front.

But I do think it’s worth understanding what we’re actually doing. America has one of the largest welfare states in the developed world, and one of the least efficient. We just refuse to call it that.

The author would like to thank his future self for not having to defend this piece at Thanksgiving dinner, since he strategically published it in January.