Does DOGE Show That There's Little Government Waste?

No. That doesn't follow.

Matt Yglesias makes a point that we will hear a lot from Democrats in the coming months. Progressives will claim that because DOGE “failed” to live up to its ambitions to slash government spending by trillions, there must be little government waste. Indeed, economist Alan Blinder even says that Elon Musk should examine Social Security as a model of efficiency, given its administration costs only 0.5 percent of benefits paid out.

Hmm.

DOGE’s ambition to cut trillions in annual spending was always wildly overambitious without the support of Congress and with DOGE’s limited focus on federal employees, grants, software, and eradicating waste and fraud. Clearly, DOGE had other goals beyond its spending ambitions anyway. But does DOGE’s smaller estimated savings of $165 billion (many think it’s lower still) really show that the federal government is a paragon of efficiency and devoid of “waste”?

I’m sorry, but no.

1. Even the Government Accountability Office itself says that “the federal government loses between $233 billion and $521 billion annually to fraud,” which is a sub-component of “improper payments,” before we even get to the broader conception of “waste.” Reducing much of this excess, especially within health care programs, is likely to require changes to systems, some of which DOGE says it has been doing. Even if DOGE wasn’t able or willing to eradicate it, though, that doesn’t mean this sort of stuff doesn’t exist. Indeed, Tyler Cowen suggests the real conclusion might be that we simply do not wish to get rid of waste in health care and defense.

2. Some of the findings from DOGE do suggest, moreover, that other programs have higher improper payment levels than people may have assumed. The American Enterprise Institute’s Andrew Biggs, for example, writes that Social Security Administration sampling of those aged over 100 seemed to find that 18.2 percent of those in receipt of payments were dead. Applied to the full population in that age range, that suggests $340 million in lost benefits per year alone. How much of this and other problems DOGE got to grips with is as yet unclear.

3. DOGE clearly made some false economies on staffing—firing people that did important work as part of indiscriminate layoffs. Yet who really knows what the necessary or optimal staffing level is to fulfil all existing government functions, especially given evolving AI capabilities? Let’s put it this way: given the incentives within government, I highly doubt that agencies, programs, and departments had precisely the “right” numbers of workers at the “right” pay rates. And the fact that transfer programs like Social Security are not particularly labor intensive in dishing out income transfer benefits proves nothing about whether government is efficient in terms of employee numbers overall.

4. “Waste” is subjective, of course. The GAO defines it as when individuals or organizations spend government resources carelessly, extravagantly, or without purpose. But I suspect most Americans understand it more expansively to include spending that’s ineffective, has costs greater than its benefits, enriches those other than its intended beneficiaries, or else muddies the waters on clear priorities. I thought Yglesias was sympathetic to “supply-side progressivism.” Isn’t their whole critique that the government imposes secondary objectives or adds social policy goals to economic projects, raising costs and causing delays? Is that not “waste”? Does DOGE failing to cut more mean Jennifer Pahlka was wrong about the inefficiencies of government that stem from layering new systems onto old, or contracting out too much? This seems a rather strange conclusion for left-leaning Democrats to make.

5. We libertarians would go much further, of course. Everything outside of government’s core functions—defense, enforcing law, the courts, and certain public goods—could be considered waste. But even if you’re unwilling to go that far, many policies have far more spending than necessary, or are frankly unsuccessful in their major objectives. Most economic development aid to poor countries doesn’t appear to improve development. Medicare pays vastly different amounts for the same treatments across different venues. Federal job training programs regularly fail to improve participants' employment outcomes. Social Security payments to wealthy retirees aren’t necessary for any safety net or anti-poverty objective. And educational achievement gaps remain stubbornly unchanged despite decades of federal spending.

It would be frankly ridiculous to assume that DOGE’s failure to cut more spending shows that there’s little fat left in the federal government. Yes, we can all agree that the big long-term driver of future deficits is the entitlement programs. Any efforts to right the fiscal ship will have to tackle them. But that’s not the same as implying the federal government is efficient.