DOGE is not worth engaging. You can’t cut your way to a federal government that does more.

The Department of Government Efficiency (DOGE) is the effort largely associated with Elon Musk. The stated goals are to reduce wasteful spending and eliminate unnecessary regulations. Unsurprisingly, the Trump administration and congressional Republicans are taking it seriously. More worryingly, there are good-faith proponents of improving the federal government’s capacity who seem willing to take it seriously.

This would be a mistake, as it’s clearly a bad-faith effort rooted in ignorance and a knee-jerk desire to shrink the federal government, both for ideological reasons and the creation of space to preserve the tax cuts for the rich and corporations that will be locked-in later this year.

It’s somewhat understandable that good-faith proponents of rebuilding state capacity might be tempted to engage with a disruptive new political effort claiming to target government “efficiency.”  This state capacity—the ability of a government to accomplish its goals (like enforcing the laws that have passed, providing the public goods and safety net programs it has committed to, and collecting the taxes it is owed)—has catastrophically eroded in recent decades.

This erosion was on full display all through the COVID-19 pandemic. When problems during that crisis required state capacity to resolve, we stumbled badly over vital tasks like screening applicants for unemployment insurance to get their payments to them in a timely fashion, standing-up testing programs, or sourcing and installing air filters and other equipment to allow faster return to in-person schooling.

Recent writing on state capacity and its failures often rebukes progressives for focusing too much attention on changing the public sector’s goals and not enough on the public sector’s ability to accomplish them. It’s a fair charge. Even worse, there is a convincing case to be made that well-meaning attempts to defend state capacity from public skepticism—attempts that prioritized procedure over outcomes—have actually backfired and have contributed to the long-running erosion of state capacity.

But DOGE does not make any useful contribution to this important discussion. DOGE’s origin story was an off-the-cuff remark made by Elon Musk that the federal government could be shrunk by a third without harm. This clearly does not reflect a drive to make the federal government more effective; it reflects a drive to starve it of resources.

And the federal government is by many measures already resource-scarce, and this scarcity has been a big driver of the erosion of state capacity. Figure 1 below, for example, shows the share of total employment in the United States accounted for by civilian federal government employees. Between 1979 and 2019—or between the time when the Reagan administration first made hostility to federal government employees a key plank of conservative policymaking and through the first Trump administration—this measure has fallen by roughly a third. 

Figure 1

Federal employment fails to keep pace with wider growth in economy: Federal civilian full-time equivalent employees (FTEs) as a share of all domestic FTEs, 1948–2023

Year Share of federal civilian FTEs 
1948 2.87%
1949 2.99%
1950 2.92%
1951 3.33%
1952 3.43%
1953 3.22%
1954 3.11%
1955 3.05%
1956 3.03%
1957 3.00%
1958 3.02%
1959 2.97%
1960 3.00%
1961 3.01%
1962 3.03%
1963 3.02%
1964 2.93%
1965 2.85%
1966 2.86%
1967 2.97%
1968 2.92%
1969 2.81%
1970 2.75%
1971 2.75%
1972 2.69%
1973 2.53%
1974 2.55%
1975 2.64%
1976 2.57%
1977 2.49%
1978 2.40%
1979 2.33%
1980 2.37%
1981 2.31%
1982 2.36%
1983 2.34%
1984 2.28%
1985 2.25%
1986 2.20%
1987 2.18%
1988 2.15%
1989 2.12%
1990 2.14%
1991 2.14%
1992 2.13%
1993 2.05%
1994 1.91%
1995 1.80%
1996 1.70%
1997 1.62%
1998 1.56%
1999 1.51%
2000 1.49%
2001 1.46%
2002 1.50%
2003 1.54%
2004 1.53%
2005 1.51%
2006 1.48%
2007 1.48%
2008 1.53%
2009 1.71%
2010 1.81%
2011 1.77%
2012 1.72%
2013 1.67%
2014 1.62%
2015 1.59%
2016 1.57%
2017 1.55%
2018 1.53%
2019 1.53%
2020 1.68%
2021 1.63%
2022 1.57%
2023 1.58%
ChartData Download data

The data below can be saved or copied directly into Excel.

Source: Bureau of Economic Analysis (BEA), National Income and Product Accounts Tables 6.5.B – 6.5.D  

Copy the code below to embed this chart on your website.

Many frustrations with government today stem largely from a simple lack of public-sector employees available to perform the work in a timely and effective way—from processing asylum claims in reasonable time frames to enforcing consequences for affluent taxpayers who refuse to pay the taxes they owe to holding employers who flout workplace regulations accountable. This frustration with government is not a problem for people like the ones pushing DOGE—it is the desired outcome.

Finally, in some ways, the shrinking share of federal civilian employment understates the strain we’ve put on federal employees from understaffing. At the same time as we have largely held the head count of federal employees constant for decades, we have expanded the activities and the money flows they are responsible for administering. Figure 2 shows the aggregate compensation of federal civilian government employees as a share of total spending by the federal government over time. This measure has fallen by 45% since 1979.

Figure 2

Stagnant federal employment administers more federal dollars: Compensation of civilian federal employees as % of federal government expenditure, 1948–2023

Year Compensation of civilian federal employees as % of federal government expenditure
1948 12.51%
1949 12.04%
1950 12.70%
1951 13.60%
1952 13.18%
1953 12.36%
1954 12.21%
1955 12.81%
1956 13.17%
1957 12.74%
1958 12.81%
1959 12.81%
1960 13.26%
1961 13.24%
1962 12.98%
1963 13.20%
1964 13.50%
1965 13.42%
1966 12.88%
1967 12.20%
1968 12.09%
1969 12.08%
1970 12.05%
1971 12.35%
1972 11.94%
1973 11.74%
1974 11.36%
1975 10.67%
1976 10.72%
1977 10.65%
1978 10.51%
1979 10.09%
1980 9.40%
1981 8.83%
1982 8.42%
1983 8.08%
1984 8.09%
1985 7.95%
1986 7.59%
1987 7.48%
1988 7.65%
1989 7.54%
1990 7.44%
1991 7.57%
1992 7.20%
1993 7.22%
1994 7.12%
1995 6.77%
1996 6.58%
1997 6.49%
1998 6.50%
1999 6.55%
2000 6.71%
2001 6.50%
2002 6.73%
2003 6.85%
2004 6.89%
2005 6.80%
2006 6.62%
2007 6.54%
2008 6.31%
2009 6.27%
2010 6.36%
2011 6.49%
2012 6.52%
2013 6.43%
2014 6.44%
2015 6.45%
2016 6.52%
2017 6.54%
2018 6.43%
2019 6.23%
2020 4.65%
2021 4.45%
2022 5.46%
2023 5.6%
ChartData Download data

The data below can be saved or copied directly into Excel.

Source: Bureau of Economic Analysis, NIPA Tables 6.2.A – 6.2.C and 3.2.

Copy the code below to embed this chart on your website.

To be clear, we should be looking for efficiencies in government, and federal employment doesn’t need to keep up exactly with growth in the economy around it to ensure decent capacity to accomplish public goals. But if one is frustrated (rightly) by faltering state capacity, it seems odd to not at least look at these steep downward trajectories over time in key federal resources and wonder if that’s a big part of the problem.

More importantly, if one wants government to perform better, supporting a strategy that hacks away at its resources is not serious.