In a recent post at OTB, Steven Taylor, in a post on measuring the size of the federal government, includes a graph illustrating the number of federal civilian employees, remarking:
Now, this is just one metric of the size of government, but it is interesting to note that despite the narrative of uncontrolled growth, on this measure the government can be seen to be either relatively steady (in absolute terms) or actually smaller (in terms of employees as a function of the US population).
The obvious retort to this is that the figures used in the graph are derived from agency 113-A monthly reports. Here’s who that does not include:
The 113-A and 113-G reports must reflect Federal civilian workforce statistics and must not include private sector contractors.
In other words, the effective size of the federal government, as measured in workers, is no lower than the figure reported and might be two, three, or ten times as high when private sector contractors have been included. There’s no way to tell from the information supplied. Furthermore, I’ve tried to determine how many full-time, part-time, and temporary private sector contractors are employed by the federal government. The task is non-trivial and may even be impossible.
Additionally, does counting heads really determine the size of the federal government? If the you measure the size of government by percentage of GDP disbursed by government at all levels, that’s been increasing quite rapidly, particularly in recent years and right now it’s just around 40%. At the state and local level, much of that is spending on education. At the federal level, nearly all of the increase in spending can be explained by military spending and healthcare spending.
Rather than enter into the food fight that predictably erupted in the comments section of Steven’s post, I’ll ask a question. I think it’s obvious that zero percent of spending devoted to the government would not be conducive to optimal general welfare while 100% of spending disbursed by the government wouldn’t, either. If it were, people would have been breaking down doors to get in to the Soviet Union rather than the other way around.
That implies it’s a Goldilocks problem: 100% would be too hot, zero would be too cold, somewhere in the middle (not necessarily a single equilibrium point) is just right. I also think it’s reasonable to speculate that at some points between 0 and 100 spending more would be better while at others spending more would be worse.
Here’s my question: where are we now? Would spending more be productive or counter-productive?
This is actually a pretty complicated subject, touching on things like Ricardian equivalence (essentially, the observation that government spending either comes from taxes or future taxes), whether increasing nominal GDP will boost real GDP, and any number of other thorny subjects.
My view is that the web of government subsidies is now so thick it is impeding economic growth pretty substantially and that increasing the thickness of that web isn’t likely to help much. This is not to say that I think that government is bad—far from it. I think we need to make better choices.






