Feed on

Today we begin a periodic series that will attempt to capture, via the numbers, exactly what has happened to the size and scope of government since the Great Society. People of all stripes have lovely narratives about “climates” of regulation or deregulation, point to increases or decreases in tax rates, point to the composition of Congress, or what leading OpEds are saying, or simply to prevailing attitudes. None of that resonates with me. Intentions and results are not the same thing. Let’s take some time to look at some “results.” We’ll begin our examination with perhaps the single best metric of the size and influence of government – how much it spends.

Tax collections and tax rates mean nothing. The true burden of government is captured by the resources it consumes. If it spends nothing every year, and merely collects taxes and gives them right back to us, the tax collections themselves may cause distortions and may themselves be costly to collect, but the burden of government is extremely small. On the other hand if governments spend trillions of dollars every year, but never collect a dime in taxes to do it, those are substantial costs, as it takes real resources to do this today, and will require payment in some form or another from taxpayers in the future.

I’d also add the caveat that “big” government in the sense I am going to depict below is NOT a priori evidence of bad or intrusive government, though it certainly is a piece of data to consider. For the purposes of this series I want to show the data without much commentary.

The following chart captures how spending at all levels of government has increased since 1960. The data are all in constant dollar terms (as of October 2011 prices). I have also eliminated the expenditures of transfers at the local and state levels – a huge chunk of their nominal spending comes from federal to state, federal to local and state to local transfers – so we need to pull these out so as to not double count. A detailed analysis of the productivity of government expenditures would be required to make any deep observations about the data, and is the topic of a longer-term research project of mine (sort of an update of this paper).

Sources: US Census of Governments, US Census, Bureau of Labor Statistics, Office of Management and Budget, NASBO, Various Years

I’d only like to make three observations on the obviously growing and large level of government spending in the US economy.

  1. I take a Hayekian view of this. I think the level of government spending is very much an emergent process. No single person has the ability to say, “aha, I am the US Czar now, so I will set expenditures on an ever upward march!” I know the right and some libertarians are prone to this kind of an argument, and I can see its allure. But it is not very useful. The level of government spending is an emergent order that is the outcome of hundreds of millions of small decisions made over many years. Sure, “we” can possibly increase or reduce it with some pen strokes, but that is a simplified way of thinking about the issue. For example, since we are so darn rich, it might be the case that “we” demand more government spending as a matter of course. Or, since we are so darn rich it might be the case that “we” can tolerate a lot more government spending as a matter of course. I really have no idea what is the reason, but it is clear that the income elasticity of demand for government spending is not a small number.
  2. For those of you wanting me to jump on this data, you might reflect on the nature of public goods (this idea is not original to me). If the government is in the business of providing truly pure public goods, then you should not see real, per-person government expenditures increase very much as population and GDP increases. Why? Because a pure public good is something for which providing it to one person is no less costly than providing it to 10 people. Think of viewing a sunset, or a missile defense system. If the US needed to spend $1 trillion on national defense to protect a population of 200 million, it should not cost 50% more to protect a population of 300 million. Of course, this is not what we see in the data.
  3. Notice the dramatic crushing of the government during the soul-crushing conservative Reagan years and the dramatic expansion of government during the bleeding heart liberal years of Clinton.

3 Responses to “Government Since the Great Society, A New Series”

  1. blink says:

    This is enlightening and suggestive. But how much of this is accounted for by transfer programs like Social Security? While the distortionary and collection costs you mention apply, these expenditures should probably be excluded from this analysis for the reasons you give in the second paragraph.

    P.S. The JSTOR like appears broken.

  2. Rod says:

    While it might be true that government spending generally grows through thousands of individual decisions to spend, Obamism grew abnormally sharply as a result of the Obama administration’s few decisions. Proverbially, Bush drove the car into the ditch, and B.O. moved the Republicans to the back seat and drove the car out of the ditch himself.

    Also, we’ve had altogether too much foolish spending recently: subsidies for green energy, the saving of mostly public jobs and the general raising of the baseline, from which future budgets are derived.

    As for whether defense spending should rise proportionally with population, it seems to me that it doesn’t matter as much how many people one needs to defend as the how many belligerent countries or, in the case of Al Qaida, the belligerence of international terrorist organizations one needs to defend against. The main principle is that you always want to have more advanced weapons than your adversaries: crossbows vs. bows-and-arrows, Parrott Rifle Cannons vs. Napoleons, Abrams tanks vs. Saddam tanks. That’s why Iran gives me the Willies.

  3. Harry says:

    One way to look at government spending is that it is all overhead. Wintercow’s graph, beginning in 1960, shows that in constant dollars per person government expenditures have quadrupled. It is as if we had one household servant in 1960 to clean the house, prepare the meals, mind the children, and now we have a manservant/butler, gardener/gamekeeper, and chauffeur/bodyguard, all part-time, but on the books from the point of view of payroll taxes, workmen’s comp, liability insurance, and pheasants and ammunition.

    As Blink points out, we spend a big chunk of it to take care of the parents with Social Security and Medicare. We spend a whole lot more per person on public education. We also spend more on public parks with ballfields that have expensive backstops and special clay for the infield. Those public goods add up.

    I did notice the rise during the Reagan years, wintercow. One might be tempted to say that rising per capita government spending caused prosperity, and that Reagan was a spendthrift.

    Wintercow is just starting a big conversation on a big topic. Thanks for a great graph!

Leave a Reply