Real per capita growth in government revenue over the preceding ten-year period

From Karl Smith, here is a discussion.


I'm sorry but that discussion seems very stupid.

We learn that per capital government revenue stopped growing in 2001 because of Bush Tax cuts. We don't see any evidence that this has or will have an effect of decreasing government spending. It's really hard to "bankrupt" government when it's government that creates the money in the first place.

And why does the author say it might not work? Because of growing number of minorities and of course we all know all these minorities are welfare queens. Geesh.

It won't work because government is necessary to prosperity. At least no one has figured out how to maintain a prosperous society without it. In fact it seems government has to get increasingly big to maintain increasingly prosperous societies; at least that's the way it has always worked out.

And if in the future a growing fraction of minorities rely on welfare, it will be because all our cuts today have denied them the opportunities to pursue productive careers. (I concede there are segments of society where the legacies of poverty and oppression seem to doom future generations along with the present, but I don't think these are the segments making up the bulk of the growth in minority populations.)

Is there a chance that opportunities to pursue productive careers could come about with a cut back public sector? I think I can see the ideal you are referencing here, but I can't help but seeing lots of opportunities rumbling about here & there that seem pretty indifferent to government activities, at best. It seems like opportunities in the Asian success stories have come about as a result of public cut backs. If we are talking about education, it only seems a shame to me that it is viewed as a public institution. It is not a public good, and does not have a positive history as a service to the poor and oppressed.

Your thesis is the corrupt, lazy, incompetent people work for government, but by laying them off the will start innovative small businesses that create tens of millions of good paying jobs because corrupt, lazy, incompetent people paid compensation tens times that of government workers will be successful?

If you see Asian government cutbacks in Asia, we need more of those cutbacks in the US: bring on the high speed rail projects under government mandate, the massive road construction projects, the massive energy investments in wind, solar, government directed new nuclear reactor designs (we have already built the dams thanks to Hoover's planning).

I do wonder what nation you live in; my parents and their peers certainly were lifted out of poverty by education, and as a first half boomer greatly benefited by public education, and I look at the computer I type this on and think of Bill Gates learning to program in public school on a computer terminal provided by a government program to introduce kids to computers. Why don't you consider the fruits of that public education to be the vast public good of greater computing power than Bill Gates had over a telephone line in your pocket? From about 1969 I expected to have my own personal computer - I was late buying one because my employer paid me to help make them. While JFK was the iconic inspiration, Eisenhower had a greater impact on my public education that got me into science and technology, and specifically into computers. In 1965 in 9th grade I had access to a better public school electronics and chemistry lab than you will find in most colleges that teach science and technology today. Eisenhower knew first hand that science and technology won the war. And the wars in Iraq and Afghanistan have depended on science and technology applied AFTER the wars began, with nearly all the huge investments of the Reagan era to be totally useless and thus wasted.

How do you work it that a big government is necessary for a more prosperous society? Why not the other way around - a prosperous society can afford a bigger government?

And do you really believe that minorities need far more government support nowadays to make it than say a century ago? Or 50 years ago?

I make more now, so now I can go out drinking every night. Just because you can afford it do not make it a good idea.

The use of the term "bankrupt" was inept, it's true. However, policy is not driven by reality. It is driven by perception and interest. If the perception of bankruptcy risk can be fostered, then the argument offered at Modeled Behavior makes sense. Watch the debate over the deficit and it becomes pretty clear that bankruptcy is central to the austerity camp's position - even if they understand that bankruptcy is not really a practical concern.

In fact it seems government has to get increasingly big to maintain increasingly prosperous societies; at least that’s the way it has always worked out.

You might be right but "at least that’s the way it has always worked out" is rather weak evidence.

So, provide one counter example of a prosperous society at any time in history which accomplished its prosperity without having big government?

Transportation seems to be a key factor in prosperous societies. The US experienced jumps in prosperity after big government policies to improve transportation were implemented: the road building in New England, the canals, the Cumberland Road, the rail roads with the game changing transcontinental railroad, the US highways, the double down of Eisenhower's Interstates, the RFD and Parcel Post which drove rural road improvement for cars/trucks, the airmail subsidies that specifically promoted regular passenger service, airport construction, government subsidies of aircraft industries through massive government purchases and pilot training, and the construction and maintenance of an inland barge way.

Afghanistan is so transportation poor that it is easier to ship grain from Kansas to Pakistan. The big improvements in Afghan transportation have come from big governments outside Afghanistan in the past decade. Afghanistan is a nation with rich natural resources greater than Japan or Hong Kong, both the product of big government.

"So, provide one counter example of a prosperous society at any time in history which accomplished its prosperity without having big government?"

Please google: cum hoc ergo propter hoc

The cum hoc is such an easy accusation to throw at any assertion and is getting quite common.

I think at least some onus must fall on the other side: do you have any reason to believe this was NOT a causal relationship? Correlation, although not definitive, is an excellent starting point to establish causality.


That's fine, but using correlation as a starting point should imply that some effort will be made to refute well known rationale for the establishment of the cum hoc fallacy, such as the casuality could run in the opposite direction, both variable could be similarly influenced by a third, or the correlation is simply spurious. Make an effort to refute these possibilities and I will consider the argument.

P.S. This is why econometrics involves more than looking at trends on graphs.

Brian, when a bunch of people get food poisoning after a church social, and 43 people got it who said they didn't eat the potato salad, and 37 people got it who said they didn't eat the fried chicken, and 51 people got it who said they didn't eat the lemon mqeringue pie, but only 3 people got it who said they didn't eat the deviled eggs, then the epidemiologists are going to say it was something about the deviled eggs.

You can say that correlation doesn't absolutely prove causation, and that's fine. But you were asked to provide one single example of a prosperous society which didn't have big government.

If you can't find any examples, then you've got deviled eggs.

Stupid? Why in the world are you talking about welfare? That's not what the government spends money on. You might as well talk about the problems caused by excessive Arts Funding. Military. Debt Payments. Social Security. Those are the only real budget line items.

It always annoys me when someone takes a perfectly good time series (first graph in the Smith post) and then re-plots it as a derivative (growth on the y axis) over arbitrary time period granularity (10 years in this case). First, it adds nothing to the discussion and insults the readers intelligence: was it not obvious from the original time series that the trend in federal tax reciepts was downward starting from 2000? Did we really need Karl Smith to compute the "growth" for us to figure that out?

Secondly, the act of taking derivative always will exacerbate any of the noisiness in the data. Do we need people to try and produce explanations for what is very well an artifact of the differentiation process?

Also, it sometimes allows unscrupulous authors to present points which aren't really "real". Karl Smith mentions that for some time periods after the year 2000, the growth actually went "negative". Now at first glance there seems something fundamentally interesting when a metric historically positive goes negative, and rightly so in most cases. It catches the eye. However, note that here the growth is computed over arbitrary 10 year periods; by tweaking this period one can give the plot interesting shapes with varying transitions to the negative. If he had used 2 or 3 year average growths instead then it has been "negative" multiple times over the last 50 years. It just allows one to make the data more sensational than it really is.

Totally agree. Governments are necessary for growth specially when countries are so vast like the USA. The GOP needs to stop messing around and start taking out their feet from that brakes that are stopping the ending of bush era's low taxes for the rich. Trickle Down economics don't work except to the eye of some fundamentalist right wing as Cato Institute and others.

Ye, i think that its nothing strange, its normal for time that we live in.

Did anyone else notice that significant falls in per capita tax receipts are uncannily good at predicting recessions? As leading indicators go, it seems you can take this one to the bank...

As far as I can see the two best periods of growth coincided with the internet bubble followed by the housing bubble. This from the first chart - the second doesn't help me much.

Fans of big government thus need another bubble (the Obubble? (Obame + Bubble)).

Not being a Big Gov fan I think that a period of no growth in government spending after years of heady expansion is to be welcomed.

Show the expenses line too

This looks like regression toward the mean.

After the internet bubble, yes- that was reversion to the mean. Just eyeballing it, it appears bubble # 2, the real estate bubble, pushed revenues back above the long term trend. Some of the decline since then is surely reversion once again, but we do appear to have overshot on the down side, though the recent uptrend might be reverting to the mean once again. However, since I believe there is a point of debt saturation, I would fully expect the long term trend to start to shallow and eventually, maybe, even start declining.

This chart is difficult to interpret but it looks like Jimmy Carter was successful in driving down revenue growth. The drops are really caused by revenue losses during recessions.

I would love to plot government deficits on top of that line

I wonder if this type of analysis is an example of sloppy scholarship's slipping slope. In his Mechanisms of Slippery Slope, E. Volokh describes six types of slippage and there is one --attitude altering-- that apparently fits well with most economist-bloggers: they are no longer shy to indulge in a race to the bottom (something that they'd have never done in academy --ah, Viva la Libertad!).

Both charts are poorly constructed: difficult to understand and interpret.

Boy I'll say. I thought it was only one chart!

They should have used Ventiles

Looks like the government needs to find another bubble to extract revenue from, and quick. My guess is that they will need to tax their bonds much more highly.

The graph ignores state and local government revenues. If you add up revenues from all governments, there is essentially nothing to see (no matter what you may think of Karl Smith's choices for his analysis).

I've produced the equivalent charts for all levels of government here at


This is a "growth" or rate of change aspect.

It is "per capita" (re: taxation revenues)

Aproximately 40% of the capitated do not pay into one of the major forms of tax revenues.

What was the relative rate of increase in the per capita population of that 40% compared to the balance? Was it constant?

Sorry hit too soon:

Remember that with the famous tax-cuts, the swath of thsoe exempted was widened.

Is that why year 2000 looks like the end of a steady trend?

Change the chart title to "Stock Market Returns Over the Past 10 Years." Same, exact, friggin, frackin, fuggin chart.

I don't think the data show what people think they do. Looking at the first chart (not the second, which is nonsense) all we see is that revenues trend upward except during recessions, when they fall. After each recession the trend continues upward from the lower post-recession level until another recession hits. If there is sufficient time between recessions the level hits a new high. If not, it doesn't. Big deal!

Note that there are many times during the course of this time series that you could have cut off the data and reached the same conclusion that Karl Smith reaches - that per capita revenues were headed down. But they didn't head down, and there is no reason to believe that they are headed down now except for the impact of the lousy economy.

The really interesting question is how someone could look at this data and conclude that it supports an assertion that it - quite clearly - doesn't support.

What I found most interesting was that, in the chart of "real federal income per capita", for each of the recessions after 1978, the graph starts to trend down a few months before the official start of a recession. So it is a leading indicator of a recession.

This shows which they last very much lengthier and thus saving you income which could otherwise are actually utilized to purchase new ones.sdsdg

Comments for this post are closed