-
Website
http://willwilkinson.net/flybottle -
Original page
http://www.willwilkinson.net/flybottle/2004/12/15/what-is-big-government/ -
Subscribe
All Comments -
Community
-
-
Top Commenters
-
Popular Threads
All Comments
A "big" government wants and tries to influence its citizenry more than a "smaller" one. So you're right - it's possible to imagine a small government that churns through an enormous amount of revenue but only serves an ornamental purpose. In the real world, however, taxes are a good indicator of government ambition.
I think that you would find most measures of taxing and spending to be pretty well correlated with any other more refined characterization of government as "big" or "small."
Why? Because almost all other relevant measures of size and scope of the state are strongly positively correlated with spending. Plus it squares with intuition.
According to OECD numbers:
Total government spending as a percentage of GDP, 2002
Sweden -- 58.3
France -- 53.5
U.S. -- 35.7
That sure squares with my intuition about the relative size of government in those three states.
BTW, in your hypothetical I think you've miscategoried a 98 percent tax refund as spending. I'd argue that effectively the net tax burden in your example is 2 percent of income, with government spending at 1 percent.
It'd count as spending only if it *altered the purchasing allocation decisions in some way* -- say by totaling up collections, dividing by population, and issuing checks equal in amount regardless of initial tax paid. The idea of spending one's income as one chooses, incidentally, is also intimately related to our intuitions about liberty.
In fact, the best description I can think of for this situation is that this isn't a big or small government; it's an enormous nonprofit corporation. It lacks the one crucial element to make it a government: the legal monopoly on the use of force.
-Billy
Transfer payments are typically treated differently than government spending, since transfers can be spent as people wish while government spending presumably alters the consumption mix.
For this reason, effective tax rates are typically calculated net of transfers. In your case, the 98 percent tax refund is arguably a transfer payment, so you can make the case it shouldn't count as spending. However, that's debatable since what constitutes a tax is a surprisingly fuzzy area when you dig into it.
In any case, it's certainly possible to have negative effective tax rates -- as with Friedman's famous negative income tax -- for some taxpayers. But not for taxpayers as a whole.
Your example I'd call small government. 'Course it wouldn't stay that way...
Another interesting example would be a similar sort of arrangement, except instead of giving back the 98% to you, they keep it and spend it for. But they have an all-knowing computer which knows exactly what you want, and which has instructions to spend the money on exactly that (and no questions asked).
Of course that one begs the question of free will. But it still ends up with the same pattern of goods and services (and money) as a free market; thus I'd still call it small government.
A government which spends 98% of gdp is vast. A government which taxes but gives back all but 2% is tiny (but why would this ever happen?).
That said, the whole big vs. small government argument is a misframed one. A more relevent and less ideologically-charged frame that I prefer to use is efficient vs. inefficient government.
If any of those measures is used, I think spending is definitely better, although not adequate. In fact, it can only be seen as a proxy to have a rough instrument to assess government strategies and goals from a time series viewpoint.
Portuguese government spending, for instance, has registered a strong increase in government size which has revealed itself to be negative for economy and mainly because it was not sustainable. In my opinion this is the main question – sustainability of public spending and, of course, public debt; hence, the spending issue should not be neglected.
This notwithstanding, for the same amount of spending, an infinite number of policy choices can be made and this is probably the question here. So, in my opinion, size should be measured not only in spending terms but also in terms of public administrations (number of employees) relative to some indicators (representing quality) for each sector.
Well, this is already a big comment, but so many things are still to be said!
the amount that it spends on all programs/extracts in taxes is greater than the level that is considered reasonable by a majority of eligible voters. Or, from the demand end, the cost of the government (sum of all programs) that the majority supports. conversely with small.
also, there are institutional rights that are unique to the state. a government that is more intrusive than what a majority would support, then government is "too big." not intrusive enough (i could imagine such a thing), and it would be too small.
so i guess according to my logic, there can't be a "big" or a "small"--just a "too big" and "too small." It could also be too big and too small at the same time, since government is made up of a myriad of departments and programs, some of which at any one time will be bigger, and others smaller, than what a majority would support.
because of the sorites argument (mentioned above), big and small need an anchor to give meaning.
In general, can we say that the cost of government is the opportunity cost of the actions it takes, and that, if we understand cost that way, cost is the meaningful metric instead of size?
I generally agree with you, and would probably say that "bigness" can probably be measured in terms of market distortion of how consumption prefrences are effected.
On the other hand, intuitively it seems like disregarding citizen's electoral preferences for government is a bit like saying consumer prefrences don't matter for markets. Citizens are consumers of government, so what if they want a lot of it? Is there a rightness or wrongness judgement corresponding to this "big" or "small" argument?
As others have said, it’s a pretty good proxy, especially in the real world.
I think your example is intriguing but not realistic. Why would it ever happen? What would be the point? If the government only needs 1%, why take 100%?
What if, however, the government was one dictator with a handful of devoted followers and he demanded 1% of the nation’s GDP or he would nuke us all into oblivion? (I’m not saying this is a realistic scenario either, but probably more probable than the Department of Fireworks scenario.)
I’m in agreement with Jason Ligon I think. The size of government is inversely proportional to an individual’s ability to act without the threat of coercive force.
As others have said, it’s a pretty good proxy, especially in the real world.
I think your example is intriguing but not realistic. Why would it ever happen? What would be the point? If the government only needs 1%, why take 100%?
What if, however, the government was one dictator with a handful of devoted followers and he demanded 1% of the nation’s GDP or he would nuke us all into oblivion? (I’m not saying this is a realistic scenario either, but probably more probable than the Department of Fireworks scenario.
I’m in agreement with Jason Ligon I think. The size of government is inversely proportional to an individual’s ability to act without the threat of coercive force.
However, the "big government" concept used in American politics also refers to a couple of other things not captured in the thought experiment. First, there is the pervasiveness of government regulation, which to many people is measure of restriction on liberty. Even if the government spent nothing on the program, people would still object to the government telling them they have to wear a seatbelt, can't fly an ultralight without a license, and many other things. Increasing pervasiveness of regulation is, of itself, a bad thing to many people's minds.
A second kind of "bigness" of government is the magnitude of the distortionary effect of the taxes and regulation. If the government is collecting the 100% tax you mentioned via an income tax, revenues will quickly drop to zero, since nobody will work, and the size of the government, measured in terms of spending, is zero. The size measured as the inefficiency caused by government, might well be 100% of what people would have earned, had the tax rate been zero.
agree'd that intrusive regulations are a part of the size of government. However, any effective regulation relies on spending to enforce it, whether through a new oversight body or the expanding of powers for existing ones. Therefore while size is not a definitive indication of intrusiveness there is a rough correlation between the two.
Actually you would not get the Laffer curve effect in this example. Since I know I will end up with 98% of what I earn, I have a huge incentive to work. In a way, this may help highlight just how weird the example is. But there may be a useful practical angle. Suppose I allowed for charitable donations to not just be tax deductions, but also REPLACE my taxes. That way, when I work overtime I can say I am actually working for my favorite charity. In that setting you would again get a much weaker disincentive effect from taxation. Of course, the reason is that I have reclaimed a big measure of control over my earnings (I can't consume them, but can direct them to my favored charity, which is a great deal more than I can say for tax dollars).
Granted, they likely are correlated, but I suspect the correlation is a weak one. Spending on the bureaucracy doesn't vary a whole lot as administrations come and go. There is a *lot* of inertia in the size of the staff. However, the rate at which the bureaucracy produces regulations varies a great deal. This I've seen first hand from the inside -- anecdotal evidence, but evidence nonetheless. It's an empirical question that would be interesting to pursue.
On my point, I think a good example would be in policing. Policing a law which is generally accepted (and therefore, I'm going to say, less intrusive) is much less costly than policing one which is controversial (prohibition being the obvious example). The more individuals feel that breaking the law is a matter of personal liberty the harder it needs to be enforced to make it effective, and therefore the more costly.
I can't immediately think of an example of a regulation which would be seriously illibertarian but not cost very much to enforce.
On a different note, I notice I read the original post incorrectly. I thought we were talking about a government that collects everyone's income, and gives 98% of that aggregate income back in equal shares to everyone. I see now that Will Wilkinson meant you get 98% of your *own* income back. In the former case, nobody works (as I mentioned in an earlier comment), in the latter case, as gerry g pointed out, everyone works, though they work a bit less than in the absence of the 2% tax.
So the intrusiveness of the state is tied to its finances.
But making the hanged man pay for the rope is merely adding insult to injury.
I'm not so worried about the government getting a good deal on its spy cams.
I just don't want any *!@?#@$%!&* government spy cams in my house, knowh'I'msain?
:D