# Number Gut Part II

Way back in 2004 I wrote about how the lack of an intuitive sense of scale prevented many people from viewing the Lancet Iraqi Mortality survey with skepticism. The same lack of sense of scale shows up in other areas such as in this article (via Megan McArdle) about ending subsidies to the oil industry instead of levying a windfall-profits tax.

The article argues:

However, one thing that I did notice when I was doing a little google-fu on the issue is that there appears to be approximately 20 to 50 billion dollars spent by the federal government per year on direct subsidies (as opposed to tax breaks) given to the oil industry each year.

I am all for yanking corporate subsidies but the author or his sources clearly pulled the \$20-\$50 billion number out thin air. It’s at least an order of magnitude off.

The entire discretionary budget runs only something like \$730 billion a year, so the numbers given would come to roughly 3%-6% of the entire discretionary budget! Even Washington’s notoriously bad accounting couldn’t miss checks that big every year. I think that (1) somebody moved a decimal point and that (2) “direct subsidy” is used in the loosest possible sense.

Why couldn’t the numbers be accurate? Several reasons. First, mathematically the wide range given is implausibly inaccurate. The range really boils down to \$35 billion plus or minus \$15 billion. That’s like saying, “I think I spent \$100 but it might have been \$55, \$145 or somewhere in between.” A sigma that big is just silly.

Second, the number implies an improbable behavior on the part of non-oil companies. How likely does it seem that the oil industry could get 5% of all discretionary spending without every other major industry getting just as big a slice? Why would other corporations support handouts to oil companies which the non-oil companies would have to pay for? (This is always the fatal flaw in conspiracy theories about oil companies. Oil companies may be huge but the aggregate size of all the companies that consume oil is many, many times larger.)

Most budgetary arguments revolve around this kind of nonsensical claim. Somebody fires off a number and others swallow it because they lack any sense of scale for the phenomenon under discussion. It would help if we could develop a standard for reporting government spending in some set ratios such as percentage of GNP. Otherwise, most discussions of government spending mean nothing.

## 13 thoughts on “Number Gut Part II”

1. assume you are right. question: why should the industry, now experience gereatest profits ever and we are paying the most money ever for the product–why should they be given ANY subsidies?

2. Fred Lapides,

I don’t thing that any company or any industry should ever be given a subsidy for any reason. Neither do I think they should get trade protection or tax breaks.

But all that is irrelevant to my point. The point is that a lot of people are making policy based on wildly inaccurate numbers.

3. You get no arguement on what you say!

4. I remember, early in the Iraq war, reading a blog post that said something like 10 million tons of munitions had been dropped on Iraq. I did a quick calculation, based on the Iraq Body Count website numbers (about 20,000 dead at the time IIRC), and computed that we’d have had to drop 500 tons (about the weight of a fully loaded 747) of explosives for every single dead Iraqi. Even ignoring the fact that most of the dead were NOT killed by bombs, the number is absurd — if you’re only killing one person per 500 tons of explosives, you must be bombing an awful lot of empty fields. I suspect it should have been “pounds”, not “tons”.

Anyway, I posted that simple analysis as a blog comment, and got flamed for it. The number had come from an official source, they said, so it MUST have been correct. There was no conception whatsoever that maybe someone in the government could have… *gasp* made a mistake… when they gave that number, or that someone in the media had misquoted it. In their eyes, the government is totally incompetent, goes to war over bad intel, isn’t smart enough to plant fake evidence to cover their arses, but NEVER makes mistakes when reporting numbers. So the number had to be taken as truth, and there was no sense in even thinking about whether it could have been absurd.

Being able to create rough estimates, and have some idea how valid your estimate is, is a key skill for understanding numbers and data. It’s amazing how many people completely lack the skill, and have no interest in trying to learn it.

5. I think I know wheere their number is coming from: royalty rebates.

They are not in fact counting budget outlays which are subsidies to the oil companies. They’re counting royalties not levied as subsidies to the oil companies.

Recall that the companies were’nt all that keen to do the deep drilling in the Gulf of Mexico? So a lower royalty rate was agreed to encourage them. That lower rate is now being called a subsidy.

I think.

6. Tim Worstall,

They are not in fact counting budget outlays which are subsidies to the oil companies. They’re counting royalties not levied as subsidies to the oil companies

I imagine they are calling something like royalties or taxes set at a lower rate than they would like a subsidy. However, a “subsidy” by definition means that the government pays money to an entity not that it simply takes less away in the form of taxes or fees than some would like. When people hear “subsidy” they believe it means that some of the money they payed in taxes is being used to cut the recipient a check.

The difference in policy between actively giving money and not taking enough are major. People should speak precisely.

7. Politics and journalism attract a lot of liars;after all if ones intentions are good,what is wrong with a few lies to help a good cause? The other thing is that a lot of these people are slobs-they don’t even care whether something is true or not. I wonder if they even think,or care about anything at all except grabbing perks.

One other thing-a number of websites repeated IQ evaluations from a fictitious website.Clinton’s IQ was 182 and W’s was 91-conveniently 50%.Aside from a feral ability to see what people want to hear,and then tell it to to them, has Bill ever shown any sign of ever having had a thought in his life? My dogs have had that ability to read and flatter people-that is what makes them very effective parasites. Politicians too.

8. The comment above ought also to note that both Bill and George got into Yale.
must both be smart. And Bill? may be this or that but he had a balanced budget and was respected throughout the world. As for all politicians: does that mean you do not vote, ever, or that you vote for liars?

9. Fred Lapides,

I think Renminbi’s point was that a disturbingly large number of people seemed to think it plausible that the President of the United States had a subnormal I.Q.

Whether Clinton was a successful president or not is a matter of opinion and not something measurable. However, people on the left did grant him the presumption of having a high intellect simply because he was a leftist. That is a systematic fault of leftist. They value their own intellects so highly that they come to believe that anyone who disagrees with them must be stupid.

10. That’s a systematic problem with any religious dogmatist. Leftists are just one specific type.

I do think it’s funny that anyone could think (1) Clinton has an IQ of over 180, or (2) Bush has an IQ of under 100. 180 is a bit over 5 standard deviations above the mean — which would make him one of the smartest ~200 people in the US. He might be fairly smart, but he’s not THAT smart. And there’s no way a below-average intelligence person would make it all the way to the top of our political system. Bush isn’t up in the 180’s for IQ, but he’s not below 120 either.

11. Obviously, Clinton isn’t that smart since he got caught partying with interns in one of the places funded by my stinking tax dollars. At least Spitzer had enough class to do it in a hotel. LotharBot brings it all home with his analysis of the faux IQ of Bubba. 180? I wonder what was Einstein’s IQ?

Shannon – I enjoy these number gut posts, hope you do more of them.

12. I think they are talking about this (CATO is as a libertarian source as you are likely to find). The figure is \$1.3 billion a year in tax breaks which are also available to other industries. Multiply that number by ten, twenty, or a hundred years to come up with the shock-and-awe number you need for a hit piece.

As I recall depreciation schedules for oil infrastructure is sometimes cited as a subsidy too. In general “subsidy” is frequently shorthand for “we could be taxing them more.”

13. Another “number gut” example came up last Friday on Best of the Web (WSJ):

“Food inflation hit an 18-year high in April, with grocery prices rising 1.5 percent for the month, the government said Wednesday. Prices rose in every aisle–dairy, breads, meats, beverages, fruits and vegetables. It means \$53 more a month to feed a family of four with a typical food budget.”

As James Taranto notes, this implies the average family of four spends about \$3,500 a month, or \$42,000 a year, on food. Seems someone miscalculated somewhere…