Our piddlingly tiny Defense Budget
I want to bring a different tack to our usually fun forays into Data Analysis: Instead of me railing about some statistical aberation, smoke coming out of my ears as I type, I want you, dear readers, to exercise those muscles.
For this, we go to a reliable source of bad data: The WSJ's OpEd page.
While editorials are by nature supposed to be opinion pieces, over time I have noted that this page has a tendency to disconnect from reality when it comes to mathematics, data, statistics. At times, the word dissembling comes to mind.
Feel free to agree or disagree with the premise of the commentary. But this is not a political exercise. Instead, I am looking for ways you find that the data has been spun or misinterpreted or otherwise omitted in support of this argument.
You may also feel free to argue the data is correct, or that the Op-Ed understates the facts. Either way, focus on the data.
Professors of Economics, Statistics, and Mathematics can feel free to make this a regular part of their curriculum. And don't limit yourself to the WSJ: The NYT, WP, LAT are all fair game. But I keep coming back to the WSJ OpEd for their long standing tendency to use economic or budgetary data, and then to distort it.
No matter: Lets see how strong your powers of observations are: Identify the errors in this Op Ed:
Our Small Defense Budget
October 20, 2006; Page A12 http://www.opinionjournal.com/editorial/feature.html?id=110009124
The full text is at the free Opinion Journal site.
>
NOTE: I do not want to make this about politics, but, rather, about how data is tortured for political ends. The most political this could get is perhaps how the WSJ Op-Ed page advocated for tax cuts during the war or things along that line which impact the budget -- but not the political arguments.
Make the focus on numbers, not partisanship.
>
Call this crowdsourcing: If anyone wants to reference charts, tables, etc, feel free to do so, and I will gather them all in a follow up post next week.
Friday, October 20, 2006 | 07:05 AM | Permalink
| Comments (76)
add to de.li.cious | 
digg this! | 
add to technorati | 
email this post
Comments
In the chart you have shown, I think that even it is literally true, it could be very misleading due to federal government debt levels limiting how much can be allocated for every area of government -- not just defense. We have run deficits almost every year since WWII. Therefore our debts have continuously grown. There has never, to my memory, been any serious attempt to pay down the national debt.
A chart showing the percentage applied to all federal government spending (including interest payments on the national debt) would give you a fuller picture. Another is what GDP was in today's dollars, vs then. Another is what the tax rate was then compared to now.
And even all that does not tell you the whole story, but it would be a good start.
Find the Truth. Do Justice.
Posted by: OkieLawyer | Oct 20, 2006 8:30:27 AM
WSJ does a lot better than NYT. I'm not sure on BBerg where you can find defese / budget data. Under ECOR, they don't really have anything about military spending. I'll assume the #'s themselves are correct and that he uses real #'s everywhere.
To be fair to the writer, I don't think he is necessarily trying to ignore things like the NPV of future US gov obligations. Ignoring those is par for the course. I mean, if we counted those, we might acutally have to do something about that, and that would involve touching a lot of "third rails".
Posted by: JoshK | Oct 20, 2006 8:59:02 AM
Did you just realize that the wsj opinion page is the worst offender of the" assault on facts"?
Posted by: marty | Oct 20, 2006 9:12:35 AM
By focusing on the share of GDP or federal outlays, he is ignoring the obvious fact that both GDP and federal spending have increased tremendoudly over the years. A smaller share of a bigger pie can still result in an enormous increase.
This argument is roughly akin to arguing that because I owned 10% of Google in its first (pre ipo and pre VC financing) year and I now only own 3% that I have somehow become impoverished.
Posted by: The Hube | Oct 20, 2006 9:22:09 AM
Marty touched on the crux of the problem. GDP has become so manupulated and overstated that these type of statistics are skewed. Defense spending as a percentage of REAL GDP is probably rather high.
Posted by: 4merRepublican | Oct 20, 2006 9:25:40 AM
I'm not sure GDP as a % of outlays is necessarily the best way to do it, but it is reasonable. I wouldn't use it as the only criteria for judging the right spending level, but it is fair. Especially so if you consider military spend to be like insurance, which you would expect to scale with earnings to some degree. eg, you up your life insurance as you make more money.
Posted by: JoshK | Oct 20, 2006 9:28:49 AM
Damn right it's a small defense budget.. still barely 47%of the whole world's military expenditures, and only about 8.6 times bigger than the UK's military spending. Oh wait, that's 2002 figures.. the situation may have improved since then.
And damnit, Saudi Arabia spends about 10% of its GDP on the military..there's a lot of catching up to do. Hell, in 2004, worldwide military spending was still an unacceptable 6% below its 1988 cold war all-time high. If we have a new all-time high for the Dow, how come military spending has gone nowhere in the last 8 years?
For those of you who know math, please immediately forget that 4% is more than a 33% uptick from the earlier 3% level. Heck, even if it had doubled, it would only be 6%, right?
Posted by: sneakypie | Oct 20, 2006 9:31:58 AM
I think this chart from the Brookings Institution (http://www.brook.edu/comm/policybriefs/pb95.htm) pretty much refutes the premise of a small defense budget. (can't paste an image here unfortunately)
Also, this article from Jane's provides some insight (it's from 2005, but relevant): US defence budget will equal ROW combined "within 12 months" -- http://www.janes.com/defence/news/jdi/jdi050504_1_n.shtml
And those are just starting points.
Posted by: Gary | Oct 20, 2006 9:33:37 AM
I don't think this thread is about how large US military spending should be compared to the rest of the world, or the EU, or whoever else. The claim Barry made is that they are misinterpreting the data.
The author is trying to make the point that military spending is lower in terms of national income than it has been in the past. The accusation is that this fact is somehow a distortion or a manipulation.
We can all sit here and talk about how the US spends more, or what the recent increase is. But, that's not the accusation above. You can discuss if this is well spent military money, but again, that's not to the point.
Posted by: JoshK | Oct 20, 2006 9:44:14 AM
I think "OpEd" stands for "opposite to editorial", the same way as "Tribeca" stands for "triangle below canal".
So it has nothing to do with "opinion". It is just another page.
Posted by: bob | Oct 20, 2006 9:47:44 AM
>Josh:
>You can discuss if this is well spent military money, but again, that's not to the point.
I think it is the point. Bottom line the journal is justifying an increase in defense spending. One has to ask weather the money spent is achieving our goals. Its like the education debate we always hear we have to spend more more more but few are looking to see what we are accomplishing with that spending. The US kids seemingly are falling behind the rest of the world in test scores. Same with health care spending, we spend almost 10x per cpaita as Costa Rica but have the same life expectancy.
Bottom line we can increase defense spending but will we achieve what we want to achieve?
Posted by: mrmanny | Oct 20, 2006 9:57:00 AM
I would be very interested to hear how this is a distortion of data.
As JoshK states the article was making the point that we spend a smaller portion of our income now on the military than we did in the past. Perhaps we should, perhaps we shouldn't. But I sense that the suggestion is that military spending should not be measured in this way. If military spending should not be measure as a percentage of the pie perhaps nothing should.
Perhaps we should keep federal spending at 1945 dollars adjusted for inflation and population growth. Does that sound like a good idea. That would leave out any increase due to productivity which would mean the federal budget would eventually tend towards zero percent of the GDP over the centuries. Sound reasonable? I don't think many would think so.
Perhaps the world is such today with technological advances and threats that we should be spending an even higher percentage of our pie on the military than we did in the past.
If the suggestion is not that military spending should be measured on a more straightline cost adjustment basis then I stand corrected, (however then I would like to know what the suggestion in fact is).
It doesn't seem to make much sense to me to not measure it as a percentage of GDP. So perhaps you could clarify Barry exactly how you think it should be measured.
Posted by: Q-Ball | Oct 20, 2006 10:10:45 AM
The obvious problem is that the Dept of Homeland Security is an obvious part of the nation's defense, but not included in the Dept of Defense budget. Likewise with the FBI, CIA, NSA, etc.
The WSJ writer wants cuts in 'non-essential' programs to spend more on security. But this point would be undercut by showing the % of the federal budget devoted to the military, especially if combined by Homeland Security, etc. (I'm assuming 'non-essentail' means welfare, education, arts, etc - which have not grown as fast as the military part.)
The writer is also trying to argue that we do not have "imperial overstretch," but neither of the measures really speak to that. That's a political discussion that needs support from different measures.
Posted by: paul | Oct 20, 2006 10:10:48 AM
"Bottom line we can increase defense spending but will we achieve what we want to achieve?"
Thats a very fair point. But it doesn't alter the fact that we spend a smaller portion of our income on military now than we used to. Again perhaps that is warranted. But I don't see how its a distortion of data. The whole point here as Barry spelled it out was not to argue with the opinion of the editorial (increase the military budget), but to show how they are distorting data to make their point.
I don't see it. Where is the data distortion.
Posted by: Q-Ball | Oct 20, 2006 10:13:29 AM
mrmanny,
>>One has to ask weather the money spent is achieving our goals.
I agree with what you are saying and that's a great discussion that can go on for days.
But, what I am pointing out is that Barry above is saying is that there is distortion in the presentation or aggregation of the data itself. That doesn't seem right to me.
Posted by: JoshK | Oct 20, 2006 10:13:41 AM
the first things to pop out are the same points that everyone above has already mentioned... ie. only listing percentages; no nominal measurements; no reference to gdp growth or changes in tax revenue, etc.
just quickly glancing at the article... i noticed one other thing that hasn't been mentioned.
"Nearly half (46%) of all tax dollars went to national security during Vietnam, and 28.1% as recently as 1987. But spending for the war on terror, including Iraq and Afghanistan, has only lifted defense to 19.8% of all federal spending today"
Now, without digging into the calculation (which might be another issue, i don't know), a the author is trying to prove that we are underfunding these recent military campaigns. However, this comparison implies that the only spending relevant to today's conflict is this years budget. what about defense spending in 1993? 1998? 2000? 3/4's of 2001? Just by looking at his chart, it appears to me that defense spending hovered around 16-20% in any one of these given years... All of which were relatively peaceful, yet we spent on defense: upgrading technology, training soldiers, building weapons (planes, ships, missiles, etc). Much of this investment is now being used.
In a corporate environment, a company may spend 1 billion dollars over 1-2 years to build a new plant. But this capital expenditure is spread across the next 15+ years of production. I see gov't military expendures in a similar way.
Without acknowleging spending in prior years, that is being leveraged in current conflicts, the author is neglected to give a true picture of what has been spent to fund the current war.
Posted by: Greg | Oct 20, 2006 10:18:47 AM
>>I think it is the point.
I think he means it's not the point of Barry's post.
My gut reaction to all data is not, 'has it be manipulated to some end' but rather 'is this data even correct.' I can usually draw my own conclusions from information but I can verify if it is valid information without understand the data collection methods. Generally if you understand the methods used your belief in the veracity of the data drops significantly. Good data is rare.
Posted by: thx1138 | Oct 20, 2006 10:20:35 AM
"Without acknowleging spending in prior years, that is being leveraged in current conflicts, the author is neglected to give a true picture of what has been spent to fund the current war."
This argument is non-sense.
Every war campaign has bee preceded by years or decades of peaceful military spending. Prior to WWII we had 20 years of military spending with no war. Yet spending spiked to 40% of GDP. I guess that 20 years of spending prior to WWII must have all been wasted since such a large buildup was needed.
Posted by: Q-Ball | Oct 20, 2006 10:25:54 AM
Think I'm nuts normally? I could ramble for hours on the US defense budget. I think Eisenhower had it right: Beware the military industrial complex.
Can someone tell me why our defense spending has to be more than the other 100+ countries on earth combined? This is choking investment in the private sector, giving Orwellian freakishness to our society, crowding out diplomacy in international disagreements, a joke when our children don't have health insurance, etc.
I'd like to see a non-Hawk win the next Prez elections, slash the military spending by 30% and get us the hell out of Iraq. So, on a very touchy subject, it's rather absurd that we feel we need to stay to teach the Iraqis how to set up a government.
So, let's say after the REVOLUTIONARY War, that's right, revolution, that the French, our co-liberators, had stayed and dictated to us the rules under which we could set up a government. How much would we have put up with that? Remember, we were a very fundamentalist country as well. For Christ's sake, we burned people at the stake. We did ok. Let the Iraqis figure out their own mess. When they come to their senses, we would be a great ally because we liberated them. And, if it's now three countries instead of one, who cares?
Posted by: BDG123 | Oct 20, 2006 10:26:59 AM
I believe the point the OpEd piece tries to make is that the US defense spending is low, especially given the threats the US is facing, and that it should be higher.
To support this view, it views the US in isolation (i.e., ignores other countries' defense budget size) and uses two arbitrarily chosen ratios.
BR said that (among other things) he was looking for ways in which data was omitted to support this argument.
One could say that these are appropriate metrics to choose, but e.g. the GDP doesn't really measure the GDP because of various distortions.
One could say these aren't the right metrics and we should use different ones e.g., median per capita income or net worth or whatever.
One could say that given the threats facing the US, UK, G8, NATO, etc., the US spends a disproportionate amount on defense relative to other countries.
Posted by: sneakypie | Oct 20, 2006 10:27:41 AM
Re: distortions....some random missives
During the Cold War, the USSR + et al was spending around 20%+ GDP on military spending (someone have exact figure?). So US military spending could be considered appropriate/symmetrical.
Consider that the insurgents in Iraq and Afghanistan are armed with AK-47s and IEDs (cost a few $1,000 per insurgent?), while the US spends $100 billion per year for the 150,000 soldiers in the Mid. East (not to mention the "routine" $300+ billion in military spending).
So in Iraq and Afghanistan, the US is literally outspending/outkilling the insurgents and a magnitude of 100+ to 1, yet the Iraq is still a mess. This is the definition of asymmetrical warfare.
Finally, re: North Korea + Iran/other conventional enemies, just like in the corporate sector, the US military has grown more productive due to technology, etc. So the US doesn't need to spending as much $$$$ to field the ability to kill each enemy.
The general downtrend of the WSJ graphs re: military spending v. GDP would look just like a graph of labor costs v. revenue for any Fortune 500 company like GE, GM.
So yes, WSJ's OpEd page loves to harp about increased productivity in the cubicles and factories, but neglects to consider the increased productivity of GI Joe.
Posted by: john john | Oct 20, 2006 10:34:23 AM
the data is correct. But let me put another spin on the analysis.
We are losing the war on terror in Iraq because this administration has been trying to fight it on the cheap as this data demonstrates. Moreover, the data overstates military spending because of the budget games the administration is playing -- for example not allowing the professionals in the pentagon to put the $40 billion in the budget to repalce and/or recondition all the equipment being worn out and/or destroyed in Iraq. this is the first war in US history that the army is weaker after 3 years of fighting then it was at the start of the war.
So why is this administration fight the war on the cheap?
Because if they were honest about what it would cost to win the war they would be forced to give up their tax cuts.
So we are losing the war because their tax cuts for the wealthy or more important to this administration then winning the war.
So someone show me one fact to demonstrate that i am wrong.
Posted by: spencer | Oct 20, 2006 10:42:40 AM
Again, everyone above has an opinion on:
1. is this enough spending
2. is it being spent the right way
3. are our foriegn policies the best
4. who's going to win the world series
But, the % of GDP argument presented by the author seems pretty tight, and that's all he's saying.
Someone above mentioned that the #'s above omit other agency spending, which is a valid point. But, that is also probably constant. I don't think that the FBI was ever in the military budget. And with the mess that congressional budgeting is, who the hell really knows what we spend on anything. But that's a different discussion, more like the ADP vs BLS argument. IMHO, go with ADP.
Compare that to a typical NYT op ed that assumes a horizontal demand curve for everything they want to tax. The integrity of the above article seems pretty good to me.
Posted by: JoshK | Oct 20, 2006 10:51:32 AM
One item that we should consider is that the last graph is of total federal outlay. On item that should be looked at is the military budget as a function of discretionary spending. The federal budget has grown tremendously in the area of mandated programs. See: http://www.thebudgetgraph.com/view.html
2007 Discretionary budget: $983B
2007 Military: $633B (64%)
2007 Non-Mil: $350B (36%)
Personally, I don't have a problem with it. I'd like to see reductions in the non-discretionary spending - but thats another topic.
Posted by: Paul | Oct 20, 2006 10:56:22 AM
If you like at the left tail of both graphs you will not that it is lower than present day percentage of expenditures. 1945 was when the Cold War began, at the Yalta conference. A relative low point, 1949, was the year the Soviets expoded their first nuclear bomb. (For a complete timeline, see here. The Cold war ended in 1991 with the collapse of the Soviet Union. Of the 66 years portrayed, 46 of them comprise the Cold War period. This means 70% of the graph represents an anomaly.
Posted by: M.Z. Forrest | Oct 20, 2006 11:13:13 AM
The comments to this entry are closed.
