F-35 Math is Hard. Analysis is Harder

Apparently Too Hard for Bill Sweetman Anyway

Bill Sweetman takes exception with Loren Thompson’s ‘math’. Let’s take a look at the complaint for any validity, shall we?
(Note: I’m not a big fan of Thompson or any ‘policy’ type for that matter that delves into the technical issues – they tend to grossly oversimplify the irreducible, but Thompson appears to be on target this time)

Taking a gander at the key bits of Sweetman’s editorial we find:

As Thompson says, “these numbers can be verified easily by perusing the Pentagon’s Selective Acquisition Reports.” The latest SARs for the F/A-18 and F-35 can be found here and here.

So let’s look at the key claims.

"Even if we include the electronic defenses and targeting systems not usually subsumed in a Super Hornet price tag, the unit recurring flyaway cost of a single-seat F/A 18 is about $80 million in today’s dollars. The corresponding cost for an F-35C is $130 million.”

The URFC of the F-35C is about right. But in then-year dollars, the URFC of the Super Hornet over 2011-13 averages $60 million (page 18 of the Hornet SAR). So what are the "electronic defenses and targeting systems” that would raise that number by $20 million? Targeting pods run about $2 million, and the ALQ-214 jamming system has been under $1 million per aircraft historically. (The SAR is not very clear as to whether those are included in the URFC.) The new Block 4 version of the jammer is higher, but any identifiable mods to the Super Hornet are still a fraction of the $20 million that Thompson is adding. Today, the F-35C costs more than two Super Hornets.

Swing and a miss!

Bill took his figures off a page (Page 18) titled “Annual Funding TY$”. For this report, we can use those numbers although I always prefer to use base year values and adjust. Sweetman’s fatal error was in not reading and understanding the totality of what he was trying to quantify,

On Page 28 of the same report, we find one entry under called “Quantity variance resulting from a decrease of 13 FA-18E/F from 565 to 552.” This entry, combined with the 2014 'blank' space in the table columns he was looking at should have prompted Sweetman to look at the previous F-18E/F SAR for more info.

It turns out, the FY2011 F-18E/F SAR had an entry (pages 17-19) for 13 units in 2014. Depending on which data you choose to use in the 2011 SAR, and in one case how you adjust from $FY2000 base dollars, it works out that those 13 units would cost between $78.9M and ~$80M each.

• Page 17 values are 13 units for $1.026B (Then Year Dollars) = $78.92308M each.
• Page 19 values, 13 units for $61.07M (Base Year 2000 Dollars) + adjusted for inflation to 2012 dollars* = $79.4M each.

*Inflation adjuster only goes to 2012, 2013 data not calculated yet

$78.92308M or $79.4M?

Call it “about $80M”, just as Thompson asserts. So why the unit cost jump? Look at the SARs. From a glance it looks to be all about Quantity and FMS price support.

Like They Say on TV: But Wait, There's More!

Sweetman goes on (in more ways than one):

Next: “When 100 single-seat Super Hornets had been produced, the unit recurring flyaway cost—with all necessary electronics included—was about $110 million in today’s dollars, which is where F-35C is likely to stand at the 100th airplane.”

The 100th Super Hornet was delivered in the Fiscal 2001 batch. According to the SAR, the then-year URFC was $61 million. A standard Pentagon inflation calculator raises that to $77 million in 2012 - $33 million less than Thompson’s figure. The F-35 is 43 percent more expensive if it is indeed $110 million.

I call 'Caviling'

Without the quantification of all “the necessary electronics included”, or estimation method used Thompson’s figures aren’t really debatable.

Sweetman citing a ‘standard Pentagon inflation calculator isn’t very descriptive, but the 2001 Superhornet values he chooses to use comes close to adjusting the 2001 F-18E/F URF the same as if using the Historic Opportunity Cost inflation adjustment ($77.6M), which is a far better choice than most make, but it still does not invalidate Thompson’s claims if he uses another recognized inflation adjustment method, such as that for ‘Economy Cost’.

Economy Cost adjustment of the 2001 URF yields $95.5M per aircraft (without electronics) in 2012.

If the Economy Cost method was used by Thompson, $95.5M without the 'electronics' probably would be equal to about ~$100M with electronics,

If anything, the Economy Cost is a more inclusive measure of a project’s value:

Economy Cost of a project is measured using the relative share of the project as a percent of the output of the economy. This measure indicates opportunity cost in terms of the total output of the economy. The viewpoint is the importance of the item to society as a whole, and the measure is the most inclusive. This measure uses the share of GDP

In Closing

Sweetman appears to be just trying to pile-on with the last complaint. Overall, his editorial fails to ‘disprove’ or cast doubt on anything except some people’s grasp of economics and defense spending. Perhaps Sweetman’s well-known target fixation on the F-35 was his undoing this time around? No doubt the innumerate will still be impressed.

UPDATE 28Jan13 : at the 'Ares' site, after trying more than once, it was still impossible to post a substantive rebuttal to Sweetman's mischaracterization of this post in the comment thread so I posted it here.