Not Impressed: NNSA Dismantlement Rates aren’t as Great as They Sound

Dec 6, 2012
By Sarah Weiner
Finally, some good managerial news coming out of the NNSA.  After a year of security breaches and cost overruns, the NNSA announced Monday that it exceeded its FY12 warhead dismantlement goal by 12%.  Don Cook, NNSA Deputy Administrator for Defense Programs, touted the accomplishment, stating, “NNSA delivered on President Obama’s commitment to reduce the numbers of U.S. nuclear weapons declared excess to the stockpile and awaiting dismantlement. We exceeded our dismantlement goals for FY 2012 by a significant margin.” 
But there’s a catch.  The NNSA may have disclosed that it exceeded its goal, but it did not disclose its original goal.  Good job clearing the bar, but just how high was it to begin with?
Best estimates say not very high.  Dismantlement goals, established in a “production and planning directive,” are classified, but we have enough information to make a pretty good guestimate.  In May 2010, the Obama administration released a fact sheet on the stockpile, including dismantlement rates, in an effort to boost transparency.  The document shows the number of retired (non-deployed, non-stockpile) warheads that were dismantled from FY94 through FY09.  I’ve produced a graph of dismantlement rates below:
Data from US Department of Defense
The data show a precipitous decline in dismantlement rates from the mid-90s to the 2000s.  From a high of 1,393 in 1995, the number of warheads dismantled dropped to a low of just 144 in 2001.  In the 2000s, rates never fell below 100 and never exceeded 650, creating an average of about 330 warheads per year.  That’s a fairly realistic point of departure for estimating FY10 through FY 12.  (Note: The graph may give the impression that changing dismantlement rates correlate with changing White House occupants, but the years don’t quite line up.  President Clinton, serving from January 1993 to January 2001, oversaw the stockpiles highest and second-lowest dismantlement rates.  Similarly, the first year of President Obama’s tenure saw a decline in rates relative to the previous year.  The only year of data disclosed for the Obama administration shows it produced just 30 more dismantled warheads than the average across the Bush administration.)  Assuming that in 2010 President Obama did not undertake a dramatic departure from the average dismantlement rate of the 2000s, we could reasonably estimate yearly dismantlement numbers around 330. 
A second bit of information roughly confirms this guess.  Both the FY11 and FY12 Stockpile Stewardship and Management Plans stated the NNSA’s intention to complete dismantlement of all warheads retired before 2009 by the end of FY 2022.  The exact number of retired warheads waiting in the dismantlement queue is unknown, but external estimates have ranged from some number “more than” 4,000 to about 4,500.  (The former estimate was made in early 2009, the later in early 2010; so the 4,500 guess may include some warheads retired in 2009 that would not count against the NNSA’s plans for warheads retired pre-2009).  For argument’s sake, let’s assume the NNSA will need to dismantle about 4,400 pre-2009 warheads.  We know that 356 of those were dismantled in FY09, leaving about 4,045 warheads for fiscal years 2010-2022 (inclusive).  4,045 warheads across 13 years comes out to about 310 warheads per year.
So we have a fairly narrowly bounded guess for the goal the NNSA managed to exceed for FY12.  If NNSA was attempting to stay on pace to meet its 2022 deadline, then its goal was likely around 310.  If NNSA was pursuing the average dismantlement rate achieved during the previous decade, its target would have been around 330.  If NNSA was trying to match the dismantlement rate it achieved in President Obama’s first year, its objective was likely closer to 355.  We could predict with some confidence, acknowledging the perils of estimating classified data, that the NNSA’s goal for FY12 was not much lower than 300 and not much higher than 375.  Remembering the agency exceeded this goal by 12%, we can estimate it dismantled 336 to 420 warheads, exceeding its goal by 36 to 45.  While surpassing any goal is laudable, we should keep these numbers in historical context.  For an agency that has in the past dismantled almost 1,400 warheads in one year, getting through even 420 warheads in FY12 is not particularly impressive. 
Some may respond that these are unfair metrics.  We should not compare this-year dismantlement rates to previous-year dismantlement rates because this year’s dismantlement budget is likely different from previous years’.  While true, it sadly doesn’t make a difference.  The chart below shows three kinds of data for FY00 through FY12: NNSA’s budget for dismantlement, the number of warheads dismantled, and the average amount of money spent per dismantled warhead.  This last number is an average, obtained simply by dividing the fiscal year’s budget by the fiscal year’s dismantlement rate.  (I have thus far been unable to find pre-2000, pre-NNSA dismantlement budget numbers, but I’d happily accept suggestions).  I estimated dismantlement numbers for FY10, FY11, and FY12 by assuming a baseline goal of 330 warheads and adjusting “actual” numbers according to NNSA statements.  The NNSA announced it exceeded its dismantlement goal in 2010 (26%), 2011 (20%), and 2012 (12%); notably, 2012 was the least successful of these three years.   
Dismantlement data from the Department of Defense. Budget data from Department of Energy yearly budget justifications.
2000 to 2012 saw slight growth in both dismantlement numbers and dismantlement budgets but a schizophrenic “value for your dollar” metric.  At its most expensive, dismantlement cost about $260,000 per warhead in FY05; at its least expensive, it cost just $69,000 per warhead in FY02.  In FY12, the NNSA spent about $153,000 per dismantled warhead – not especially bad in historical terms, but not especially remarkable either.  Exceeding its goal by 12% did not translate into significant savings. 
This does not mean the NNSA has failed by any measure, even in a historical context.  The NNSA must use its personnel and facilities to meet many diverse objectives, of which warhead dismantlement is only a small part.  Lower goals for warhead dismantlement may simply reflect a reshuffling of priorities between dismantlement, research projects, life extension programs, and other stockpile stewardship activities.  Slower or more costly dismantlement may also reflect better, safer, or more secure procedures that necessarily require more resources.  But, granting all of these caveats, we still can’t be too impressed by the NNSA’s most recent announcement.  Achieving 112% of a pretty low goal is nothing to write home about.  
Sarah Weiner is a research intern for the Project on Nuclear Issues. The views expressed above are her own and do not necessarily reflect those of the Center for Strategic and International Studies or the Project on Nuclear Issues.

Not Impressed???


Suggest you keep your "not impressed" look to yourself until you understand the spectrum of work being done. Your research above, while intriguing, is superficial in a way I wouldn't expect from a CSIS interrn. Without sharing anything out of school, I can tell you the technical differences between the systems dismantled when they reached the highs you quote and the ones they would be working now are so significant that it is little surprise that they take longer on each system. When you factor in the incredible increases in safety, security and environmental restrictions that have been levied over the last several years, it's actually amazing the NNSA is able to complete any dismantlements at all.

Bottom line, your assessment is very wide of the mark and you owe the hard working, under-appreciated technicians in the NNSA community an apology for disparaging their diligent efforts to meet our national policy objectives.

You are right, the intricacy

You are right, the intricacy of systems has increased along with safety and security requirements, and I’m sure dismantlement is a complex and difficult task.  I in no way mean to criticize the work of NNSA technicians or the organizational priorities set by the NNSA.  And I appreciate your contribution to the conversation.

That said, announcing dismantlement rates in terms of percentages instead of actual numbers is a sleight of hand that, while perhaps important for security and secrecy, may lead the public to draw the wrong conclusions.  I do think it is important to break down the NNSA’s latest numbers in a historical and budgetary context.