Tuesday, July 12, 2011

USAF AADs: Human Capital or Cheap Signal? or "Why I stopped worrying and learned to love Tauro University?"

In the first decade of the 21st century, the United States Air Force experienced a significant policy debate regarding officer education. The question at hand was why officers attain graduate level education or advanced academic degrees (AAD) and how should these effect promotion. On the one hand it is obvious why some officers need graduate level education. For example, those serving in research or test units need some specific education above and beyond their undergraduate training. The level at which they work is more specific than general. However, it is not completely obvious why the vast majority of Air Force officers, those serving as aircrew, personnel, finance, etc., need to obtain a level of education higher than required to assess into the officer corps. 
 
It is this second group of officers, the generalists, that are the source of contention and debate. Moreover, this debate led to conflicting policies from the most senior leadership and left the issue muddled and confused for junior and mid-level officers today. In this article I lay out the main points of each side, discuss their merits and then offer a third perspective. I argue that the Air Force has left behind the original intention and understanding of graduate education and instead uses AADs principally as a promotion screening and selection mechanism. 

In 2004, then CSAF General John P. Jumper wrote a Letter to Airmen[i] describing a significant change in promotion procedures and how the Air Force would treat education in general. In that letter he directed AFPC to mask officer education data on promotion boards through the rank of Colonel. Promotion boards would not see officer education information until being considered for Brigadier General. Jumper’s stated intention for masking education data was to prevent and stop the practice of company and field grade officers pursuing AADs for the sole purpose of increasing promotion probabilities, also known as “square-filling” or “checking the box”. Although Jumper acknowledged the value and importance of education in an officer’s career, he felt that the Air Force’s education culture had evolved into a mass-production of credentials of dubious value.

Nearly one year later, the succeeding CSAF and SECAF, General T. Michael Moseley and Secretary Michael W. Wynne, issued a Letter to Airmen[ii] reversing Jumper’s decision. Gen. Mosley lauded the importance of education in his letter and stated that the value for the Air Force was in having “intellectual throw-weight”. Gen Mosley announced that officer education data would be unmasked starting with the promotion boards in  2008. And so, because of a sweeping policy change, and then a rapid reversal, the Air Force held promotion boards between 2005-2007 that excluded any and all information about an officer’s education attainment.

How could these two men have claimed to grasp the importance of post-graduate education in a similar way yet employed policies that were absolutely opposed to each other? What did they see differently about the modern Air Force and officer advanced education? The economic theory behind education offers two distinct exclusive ideas to explain this- the theory of human capital and signaling.

The modern economic theory of human capital looks at workers in the labor force as a sum of acquired skills and knowledge[iii]. Some of our personal human capital is useful in any setting such as the ability to read, write and do simple math. Other dimensions of our human capital are specific and only useful in very narrow settings such as the ability to operate a fighter-jet in combat. We acquire specific and general human capital both through formal education and through experience. Applied to the Air Force we could say roughly that our specific human capital is acquired through formal training courses while our general human capital increases in educational and PME programs. For example, take a hypothetical A1C crew chief who went to a training course to learn to work on C-130s. His training course did not teach him how to be a finance troop or even how to work on and launch F-16s. The human capital he has in being able to fix C-130 aircraft is very narrow and specific. On the flip side, the education he acquired in Airmen Leadership School increases his general human capital in that it would serve him in any AFSC or in the civilian sector. 

For the rest of this essay I will simplify things and only talk about human capital as the composite of these two distinctions, general and specific. In reality, most sources of human capital are a mix of general and specific and do not divide neatly into one or the other. However it is helpful to keep both areas in mind when evaluating the education programs available to military members. We want to ask ourselves if a particular educational program increases a student’s general or specific human capital, or both. 

When Mosley emphasized the importance of education and justified his decision to reverse  Jumper he was speaking to the value of human capital acquired through the pursuit of advanced education. In his mind, the Air Force should be able to identify officer education levels during promotion boards because advanced education represents higher levels of human capital. As officers advance in rank, responsibilities require even greater abilities in communication, leadership, critical thinking, and knowledge of the Air Force organization and doctrine. From Gen. Mosley’s perspective, masking education data (both undergraduate and graduate) removed the promotion board’s ability to identify officers with high levels of human capital. 

What did Jumper see in the education process of an officer’s career that led him to order educational data masked on promotion boards? When he complained of acquiring master’s degrees in order to “square-fill”, Gen. Jumper believed that the path most officers take towards an AAD, namely distance-learning and off-duty education, is void of nearly any educational benefit and did not increase human capital. Jumper neither was, nor is, alone in his sentiments towards AADs acquired through distance-learning and off-duty education programs. I believe through personal experience and contact with hundreds of peers that any serious survey of Air Force officers would report the follow results: a) most AADs acquired at the ranks of Major and below grades are from off-duty education programs, b) most officers do not believe these types of AADs increase their human capital and c) most officers enroll in distance-learning and off-duty education for the sole purpose of increasing their promotion chances.   

In fact at a recent MAJCOM officer call, the Commander reported that at the just-concluded CORONA conference the Air Force’s top generals discussed, among other things, the constant complaint from mid-level officers about having to do master’s degrees off-duty in order to be promoted or be competitive for in-residence civilian and military graduate programs and likewise with PME in-correspondence courses.
This particular general officer did not attempt to defend the current practice by extolling the virtues of a well educated officer corps or discussing how these practices increase our levels of production and human capital. Without apology, he explained away and justified the current practice as a mechanism for sorting and selecting officers for promotion. 

What this general officer and Gen Jumper were referring to was the economic phenomenon called the signaling effect[iv]. In short, a signal is an indirect means of communication when one person wants to convey information about himself to another but is unable to do so directly. The Air Force promotion board wants to know the intelligence of a candidate, how much human capital do they have, and their capability for the next rank. However, the board does not have information such as IQ, AFOQT, GRE or SAT scores to help them understand the cognitive abilities of the officers in the pool.  Theoretically, a master’s degree would send information about a candidate’s level of human capital relative to his peers. 

For example, a hypothetical Harvard graduate’s diploma is a very powerful signal when she applies for a job. The hiring company knows well that prospective students at Harvard are screened heavily, had excellent grades in high school, astronomical College Board scores and a high percentage of applicants are rejected. As far as signal efficacy, an Ivy League diploma is very effective because it conveys a large amount of information and is very difficult to obtain. 

When a signal is not difficult to obtain then it loses its ability to convey information and causes confusion. This situation is called a “cheap signal”. Prior to the Iraq invasion, Saddam Hussein relentlessly signaled to the United States, through endless speeches and by kicking out WMD inspectors, that he had and was willing to use NBC-type weapons. Hussein engaged in relentless “cheap signaling” to confuse the international community and dissuade the US from invading. This signaling was cheap because nearly anyone can engage in this behavior (think North Korea and Iran) and some people will believe it. In the end, his cheap signaling did not work to deter the US but it did succeed in creating the illusion that Iraq had WMDs.

Even if no human capital production took place as Air Force officers toiled away to produce graduate level diplomas, there would still have some utility if AADs were difficult to obtain. If one had to be highly intelligent, insightful and a cut-above to complete a master’s degree at one of the on-base programs or through distance learning then the diploma would be a powerful signal of an officer’s level of human capital and abilities. This is hardly the case. In fact, the Government Accountability Office (GAO) released a report[v] in March 2011 critical of the DoD’s lack of oversight into the quality of education received by service-members for on base education programs. This GAO study incorporated data from all four services through extensive study of base educational centers.
DOD verifies whether a school is accredited; however, it does not gather some key information from accreditors when conducting its oversight activities, such as whether schools are in jeopardy of losing their accreditation. Accreditors can place schools on warning or probation status for issues such as providing inaccurate information to the public and poor institutional governance. Schools can experience various problems within the 3- to 10-year accreditation renewal period, and these problems can negatively affect students, including service members. Additionally, DOD does not require schools to have new programs and other changes approved by accrediting agencies in order to receive TA funds. Currently, students enrolled in unapproved programs or locations are ineligible to receive federal student aid from Education, but can receive TA funds.
In short, the DoD allows military members to use tuition assistance (TA) funds at institutions that have met the bare minimum of education standards and that may be experiencing other problems. This report states that it did not even begin to address distance-learning programs which made up 71% of courses taken in 2009. The evidence contained in the GAO report is not prima facie evidence that on-base and distance learning graduate programs offered to military members are void of human capital production but it should at least give us pause.

Even before the GAO report came out one could have looked down the list of off-duty education programs offered to military personnel, such as American Military University, Embry Riddle, University of Phoenix, or Troy University, to see that the education opportunities available to the majority of Air Force officers are not high quality. If one were to cross-reference on-base programs with the US News and World Report Graduate School rankings or any other reputable ranking system as I did, one would find these institutions absent from the listing. The fact that the US News and World Report does not even attempt to rate most of the graduate programs in which military members are enrolled speaks volumes about their reputation and quality. It is abundantly clear that the path to a master’s degree from these institutions is not a trial of intellect to military officers. These AADs are in fact cheap signals. Because they are not difficult to obtain nor are admissions particularly selective these degrees do not in any way distinguish cognitive abilities between those who possess this signal and those who do not. The only information conveyed by the on-base or distance learning AAD is that the officer was willing to sacrifice lots of personal time to do it. 

Further evidence of this can be found by looking at the promotion statistics published by AFPC, most prominently in the statistics published for Lieutenant Colonel and Colonel promotion boards (See Table 1). One can see that in 2005-2007 IPZ promotion rates for officers without an AAD shot up dramatically. These were the years that promotion boards could not see information about an officer’s education. For example, promotion rates to Lieutenant Colonel without a master’s degree went from around 15.7% to 48.6%. While it is true that more officers approached their promotion boards without having completed their master’s degree (from 7.6% average from 1996-2004 to a average of 16.2% from 2005-2007) this cannot explain the more than tripling of promotion percentages for non-AAD officers. One could challenge this assertion by saying that the AF must have been promoting more officers to Lieutenant Colonel but this was not so. Over the last ten years promotion rates to Lieutenant Colonel have been steady at 73-74%. If an AAD increased human capital then promotion rates should have remained unchanged because those with AADs, armed with more skills and more productive capability should have outperformed those without AADs at the same rate as before. But, they did not. Officers with AADs became undistinguishable from those without AADs.   
Looking at promotions to Colonel we see more evidence, albeit less powerful statistically. In the years 2000-2004, no officers were selected to be promoted to the rank of Colonel without an AAD. To be fair, very few officers, upon reaching the promotion board for Colonel, had not obtained their AAD. However in the time period 2005-2007, a few slipped past without AADs and were selected for promotion to Colonel. Once Mosley’s policy was enacted officer promotions regressed back to the trend and no officer since 2007 has been promoted to Colonel without an AAD. But those officers promoted to Colonel without an AAD must have had excellent performance records as only around 45% of IPZ Lieutenant Colonels are selected for promotion.   

What this tells us is that once education data was masked, the promotion board could not use AADs as a discriminator for selection to the next rank. Unable to use education information the selection boards ended up promoting people that would not have been selected in previous years. In 2005-2007, those promoted to Major, Lieutenant Colonel, and Colonel had better performance records than those not selected for promotion. Before and after this period not every officer selected for promotion had a better record than those not selected. If that statement were false then promotion rates between AAD and non-AAD officers should have remained unchanged regardless of the availability of education data.

In truth, a promotion board does not need education data to determine promotions because nearly all the information needed about a person’s performance, intellectual strength, and prospects for success at higher levels of responsibility can be found in training reports, evaluations, decorations, and records. However, that data does not help a promotion board distinguish levels of commitment to the Air Force. This is what a master’s degree from Trident University tells an officer promotion board and it is an effective signal in that regard. Instead of the AAD being a signal of higher levels of human capital it is a signal of loyalty to the Air Force.  

In trying to decide to whom should be given a valuable Definitely Promote (DP) on a Promotion Recommendation Form, or who should be selected for a special assignment or program or should be designated a school select, the Senior officer or selection board would like to know if the candidate is dedicated to the service, if they plan on serving at least twenty years, and if they are striving to be a senior leader. With a limited supply of DPs, developmental education slots or positions for promotion it is not unreasonable for selection boards and leadership to want to adopt commitment into their decision calculus.
Simply asking subordinates how committed they are to their career and to the Air Force would be useless. Being truthful about one’s career plan is not always the best strategy since any answer less than a desire to be a commander might hurt the subordinate’s stratification or leadership support for special programs and jobs. This is why non-traditional AADs are so efficient as signals. A unit commander does not need to ask his subordinates about their career intentions. The commander knows those officers who want to be promoted will complete their off-duty AADs and those that are less committed to being promoted will not.  
Similarly with AADs, our Air Force leadership now uses PME in-correspondence courses as signal mechanisms for commitment. Like most off-duty AADs, our PME in-correspondence courses are not difficult to complete but they do require a commitment of time. Thus they are cheap signals for knowledge and human capital. That is, they convey no information about an officer’s intellect relative to his peers. But, they are excellent signals for commitment because they require many hours of time to read the materials, write papers, and take the required exams. An officer who is looking to get out of the service at the end of their ADSC would have little reason to finish a PME in-correspondence course even if he or she would not admit this to their boss.   

The original intention of PME in-correspondence  programs was for officers who were unable to attend in-residence to obtain the knowledge necessary for the next level of leadership and remain competitive with their peers for promotion. It was never conceived as a pre-requisite for attendance in a full-time PME program. Yet, that is exactly what PME in-correspondence has become. How many times a day do our captains think to themselves “Why do I have to do SOS in-correspondence just so I can go and do it again in-residence?” Or likewise our majors who ask themselves or their commanders “Why do I have to do ACSC in-correspondence just so I can go and do an IDE program in-residence?” I have never heard a justification for this practice except that it helps with promotion boards and PME selection.

Another benefit, although less tangible, to full-time graduate studies is a peer-effect. Officers enrolled in civilian programs are exposed to ideas and people that are far removed from their normal sphere. When an officer is enrolled in an off-duty education program it is either in-correspondence, and therefore there is no peer interaction, or it is on-base or near the base. His class will probably consist of other military members and DoD civilians. This does not, ipso facto, mean that the class will lack a discourse of diverse thought and opinion. But, knowing that nearly all the students will bring a relatively similar background and set of experiences to the classroom the probability of a cross-pollination of ideas is low. 

In contrast, at a civilian institution our student-officers will most likely be the only person with any military experience. This is not only our chance to learn from civilian peers who have experience in industry, business, government, and academia but also to share and impart our military experiences with people who may not know anyone who has worn the uniform. One cannot understate the importance of exposing future civilian leaders to the culture of our defense institutions for which they will be charged with developing and implementing policy. Officers attending full-time graduate studies are not just students, they are ambassadors for a culture that has become increasingly alien to the rest of America, particularly the well-educated elite. 

Let me offer an alternative vision to a world where our Air Force officers spend too much of their time producing degrees that are considered nearly valueless in the labor market writ large or half-heartedly studying PME material for courses that most of them will repeat as full time students. This vision restores education to where it should be, a human capital producing venture that creates a good signal of ability and commitment. To do this all we must do is take a page from the play book of our sister service, the Army.
Because of historically low retention rates among junior officers the Army has not only been unable to fill positions that require senior captains and junior majors the Army lost the ability to retain its most talented officers[vi]. In 2005, ROTC and West Point implemented a program called the Officer Career Satisfaction Program[vii] (OCSP) to retain their officers. One of the incentives offered to cadets was a fully funded graduate school option that vested after cadets had completed their initial active duty service commitment (ADSC), an extra three years of service as the price for the option, and had served well and had attained the rank of captain. When the graduate school option vests at 7-8 years of service the officer can either a) leave the Army, b) remain in the Army but decline the opportunity to go to graduate school or c) attend a civilian graduate school program of her choice for two years to obtain a master’s degree with the Army continuing to pay her salary and benefits as well as tuition. In return for option c) the officer would “pay-back” with an ADSC of 6 years. This typically will take the officer to 15-16 years of active duty service and at that point she will almost certainly stay until twenty years.

The advantages of this systems are plentiful. First it very clearly identifies the commitment levels of young and mid-level officers. Those who are willing to contract for the graduate school option are obviously serious about their career in the Army. These officers are worth the investment of additional resources because they are not leaving anytime soon. Secondly, junior and mid-level officers do not have to allocate an inordinate amount of time away from their work and personal lives. They can focus on the mission, their soldiers and their families. Contracted officers know that at a certain time they will be freed up from their day-to-day duties and are guaranteed funding and time to be a student. Finally, the knowledge and abilities obtained during the two years of study will allow the Army to reap the increased human capital for its own benefit as well as the officer’s. Because the officers must serve six years after finishing grad school the Army is guaranteed to employ a smarter and more educated officer in positions of higher authority. Because officers exercising the graduate school option are not limited to only on-base or distance learning programs they can apply for and complete degrees at world class universities like Stanford, Johns Hopkins, or the University of Michigan. 

Upon implementation of this program the Army discovered that among its cadets and officers there existed a nearly insatiable demand for incentives like the graduate school option. Cadets were very willing to make a commitment to the Army above and beyond their initial ADSC in exchange for the Army making a commitment to them. Obviously the Air Force is not the Army and we have unique circumstances that would make it imprudent to simply mimic what the Army has done. Our Air Force senior leaders might look askance and say that this may work for the Army but there is no conceivable way to allow our officers a two year sabbatical for graduate studies. While this type of program would require a lot of personnel flexibility and career juggling it is impossible to dismiss this idea outright unless the Air Force is only paying lip service to the importance of education. When we consider that the US Army’s soldiers have taken the lion’s share of sacrifice and pain during our wars in Iraq, Afghanistan and the Global War on Terror this argument does not hold water. In the midst of massive junior officer shortages, if the Army can commit to its officer’s education and extract a likewise commitment from those them, then so can the Air Force. 

For reasons that probably began in the 1990s during the military drawdown, the Air Force lost its way regarding the value and importance of graduate level education for its airmen officers. Instead of AADs being something of value that increased the skills and knowledge and signaled higher levels of human capital the process devolved into a test of loyalty or a sign of commitment to an Air Force career. Likewise with PME in-correspondence courses. There is of course nothing wrong with the Air Force wanting to know the commitment levels of its officers before determining promotions, in-residence PME positions and how to fill important jobs. However, in allowing our advanced education and PME in-correspondence process to become a race to the bottom, the ability to discern commitment levels has come with a huge cost to the Air Force and its officers. While Gen Jumper may have enacted an extreme policy by masking all education data on promotion boards his instincts were correct. Thankfully we do not need to return to measures like these to break the cycle. Adopting measures like the Army’s OCSP, would allow the Air Force to invest in human capital and enjoy a much larger return while at the same time be able to receive a strong signal of commitment. We would then truly be able to call our officer corps “well-educated” and have real “intellectual throw-weight” available to fight the wars to come.  



[i] Jumper, John P, “Force Development: Changing the Education Mindset”, 2 February 2005

[ii] Wynne, Michael W. and Moseley, T. Michael, “Letter to Airmen: Advanced education key to global mission”, April 2006

[iii] For an understanding of the modern theory of human capital see Gary Becker, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. 2ed. New York: Columbia University Press for NBER, 1975.

[iv] See the following two papers for an understanding of the development of signaling and its role in economics.
Michael Spence (1973). "Job Market Signaling". Quarterly Journal of Economics (The Quarterly Journal of Economics, Vol. 87, No. 3) 87 (3): 355–374.
Michael Spence (2002). "Signaling in Retrospect and the Informational Structure of Markets". American Economic Review 92 (3): 434–459.

[v] GAO 11-300 DOD Education Benefits: Increased Oversight of Tuition Assistance Program is Needed, March 2011

[vi] For a good background paper on this subject, especially on the Army’s shift to a human capital and market based thinking about screening, training, and retaining talent see: Wardynski, Lyle and Colarusso, (2010), Towards a U.S. Army Officer Corps Strategy for Success: Retaining Talent, http://www.strategicstudiesinstitute.army.mil/pubs/display.cfm?pubID=965

[vii] http://www.career-satisfaction.army.mil/

3 comments:

  1. I've read your paper in A&S power journal and i'm trying to do some follow up institutions work to further show the signaling nature of this process but have run into a problem. How do you know data masking occured from 05-07? Other papers claim data masking occured from 2003 onward...i have been unable to verify (through AFPC, AFHRA, or other seemingly relevant orgs) when the data masking actually occured. By what means (other than the chiefs letters/messages can you validate the 05-07 claim?

    ReplyDelete
  2. Hi, I bumped into the same problems. There is no source documentation that I could find either to verify that the 05-07 boards were masked. I took General(s) Jumper and Mosley at their words. AFPC was less than helpful.

    The other piece of evidence is that something clearly happened in the 2005-2007 boards because the promotion results are far off trend with respect to AADs and in 2008 they return to trend. This of course could be completely unrelated to the masking of data but I took a small leap of faith and consider the change in promotion rates to be the result of data masking. It is entirely possible that I was wrong but I cannot find another plausible explanation.

    I'd like to read any other papers you found regarding data masking or signalling with AADs. Please send them if able.

    ReplyDelete
    Replies
    1. will do. i have also found AFPC unhelpful (shocking I know) but am trying to go through HAF A1 to get an answer since my wife currently works there...will see. There are tons of NPS papers out there, and not just papers but master thesis' that all claim such education "increases productivity." the problem is they measure the effect on productivity by equating promotion with productivity for a LHS and then having other factors, including education on the RHS. but if it's signaling instead you'd still expect the same result. my point is to prove that it can't (at least purely) be a human capital story as so many are wont to claim.

      Delete