May 5, 2019: The U.S. Air Force has become increasingly desperate to find a solution to its persistent and growing pilot shortage. Many studies have been commissioned, many experiments conducted and while there has been some progress the problem persists. The latest study sought a cost/benefit analysis of paying pilots more to remain in the air force versus recruiting and training new pilots. The air force has tried cash bonuses to persuade pilots to stay, but it has been pointed out that not all pilots cost the same to train and you still need new pilots to replace those that retire or are shifted to non-flying jobs. The key issue here was the need to hold onto more of the older and experienced pilots.

The study found that there were considerable differences in what it cost to train pilots depending on what type of aircraft they flew. Transport pilots cost the least. A C-17 (four engine jet transport) pilot cost $1.1 million to train, and these were the pilots who could most easily shift to being commercial pilots. Other types of transport pilots were somewhat more expensive train. The C-130J (four engine turboprop transport) cost more ($2.5 million) because these aircraft often operated from more primitive airbases and under combat conditions. Special purpose transports, like the RC-135 (an older four engine jet transport) used for electronic warfare and intelligence tasks cost $5.5 million for pilot training.

Pilots for combat jet aircraft are the most expensive to train. The least expensive are F-16 pilots ($5.6 million) with the most expensive being F-22 pilots ($10.9 million). Bomber pilots are more expensive with the older bombers (B-52) more expensive ($9.7 million) than more recent models like the B-1 ($7.3 million). Much of pilot training cost is the expense of taking their aircraft into the air, which is essential (as experience has shown) compared to alternatives like realistic simulators. It was also found that aircraft with better flight control software and more efficient cockpits cut training costs. The best example of this is the F-35 but the trend had been noted in other aircraft when cockpits and flight control software were upgraded.

The air force (and the military in general) have another problem and that is the long tradition of pay based on rank and time in service. This has long been recognized as archaic and inefficient. Commercial firms, even those providing military contractors for combat jobs overseas, pay according to what the market demands. There is a premium on skills regardless of age or rank. For example a former special forces NCO with lots of combat experience is going to get paid a lot more than a senior infantry officer with limited combat experience. The special forces NCO not only handles a more dangerous job but one requiring superior intelligence and training levels. That can also be seen in what professional athletes get paid versus what support staff and their managers earn. Tech industries also have to face the fact that special, and relatively rare, talent has to be paid the market rate if you want to have the people you need.

The air force, being the most technical of the military services, has long had problems with this issue and nowhere was that more obvious, and painful in the relatively new job of operating remotely controlled aircraft (UAVs). In September 2018 the U.S. Air Force released their study of using warrant officer ranks for career pilots as part of a solution to the growing pilot shortage. The air force study concluded that the army use of warrant officers for those who prefer to be career pilots (mainly of helicopters) would not work but having officer pilots specializing in flying who could advance up to the rank of colonel (O-6) might. Actually, there was no agreement on what might work but air force leadership agreed something had to be done. Many departing pilots just wanted to fly and current air force leaders do not like the idea of warrant officer pilots even though it works fine for the Army.

The study quoted departing pilots listing why they were leaving (lack of opportunity to fly) and why they would stay (belonging to a combat squadron full time and concentrating on flying). The air force continued to see the problem as one of money and pilots spending too much time away from their home base because of foreign deployments. The study missed the point. The departing pilots were not concerned about rank and were willing to accept the relatively rank-less commercial pilot jobs where pay depended on experience and the number of different skills (different types of aircraft) you were qualified to fly. Departing air force pilots wanted that sort of thing in the military.

The air force could easily adopt a similar system while retaining current ranks. You have pilots holding ranks O-1 (2nd lieutenant) to O-4 (major) but with some symbol to indicate these are flight officers and have them eligible for additional pay (as is not the case with current flat rate flight and combat pay) for skills. Since squadrons are commanded by O-4s or O-5s you would have no problems with these “flight command” officers having to deal with O-5s and O-6s under their command. Flight officers with leadership skills could leave the squadron duty for higher rank (and fly a lot less and lose some of the valuable skills flight officers got paid for). The air force has to be flexible in this area but the main thing is to follow the commercial pilot and airliner solution of paying for skills, not rank. This is largely what the army warrant officer system is. There are four warrant ranks roughly equal in pay to the commissioned officer ranks O-1 to O-4. On top of that, there is flight and other special pay and the guarantee that you will spend most of your time flying and belonging to an army aviation unit. That is what most of the departing pilots want to do; fly a lot and belong to a squadron that does it well because they get lots of practice.

Meanwhile, the air force has other “flight officer” problems which it has been forced to solve by returning to older (but long discarded) solutions. In mid-2017 the air force graduated its first female enlisted UAV pilot. What was notable about this was not that this sergeant was a woman but that she was one of the few sergeants the air force had reluctantly allowed to apply for UAV pilot school. The first three air force sergeants graduated from the 34 week enlisted UAV pilot course in May 2017. The enlisted course is longer than the six month one most officers take because the air force insists that all UAV operators must be able to pilot a manned aircraft. This is taken care of by sending most enlisted applicants (and some officers who are not pilots) to flight school where must qualify (as pilots) on a single engine propeller driven trainer aircraft. This is done in order to learn basic flight skills in the air.

The air force expects sergeants proceeding to UAV operator school to suffer the same attrition (fail) rate (15 percent) as officer trainees. This has been the experience of other services and nations that allow NCOs to pilot large UAVs. The air force NCOs (sergeants) get flight pay, like officers do, for piloting UAVs and that is now justified because operating a UAV, especially an armed one, is a lot more intense and stressful that piloting most manned military aircraft. Originally the air force agreed to give their UAV pilots (back then all pilots transferred from fighter, bomber or transport squadrons) flight pay while operating UAVs to maintain morale.

Initially, the sergeant pilots will not be operating armed UAVs but just the ones that carry out surveillance and reconnaissance. For the moment this means UAVS as large as the RQ-4 Global Hawk, which is the size of a small airliner. Thus all RQ-4 pilots must qualify to operate an aircraft in commercial airspace in all weather. To help with that the NCO pilots are drawn from the ranks of Air Force enlisted personnel who regularly fly as crew on aircraft (loadmasters, flight engineers or electronic equipment specialists) or those who have experience as a UAV sensor operator (nearly all of these are sergeants).

By 2020 the air force plans to have about a hundred sergeants piloting unarmed RQ-4 UAVs and that will comprise about half the 200 RQ-4 pilot jobs.  Beyond that, the air force admits it is open to allowing NCO UAV pilots to operate armed UAVs and the continued shortage of officer pilots for all aircraft (manned and unmanned) is moving the situation in that direction.

In 2015, after years of fierce resistance air force leadership finally relented and agreed to allow NCOs to be UAV operators. This decision was mainly driven by the fact they had tried everything else to obtain enough people to operate the growing UAV force and they were still unable to cope. Since 2001 the Air Force has failed to solve the problem mainly because there were not enough officers available who were capable or willing to be UAV operators. As a result, the UAV operators the air force had (currently about 1,300) were often being overworked and many were not willing to continue as UAV operators. Many were leaving the air force as well because of the heavy and intense workload encountered as a UAV operator. Only about a quarter of those UAV pilots operated unarmed UAVs (mostly RQ-4s). The air force retired the last of its Predator UAVs in 2018 and from then on all the armed UAVs will be MQ-9 reapers. The MQ-9s also do a lot of unarmed surveillance missions but pilot sensor operators of armed UAVs are under the most pressure.

The only way to ease the UAV pilot shortage was to obtain a lot more operators and that could only be done if NCOs were allowed to operate UAVs. The air force decision did not solve the problem because the air force insisted that enlisted UAV pilots could not operate armed UAVs. More and more air force UAVs are MQ-9 Reapers, which were built to carry weapons and replace combat aircraft in some cases while also performing the traditional UAV specialties of surveillance and reconnaissance. Yet the air force already acknowledges that the NCO pilots could probably handle it if only because many of the NCO pilots  have years of experience as a UAV sensor operator, literally working next to the pilot at the wall of flat-screen displays, keyboards and joysticks that serves as the cockpit of a large UAV.

The air force also tried using civilian contractors to handle the pilot shortage. By mid-2016 that was seen as counterproductive because the contractors were paid up to three times what officer operators made. The contractors were all former air force pilots with experience operating these UAVs and word quickly spread that if an officer was fed up with being stuck with another tour of duty as a UAV operator they could retire (if they had at least 20 years’ service) or resign and apply for a contractor job.

Until late 2015 the air force had hoped that higher cash bonuses would solve the problem but it didn’t. The air force offered $35,000 a year in bonuses for air force pilots who volunteered or were persuaded to serve as UAV operators. The job was simply something most air force pilots did not want to do. Many in Congress expressed reluctance about just throwing money at the problem, especially when there was a proven and cheaper solution for this; allow enlisted (sergeants) airmen to operate UAVs and make a career out of it. The air force had done this before, during World War II when it was still part of the army. But that was changed during World War II and the air force refuses to consider going back to what worked in the past, even though it works fine for the other services and some other countries.

One thing that prompted the air force to change its mind in 2015 was political pressure. Congress had asked the GAO (Government Accounting Office, the investigative arm of Congress) to look into the matter. This involved interviews with a representative sample of UAV operators and that documented how UAV operators were overworked and the air force was unable to get as many as it needed. This meant that existing crews had to work longer hours (60 or more a week). This caused a lot of stress. UAV operators each spend about 1,200 hours a year controlling UAVs in the air, versus 450 hours for army helicopter pilots and even less for air force pilots in the combat zone. The problem was that UAV operators (all of them pilots of manned aircraft) get none of the enjoyable aspects of flying (operating a jet, especially a fighter) and a lot more of the drudgery (constantly monitoring instruments and what is going on below). Operators did report that the air force had addressed a lot of the earlier problems; poor training, loss of career opportunities, especially promotions. The main problem was that few officer UAV operators wanted to be UAV operators. And those few who did choose it as a career were just as worn down by the grind like everyone else.

By 2015 it was obvious that UAV operators were a growing segment of the pilot population. In 2013 UAV operators were nearly nine percent of all air force pilots, triple the percentage in 2008. The air force was unable to get enough manned aircraft pilots to “volunteer” to do a three year tour as a UAV operator and could not train non-pilot officers fast enough to be career UAV operators. UAV operators were leaving the air force at three times the rate of pilots of manned aircraft. Worst of all, UAV operators were not shown the same respect as pilots who go into the air aboard their aircraft. Despite the GAO study the head of the air force continued to insist that all UAV operators be officer pilots. But a growing number of air force officers became aware that a lot of Air Force NCOs could handle this job and that NCOs in the other services had long been doing just that. So in late 2015 the air force brass finally relented and soon hundreds of air force NCOs were becoming UAV operators and by the 2020s thousands will.

Until 2015 only the army allowed enlisted troops to handle larger, and armed, UAVs. It was no secret that air force NCOs were eager for this kind of work and often are better at it than officers who are experienced pilots of manned aircraft. This is believed to be caused by the fact that operating a UAV is more like using a consumer-grade flight simulator game than flying an actual aircraft. The NCOs often have lots of experience with video games and get better the more they actually operate UAVs. This is especially true with the widely used army Raven micro UAV.

Most of the army operators use the small (2 kg/5 pound) Raven UAV, which provides platoons, companies, and vehicle convoys with aerial reconnaissance. The Raven training only lasts 80 hours but this tiny UAV was designed for ease of use. It takes about five times longer to train operators for larger UAVs like Shadow and Predator. The air force points out that the largest UAVs, like the Global Hawk, can cross oceans and require a high degree of training and skill. But it's much more dangerous to fly a Raven within rifle range of enemy troops and keep the little bird alive long enough to get the video feed needed to win the battle. Many of these army Raven operators are very, very good, mainly because they have hundreds of hours experience operating their UAVs while under fire. Few air force UAV drivers can claim this kind of experience.

Another argument in favor of NCO pilots was the fact that most special operations troops (Special Forces, SEALs and pararescue) personnel are NCOs. These troops undergo much more strenuous selection and training than pilots and are quite satisfied with being an “operator” all the time without any mandatory detours in the name of being “well rounded.” For a long time, the air force leadership was not swayed by this; for them, there is something undefinably wrong about putting NCOs in the pilot’s seat.

Commanders closer to the action believe NCOs could do the job and that would eliminate the shortages and morale problems with officers doing it. In large part, this is because of expectations. NCOs know what they are getting into and consider operating UAVs as a step up and a rational career choice. This is nothing new and the controversy over NCOs or officers being pilots began at the start of World War II when the army air force (there was no separate air force yet) and navy both had enlisted pilots. These men were NCOs ("flying sergeants" or "flying chiefs" in the navy) selected for their flying potential and trained to be pilots. Not leaders of pilots but professional pilots of fighters, bombers, and whatnot. Officers trained as pilots would also fly but in addition they would provide the leadership for the sergeant pilots in the air and on the ground. This worked quite well and many countries continued using NCO pilots throughout the war.

The “officer only” policy began in the United States during World War II as the Army Air Corps changed into the mighty AAF (army air force, 2.4 million troops and 80,000 aircraft at its peak). Back then the capable and persuasive AAF commander general Hap Arnold insisted that all pilots be officers. Actually, he wanted them all to be college graduates as well until it was pointed out that the pool of college graduates was too small to provide the 200,000 pilots the AAF eventually trained. But Arnold forced the issue on officers being pilots and the navy had to go along to remain competitive in recruiting. When the air force split off from the army in 1947, the army went back to the original concept of "flying sergeants" by making most pilots "Warrant Officers" (a sort of super NCO rank for experienced troops who are expected to spend all their time on their specialty, not being diverted into command or staff duties).

Many air force pilots envy the army "flying Warrants" because the Warrant Officers just fly. That's what most pilots want to do; fly a helicopter or aircraft, not a desk. But a commissioned officer must take many non-flying assignments in order to become a "well-rounded officer." Many air force pilots don't want to be well-rounded officers, they just want to fly. So a lot of them quit the air force and go work for an airline. But often they stay in the air force reserve and fly warplanes on weekends and get paid for it. This is considered an excellent arrangement for the many pilots who take this route and the air force has not been able to cope with this source of pilot attrition.

Unlike the traditional "pilot and crew" arrangement for aircraft, larger UAVs, like the Predator and Reaper, are operated by a team. Typically each of these UAVs is attended to by a pilot and one or two sensor operators (NCOs), who monitor what the cameras and other sensors are picking up. Because a Predator is often in the air for 24 hours at a time, and is often flying over an active battlefield and is looking real hard for specific stuff, the "crew" has to be changed every 4-6 hours to avoid fatigue. Moreover, each Predator unit might have several UAVs in the air at once. The pilots also operate the weapons for Predators carrying missiles. But most of the time Predators fly missions without using missiles. That is less the case with the larger Reapers, which are considered combat aircraft because of the large range of weapons they can carry (including smart bombs).

Another aspect of the UAV pilot shortages is the fact that software is replacing a lot of pilot functions and, eventually, taking the place of human pilots. Many larger UAVs already have the ability to take off, follow a predetermined course, carry out a mission, and then land, all by itself (or "autonomously"). One can make a case for officers being in charge here but as commanders of the autonomous UAVs, not their operators. This is the ultimate solution and probably one reason why the air force keeps insisting that UAV pilots be officers. Flight control and pattern analysis software takes a lot of the work out of operating a UAV. Pattern analysis software can spot what is being looked for on the ground and is rapidly approaching the point where it does the job better than human observers. Thus the future was seen to be officers commanding several UAVs, each largely “operated” by software. Each officer would then be assisted by one or two NCOs to help deal with any situations requiring human intervention. The trouble is that sort of software is not here yet and not for another five or ten (or more) years. In the meantime, the air force brass was forced to do the unthinkable and return to the World War II practice of using NCOs as pilots. Alas, this worked in World War II because the NCO pilots could shoot back at the enemy and many proved quite good at it. That may have something to do with the air force reluctance to change but now they must and, as the army has already demonstrated NCOs, can handle armed UAVs quite well and that warrant officer pilots are the pilot shortage solution the air force is looking for but will not adopt. That is a leadership problem, not a pilot retention problem.