Maladjusted, Part I: 21st-Century Attack

February 13, 2017

For special access to experts and other members of the national security community, check out the new War on the Rocks membership.

Once upon a time, the Air Force was flush with attack aircraft. Late in the 20th century, the Air Force maintained large numbers of attack aircraft, peaking in 1983 with a total inventory of 1,115 attack aircraft in three different models. Today, most mentions of the attack aircraft relate only to the A-10 Warthog, and underappreciated aircraft perennially in danger of retirement. Even among Air Force officers who should know better, the A-10 is referred to as a close air support (CAS) aircraft, discounting the myriad other roles that the aircraft has performed. Attack aircraft are the elderly second cousin at the Air Force dinner table, rarely spoken of and relegated to the status of historical relic, banished even from the official Joint Dictionary. This unfortunate condition makes it exceedingly difficult to discuss the utility, role, and historical contribution of attack aircraft to combat airpower. And it makes it especially difficult to discuss the most critical aspect of all — the Air Force needs to bring them back.


Historically, attack aircraft were distinct. They were dedicated to the ground attack mission, focused on attacking things on the surface of the planet. Unlike fighters, they were not intended to mix it up with other aircraft. They were also not expected to haul large bombloads over long distances, which was the purview of the bomber.  As attack aircraft numbers shrank, their classification fell into disuse.  It is possible to look it up — because the Department of Defense has to have a regulation for everything — under AFI 16-401, which is also Army Regulation 70-50 and NAVAIRINST 13100.16, titled Designating and Naming Aerospace Vehicles

These regulations use the following definitions:

A (Attack) — Aircraft modified to find, attack, and destroy enemy targets using conventional or special weapons. This symbol also describes aircraft used for interdiction and close air support missions.

B (Bomber) — Aircraft designed for bombing enemy targets.

F (Fighter) — Aircraft designed to intercept and destroy other aircraft or missiles. Includes multipurpose aircraft also designed for ground support missions such as interdiction and close air support.

This is a clear, but indistinct articulation of the mission classification, because the definitions for fighter and attack share an identical second line. And therein lies the conundrum. When all fighters became fighter-bombers, the niche formerly occupied by attack aircraft became even less distinct. The expanding capabilities of multirole fighters such as the F-4 Phantom presaged the death of the attack aircraft: why purchase an attack aircraft to attack ground targets when you could purchase a fighter-bomber that could attack ground targets and shoot down other aircraft? That very argument is what killed the A-7F Corsair III, a supersonic attack aircraft that found itself competing unsuccessfully with the multirole F-16A. So, to return the attack aircraft to its proper niche, we need to define it. Here is my proposition:

An attack aircraft is a multirole combat aircraft that performs tactical counter-land and surface attack missions but is not intended to perform counter-air missions against airborne enemy aircraft.

Defining the aircraft is one short step toward justifying it. But why on earth would anyone want a combat aircraft that doesn’t shoot down other aircraft?

Tradeoffs: Cost and Capability

One Word.  Cost.

Air-to-air capability does not come cheaply. You pay for it either with higher unit cost or design tradeoffs — or both.  In 1971, the Air Force reported to the Senate Armed Services Committee that an A-7D attack aircraft had a unit flyaway cost of $2.6 million compared to $2.5 million for the multirole F-4E, but that the A-7 had “twice the range and loiter time while carrying a 25% greater bombload.” In that same year, the Navy was paying $2.4 million for an A-7E attack aircraft and a whopping $13.1 million for early F-14A fleet defense fighter.  Baselined again in 1985 dollars, the flyaway cost for the attack aircraft produced in the 1970s was significantly cheaper than for their fighter counterparts, including fighter counterparts like the F-105 and F-106, built decades earlier.

Fixed-wing Costs in 1985 dollars (ADA-A151 575)

One reason for the cost differential is obvious: The purchaser doesn’t pay for systems (for example, air-to-air radars) that aren’t on the airplane. What you don’t need, you don’t have to test, buy, maintain, or upgrade. Cost savings made from design tradeoffs are less obvious. The A-10’s design philosophy focused on responsiveness, lethality, survivability, and simplicity. This allowed the Air Force to avoid cost drivers such as high aerodynamic performance. The Hog’s design tradeoffs sacrificed speed for lethality and survivability; its slow speed was a feature, not a bug. In the now almost-forgotten 1974 fly-off between the A-7D and the YA-10, the A-10 had several times the endurance of the faster jet and could operate under weather ceilings where A-7 pilots dared not go.

Loaded YA-10 Prototype, 24 December 1972.  Note that the iconic GAU-8A 30mm cannon is not installed yet; an M61A1 Vulcan is mounted in its place. (U.S. Air Force)

But that’s not the only reason why the Air Force might not want a combat aircraft that doesn’t shoot down other aircraft. It also comes down to capability. In a service that always seeks to fly higher and faster, the benefits of building aircraft without the air-to-air baggage have been forgotten. There are other attributes that might be desirable in an attack aircraft that cannot be designed into a high-performance air-to-air machine. For example: The OV-10 Bronco, which the Navy used as a light attack bird, was designed to take off from and land on roads. It could carry and drop three parachutists from the back. The A-7D could carry more bombs than the Phantom, further, and just as fast, while hugging the terrain on the way.  The A-37 Dragonfly had screens to protect its engine from debris, was easy to maintain, and could swap out an engine in an hour.

Colombian A–37 Dragonfly (Lt Col Mike Pietrucha)

The Air Commando versions of the A-26 could carry 12,000 lb. of ordnance (more than twice that of the B-25 bomber) including the ammunition for 8 forward firing guns, and deliver it all from a 30- degree dive.

The Long Game

The fight against non-state militant groups is not going away. The Air Force cannot look ahead and say with confidence that it will no longer be involved in the kind of air operations that have defined the last 15 years. And for almost 100 percent of the post-9/11 air operations, air-to-air capability has just been along for the ride. All of the investment in air-to-air design features are useful for finding the tanker when the weather is bad and for avoiding traffic, but not for guiding air-to-air missiles to a hostile aircraft or outmaneuvering enemy fighters. In the skies over Iraq, Afghanistan, and sundry other nations plagued by violent extremists, the aircraft we continue to use are a ridiculous overmatch for the missions we need them for. This method of operations comes at a high cost in premature aging among the U.S. legacy fighter fleet.

A-26K restored to its Vietnam-Era Appearance (Air Force Museum)

Why does this continue? The answer is simple: We have no other aircraft to use. The U.S. military put flight hours on fighter aircraft because they were capable of doing the mission, not because they were well-suited to the mission, and certainly not because they were the most cost-effective choice. In fact, the only way to make the fighter option less cost-effective would be to use the F-35. The sole focus on expensive multirole aircraft carries with it significant collateral damage on training, readiness, and the industrial base.

Historical Attack / Observation Total Aircraft Inventory by Type, 1963-2015

The post-Gulf drawdown demolished the Air Force’s attack aircraft inventory, and the portion of the industrial base that should have been building attack aircraft is moribund. It’s a minor miracle that the F-15E and F-16 production lines are still open, and only because they have been entirely sustained by foreign sales for 15 years. In the process, the United States lost the ability to produce aircraft suitable for partners with smaller budgets and limited airpower capabilities. If you cannot afford an F-16, you’re not going to be shopping for a U.S. combat aircraft. With the loss of the lower-end fighter/attack market comes a loss of ability to build partnerships and support foreign aviation capabilities.

It also means that the workforce that builds and equips combat aircraft is smaller and much less redundant than it should be. In 1956, the United States had six open production lines producing fighter and interceptor aircraft — a healthy and competitive market. By 1966, this was down to five. 1976, four. Every ten years, the United States lost a production line and often an aviation company until 2006, when there was only one production line open and only one company (Lockheed Martin) producing fighters for the Air Force. Boeing produces fighters for the Navy and for foreign F-15 customers, and Northrop hasn’t built a fighter since the YF-23, 25 years ago.

Air Force Fighter / Interceptor / Attack Production Lines, 1954-2016


The shortage of combat aircraft is substantial, pervasive, and it has an impact on warfighting potential, pilot inventory, readiness, and the industrial base. A vision of exactly how bad is can get is easy to find: Look no further than the Marine Corps’ fighter/attack enterprise – where two thirds of its fighters are regularly grounded -to see a real-time example of how failure to recapitalize and a high operations tempo can shatter a force. Replacing the F-18 and AV-8 with an aircraft that costs more to buy, fly, and maintain is not a recipe for success in an environment where the ops tempo remains high, year after year. The Air Force has a solution to many of its tactical aviation woes, and it’s one that we have used before — returning to a high/-low mix of aircraft. But while the term is often tied explicitly to the F-15/F-16 purchases of the 1970s, it could apply today in terms of aircraft roles rather than aircraft models. Leave the high-end fight to the fighter/bomber aircraft. There is plenty of room for attack aircraft in the wars we have actually been fighting for 25 years. In the Vietnam Era, the United States solved a great number of tactical aviation shortfalls with the introduction of attack aircraft tailored for the war it actually fighting instead of the wars it expected to fight. The United States can do this again. Twenty-five years after America gave away all of its attack aircraft except for the A-10, it’s time to bring some new aircraft home to roost.


Col. Mike “Starbaby” Pietrucha was an instructor electronic warfare officer in the F-4G Wild Weasel and the F-15E Strike Eagle, amassing 156 combat missions over Iraq and the Former Republic of Yugoslavia and taking part in 2.5 SAM kills over 10 combat deployments. As an irregular warfare operations officer, Colonel Pietrucha has two additional combat deployments in the company of US Army infantry, combat engineer, and military police units in Iraq and Afghanistan. The views expressed are those of the author and do not necessarily reflect the official policy or position of the Department of the Air Force or the U.S. government.

Image: U.S. Air Force photo by TSgt Frank Garzelnick

We have retired our comments section, but if you want to talk to other members of the natsec community about War on the Rocks articles, the War Hall is the place for you. Check out our membership at!

14 thoughts on “Maladjusted, Part I: 21st-Century Attack

  1. Mike,
    As a long ago once were Navy guy who started out in F-8 Crusaders but went to war in an A-7B, let me say just excellent article. I have been following your articles with only one bitch comment – you never talked true attack and Navy Marine Corps. Here you nail it.

    I run a website with an ongoing analysis series on this very subject and would like to connect with intent to possibly republish some of your stuff either in part or in whole. Was refered on this particular piece by Brian Laslie author of Air Force Way of War as part of ongoing discussion on the “attack bidness.
    Ed Boris Beakley

  2. Well, yes, but…. The world’s changed since Vietnam. A couple of particulars make the cost argument difficult to sustain:

    Night. U.S. ground forces have made a deliberate shift to take advantage and fight at night and in poor weather. That means fitting attack aircraft with expensive sensors to find targets and avoid the ground. Many of the A-10’s upgrades have been to improve it’s ability to fight at night. Both the Air Force (AC-130, F-111, F-117) and the Navy (A-3, A-5, A-6) operated night/all weather attack aircraft, but with exception of the AC-130, all were really light/medium bombers, and none were cheap.

    Culture: Not Air Force fighter culture, but DoD acquisition culture. The A-37 and F-5 programs succeeded because the Air Force bought bunches of them and either gave them away or threw them away when done. In any case, 10 years was considered a decent service life for a combat aircraft. No more…now we expect aircraft to last for decades, which doesn’t make them cheap.

    1. Completely agree. Quality beats quantity, as we saw in Iraq, and in contested airspace I’d sooner have one F-35 pilot controlling an F-35, 3 UCAVs and 10 missiles launched from other platforms, than 14 pilots flying 14 A-10s.

    2. Night-fighting sensors were expensive. They aren’t as relatively expensive anymore, and where they used to be only useful at night, now they far outperform the Mk 1 eyeball in all conditions, so even day missions require them.

      The real reason we need observation/attack aircraft is to be FAC(A). A Forward Air Controller (Airborne) can call in strikes by other aircraft, so the only thing an F-15E, F-16, or F-35 pilot needs to do is punch the provided coordinates into the computer and drop the bomb when the jets says to. That allows a large force of fighters to be multirole without much effort or training time, while maximizing the firepower available to the specialized FAC(A) pilots. For that mission, an OA- aircraft needs long on-station time, several radios/datalinks, targeting sensors, and a well-designed cockpit, but it doesn’t need expensive speed, air-to-air radar, or even that large of a weapons load.

      1. The FAC task is much better performed by the F-35 than any other aircraft in the world. There is no aircraft in the world with a better command of a battle space. The F-35 has an unmatched breadth of a sensor suite, state of the art communication equipment, unparalleled situational awareness, to boot it is stealthy. Besides the F-35 was designed for the FAC role.

  3. Col. Pietrucha ignores the following developments since 1991:

    1. Nearly all US airstrikes are now conducted by precision-guided munitions, which wasn’t the case even as late as Desert Storm (~10%).

    2. UCAVs are increasingly used by all service branches to provide low-end observation and land-attack, including CAS for ground forces. The UCAV force is growing or expected to grow by at least +3.40% per year.

    4. Advanced air defenses are proliferating, making the A-10 useless in a high-end environment and perhaps even too vulnerable against a middle or low-end threat such as Iran or North Korea. Note that the A-10 was originally intended to bust high-end Soviet tanks, not to act as flying artillery for open-ended COIN missions in the Middle East.

    5. Where is the DOD going to find enough pilots for high-end, low-end and unmanned platforms?

    6. Why does Col. Pietrucha continue to argue against the F-35 when he only has access to OSINT on the project? In 1990-1991, OSINT would have told us that Desert Storm would be a bloody affair for US aviators, wouldn’t it?

    7. Col. Pietrucha mourns the loss of various aircraft producers, but ignores the fact that the US aircraft inventory was unprepared for the Vietnam War, and was only adapted after North Vietnam had established a high-end and dense IADS.

    8. Col. Pietrucha refers to cost, and yet he ignores that low-end aircraft need to be built and deployed in greater quantities, whereas the F-35 is not intended to replace a single 4th Generation aircraft nor match opponents 1:1 but more likely 1:4.

  4. Col Pietrucha,

    I think this is a good article and an especially great look at the cost effectiveness of using multirole fighters nearly exclusively for our dynamic targeting/CAS mission sets. However, I think you’ve missed the Air Force’s biggest reason for not producing more attack aircraft – they’re replacing attack aircraft with unmanned assets. The MQ-9 has all of the advantages that you mentioned (long loiter time, slow airspeed, specialization) with the notable exception of payload (typically flying with “only” four Hellfires and a 500lb bomb). What’s more, the operating costs per hour of MQ-9s are far under that of even the cheapest attack aircraft.

    But it’s also important to look at developing needs in attack aircraft beyond what older models could provide. The biggest challenge for counterinsurgency operations is rarely having an asset overhead with the proper munitions, thanks to unchallenged air refueling and the ability to launch aircraft on short notice and without the need for a defensive CAP, not to mention the generally air-to-ground focused standard combat loads that air supremacy allows for. Instead, the challenge often lies in finding and fixing the targets to strike. This is where the MQ-9 especially can outperform tradition-style attack aircraft including the A-10. A better pod, large screens, and a dedicated imagery analyst gives a much better chance of actually finding a target to strike. This also leads to trickle down efficiencies – with MQ-9s identifying targets, all aircraft need to loiter for less time, and can instead be passed targets are they ingress, strike them when on station, and leave. This drives down their number of hours needed aloft, increasing efficiency for the fleet.

    There’s more to be said here, and there are downsides to unmanned aircraft, especially in contested environments. But as you say, we need to win their battles we’re fighting before we can get to tomorrow’s fight. And unmanned aircraft are filling in now for the niche that you’ve identified.

    1. He’s only interested in making a case against the F-35, unfortunately. That is what this is all about. If you add the COST of providing a permissive airspace for the A-10 to operate in (SEAD, OCA), its true COST actually rises…

      1. The A-10 should be retired. In the environments it currently can operate in a prop driven plane would make more sense. The A-10 requires $10,000 of maintenance for every hour it flies. A prop based plane requires only $1,000 of maintenance for every hour it flies. The A-10 was designed to take out massive columns of T-72 Russian tanks. Its lifespan even 30 years ago in such an environment would probably not be two minutes. And the job of tank buster disappeared 30 years ago with the Cold War.

  5. Are drones the “low” of the modern “high/low” mix? Circa $17m for a Reaper compared with $100m+ for a F-35? 4x Hellfire/Brimstone missiles and 2x 500lb smart bombs may be much less in bulk terms than say an A-7, but add in the precision elements and the leathality , probably similar?

  6. Starbaby,

    Your math on the cost of air-to-air capability and high speed in jets that don’t use it is spot on. Also, multirole fighter pilots need to stay current in all their missions, which means more flight hours in a more-expensive aircraft.

    However, you are far to generous to the writers of “Designating and Naming Aerospace Vehicles.” Under the definitions for Fighter and Attack, aircraft with any air-to-air capability are biased heavily towards the Fighter classification, and the Attack classification is stated as a modifier rather than a primary type. Attack as a primary designation is only defined negatively: used for interdiction and close air support missions, but not a Fighter.

    A new definition of Attack is a starting point, but all the official definitions should strip “multipurpose/multirole” language out. Modifiers and split designations (F/A-18, e.g.) exist, so there is no reason any classification should be using multipurpose language to take over the conversation. The definition for Fighter should note the historical ability of fighters to employ air-to-surface weapons, but the current definition allows the fighter community to completely crush the debate over air support.

    1. There seems to be a belief that the “Attack” designation favors aircraft designed with CAS in mind…the A-10 (and its A-9 competitor) are the only examples where that’s true. When the Navy picked up the “Attack” designation after WWII, it was essentially a way of avoiding the term “bomber” and attracting unwanted attention from the Air Force. In spite of that, the A-1, A-4, A-6, and A-7 were designed as ship-killers; the A-2, A-3, and A-5 were nuclear bombers.

  7. Small note: About the only way you could get a “Sluf” supersonic was to take an absolutely clean (no drops, no nothing on the outboards) a/c to 50K, roll over and pull the nose straight down while two-blocking the throttle. Even then it would start to back out through the Mach at 20K as the density altitude began to build drag. Transonic maybe, supersonic – not so much.

    I’ll bet it would be interesting to put the A-4F (w/ J-52-P408) or the A-4M in your mix. Another interesting historical comparison might be the way the 8th and 15th Air Forces used the P-51 / P-47 / P-38 mix in the ETO. Mustangs high w/ the bombers, Lightnings in the middle. and Jugs of the 15th AF shooting up everything that moved on the ground. Might be a good model for how to use a large conflict hi/low mix.