With the dust settling in Gaza and Israel, one of the big take-aways in the Western media seems to be that the Iron Dome CRAM (counter rocket, artillery, mortar) system was a smashing success, and has revived interest in missile defense as a viable strategic goal. I'm a bit more skeptical, but I'm curious what some of you folks out there think about the cost/benefit analysis of the US (or Western allies in general) aggressively pursuing a comprehensive missile defense strategy. In this article, I will examine missile defense systems currently in operation and development in order to assess their efficacy and viability. Given the economic and technical realities, I will argue that certain forms of missile defense are more worth pursuing and developing than others. In particular, I believe that CRAM systems like the Iron Dome and tactical ballistic missile defense like Aegis and Arrow are relatively proven and reliable technologies with reasonable costs for the political and military benefits. They benefit from flexibility, modularity, and a clearly articulated need. However, I will show that other missile defense systems in development - notably those aimed at medium to long range rockets and cruise missiles such as Magic Wand, as well as strategic ballistic missile defense against ICBMs - are largely overpriced and irrelevant for most governments. In particular, I single out the Ground-based Midcourse Defense system as a fundamentally flawed technology with little strategic or tactical value which is being implemented at enormous cost.
Beats CRAMming into bomb shelters
There is no question that the Iron Dome system is pretty fancy technology. In the latest round of fighting, it appears that the system had an 84% successful intercept rate (with the radar and C&C system determining about 2/3 of incoming rockets were not significant threats and choosing not to attempt an intercept) - some 400 intercepts have been recorded to date. There were some snafus, mostly with the control systems, but overall it appears that the system will approach 90% interception success for rockets with ranges up to about 70 or 75 km. Further upgrades might push this interception rate a bit higher, but likely not too much further - the short intercept time complicates matters. This is by far the most successful missile defense system to date. It is true that the mortars and rockets being intercepted by the system aren't particularly large, heavy, or fast (making the job of the interception a bit easier), but the time to intercept is also incredibly small, generally ranging from about 15 seconds to a minute. That's part of why Israel has to deploy so many batteries to protect its cities - more centrally located batteries wouldn't be able to launch an interceptor in time to meet the incoming threat.
I would say that this is indeed a more capable system than the earlier proposed alternative, a land version of the Phalanx CIWS for point defense of ships (it also is far more practical than an earlier idea using chemical lasers called Skyguard/THEL). It has greater flexibility and range, and has the ability to attempt multiple intercepts. The unit cost per intercept is certainly higher, but a single Iron Dome system can cover much more ground against more varied threats. An issue with both systems is the danger to surrounding areas from the system itself - the Phalanx system is supposed to have self-destructing rounds, and the Iron Dome is supposed to direct unexploded missiles to uninhabited 'graveyards', but there's still a lot of danger there. The Phalanx/Centurion ammunition is likely to have some duds that don't self-destruct, and shrapnel from successful Iron Dome intercepts has cause damage and some minor injuries in Israel. I would imagine that protecting a single small region (such as a military outpost) is still probably a job for the cheaper-per-intercept Phalanx system, but any more comprehensive solution would need something akin to the Iron Dome (or, potentially, a heavily re-engineered land-based version of the SeaRAM).
The reason this kind of technology hasn't been implemented until now has largely to do with speed. Until recently, we didn't have the computer power or sophisticated enough control systems to do the necessary calculations for intercept in the allotted time. It also helps that missile technology has gotten pretty fancy, and it's possible to build relatively cheap interceptors that are still quite maneuverable and sophisticated. $40k a pop might not sound cheap, but it's a drop in the bucket compared to other modern missiles - a modern air-to-air missile will run about 10 times as much, and a cruise missile will cost maybe 50 times more. Obviously it's more expensive than an asymmetric threat (a rocket that costs at most a few hundred dollars), but given target discrimination abilities and the savings in lost property/lives, it probably approaches being cost effective.
So much for the CRAM threat; I think given this new reality it makes sense for Western militaries to invest in these systems for specific needs - generally, point defense of military ships and bases as well as area defense against asymmetric threats in relevant environments. These systems are clearly effective and change the strategic calculus when face with short-range asymmetric threats, giving governments a much greater degree of political freedom in responding to such threats. Cost is a significant concern, but marginal costs are low enough compared to alternatives (such as much more costly ground invasions or extensive fortification) that it is probably worth it for most scenarios. Since the technology is now proven, development costs and risk are fairly minimal. The biggest buyer I'd see (now that the US is winding down Afghanistan) would be South Korea, though some other countries might find it a useful investment as well for point defense of military bases and sensitive installations.
Wanna see my Magic Wand? That will be $1 million
The real question, though, is whether longer-range systems should also be pursued. There's several steps between a CRAM system and a full ICBM shield, so I'll address each one:
In the sub-ballistic missile range there are long range artillery rockets (up to a few hundred km) and cruise missiles. These are a more challenging target in that they tend to be faster, larger, and potentially can be steered, but you also can get away with fewer batteries and more time-to-intercept (allowing for multiple shootdown attempts). The cost disparity between interceptor and target may decrease, depending on the particular threat, which makes the economics more favorable for an interception system. So far there is precious little data out there on functioning systems. Israel has been aggressively pursuing this with its Magic Wand system, which just achieved a notable milestone with its first test interception, and it seems likely they will be able to successfully replicate their success with the Iron Dome - much of the basic technology is the same, given that the target is not all that different. I am more skeptical of their claims that this will form the basis of a comprehensive air defense system, which Magic Wand interceptors being able to engage cruise missiles, aircraft, etc. - those are much more maneuverable, and it may drive costs up to have to engineer for those threats as well. Missiles that are too broad in design will often fail to be very effective at any one task.
The only other vaguely similar system is the bottom end of the Patriot engagement envelope. Though envisioned mostly as an anti-ballistic missile system (for the PAC-3; earlier versions are better optimized for engaging aircraft as well), it does have some capability of medium to long range rockets as well as cruise missiles. I am still quite skeptical of the efficacy of the Patriot after its relatively lackluster combat performance in the past, but I do indeed recognize that the system has been dramatically upgraded in the last twenty years, and it may indeed be capable. The real challenge is the economics - Patriot missiles are too expensive to waste on intercepting a Zelzal rocket, so they're not really suited for this kind of engagement. It's a bit better of a cost differential with cruise missiles, but the PAC-3 is not really suited for that role.
I'd say the jury is still out on whether it's worth it to invest in medium-to-long range rocket defense systems like Magic Wand. I understand Israel's interest (and South Korea's), but it seems to fit in the uneasy space between point defense (very useful for the military) and theater defense (very useful against unconventional warheads). It's good for protecting large civilian populations against long range asymmetric threats, but probably isn't very good at much else. The US and most of its allies should probably not pursue this very aggressively.
Sheltering under the Aegis of America's warm (and expensive) embrace
Now we get to the meat of the debate: ballistic missile defense. This is by far the most expensive part of the whole story, and the most prone to failure. The strategic stakes are also far higher. I would distinguish between two categories here: theater ballistic missile defense and strategic ballistic missile defense.
Theater defense has a pretty long and varied history - many of the early systems, like the Patriot and various Soviet systems, evolved out of sophisticated surface to air missile systems which were given dual status in the vain hopes of some level of ballistic missile defense for a given theater of war (say, covering a small country with a single system against most ranges of ballistic missiles). Most of these early systems were pretty bad at their job, mostly because it's easier to make a plane not airworthy than it is to keep a relatively hefty missile from screaming out of the sky and exploding. Given the relatively small amount of accuracy these ballistic missiles need if equipped with unconventional warheads (the only kind of warhead that makes sense for a medium to long range ballistic missile), you need to have a pretty lethal and accurate kill vehicle. Your defense system also needs to have a very low leakage rate against fairly large salvos (say, at most letting 1 in 1000 through in a salvo of dozens or even hundreds). A single miss will likely make your system a moot point, so it has to be very good.
This makes these systems very challenging (and expensive!) to develop. It also means that outside of a pretty big shooting war between fairly sophisticated enemies, there's no way to test the system in a realistic setting. The big advances since the early operating theater ABM systems in the 80s are threefold: better tracking, better maneuverability, and better lethality. Radars have gotten a lot better, as have computers - this helps a lot. Also, as they have expanded the interception envelope higher into the atmosphere (or even into space), it gives systems the possibility to do things like shoot-look-shoot to have multiple interceptors tracking a single target (90% interception rate jumps to 99.9% if you shoot three interceptors). Radar systems are also much better at tracking multiple targets in real time. Missiles are now much better at moving, using clever (but fancy) technologies like thrust vectoring, fancier terminal seeking, and better on board computers. This also is somewhat useful in better target discrimination from decoys or MIRV designs, though not as much protection as one would like. Lastly, the missiles are more lethal to targets - they have learned from the near-misses of the Patriot during the Gulf War which tracked Scuds decently but were unable to take them down. There are a variety of kill vehicle approaches - fancy proximity fuses, hit-to-kill, etc. - but they all take into account the early design flaws. It's hard to know if hardened terminal stages of medium and long range ballistic missiles will be able to withstand new kill vehicles, but at least it's an area of significant concern to current designers.
So - the technology is better, but is it better enough? Early ABM tests in the 90s using new theater ABM systems were notoriously iffy, but results in tests and simulations have gotten a lot better - and not just from the designers making easier tests. Of particular interest are the US' two major theater systems and Israel's Arrow system. The US' best tested system is the Aegis BMD, which was originally developed for naval defense on cruisers and destroyers using the SM-2 and SM-3 missiles. The missiles are pretty sophisticated, albeit expensive, and can go exoatmospherically for midcourse as well as terminal phase intercepts. A single battery has coverage for a decent area, and this is the basis for the 'interim' phase of European missile defenses (several Aegis-equipped US Navy ships hanging out in the Mediterranean etc.). It has the advantage of being more flexibly deployed than ground-based systems, and it is based on relatively proven technology. The big challenge is that tests so far haven't been that realistic against emerging threats (more sophisticated re-entry vehicles coming from Iran, NK, China), and the missiles are quite pricey (over $10 million a pop in most cases). The US also has the THAAD system, which is solely a terminal interception system from ground units. It has a very troubled development history in the 90s, though more recent tests have been more promising. It remains to be seen whether it will gain the desired capability, but bits of it have been deployed already - the US Army has a small initial operating capability, multiple X-band radars have been installed (including one in Israel), and the US is selling the system to some Gulf region countries (notably the UAE). I'd say the jury is still out on THAAD compared to the more mature Aegis system. The Israeli system, Arrow, is more rigorously tested than either and the missiles may be a bit cheaper. It also is the first theater ABM system in operation. It has similar caveats to the Aegis system wrt realism of the tests, but the new interceptors in the later blocks of the Arrow 2 and the new Arrow 3 all appear to be quite capable.
What's the verdict? I'd say that the technology remains to be tested and proven in battle, but that it fills a glaring need for defense against unconventional threats from sophisticated but not near-peer competitors (e.g. NK, Iran), and against conventional threats from belligerent ones (China). The problems of intermediate range ballistic missile defense are a bit easier to solve than a strategic shield, and the systems tend to be a bit more modular, cheaper, and flexible. Tests in the last decade or so have been increasingly showing successful intercepts against more and more sophisticated engagement scenarios. I'd say it's worth it. Lots of countries agree with me - bits of the Arrow system have been sold to India and South Korea, and the US has sold Aegis to a boatload (heh) of various Navies - Japan, NATO, Australia, possibly SK in time. THAAD has also been sold to UAE and is evincing interest elsewhere. I don't think these systems satisfy leakage requirements for unconventional warheads yet, but they are getting there. The obvious regions of concern are Asia (NK and China) and the Middle East (Iran) right now, and a policy of containment is probably better than outright confrontation. These systems give us a bit of flexibility in discussing other actions, whether military or diplomatic, which is not a bad thing.
Good Morning Deficit
The last type of missile defense is by far the most expensive - strategic ABM shields for global coverage against ICBMs. Currently, this is only a threat from near-peer competitors - Russia, China, and technically some Western nations we wouldn't consider threats (Israel, Britain, France). India will get there eventually, but Pakistan, Iran, and NK will still need quite some time before they have a real threat there. These threats also tend to have the most sophisticated re-entry vehicles, are extremely fast, and almost all carry unconventional warheads. To date there is no functioning system that can convincingly intercept these threats. The closest we have is the US' Ground-Based Midcourse Defense (GMD) program, which uses a series of very powerful radars to detect and track missiles, and a handful of ground-based silos to launch interceptors for a 'midcourse' intercept in space. This has the advantage of being able to provide much larger coverage (you'd only need 2 or 3 silos to convincingly cover the US), but is an incredibly hard problem and requires very large, fast, and sophisticated interceptors and kill vehicles. To date, the US has sunk tens of billions of dollars into this program with extremely mixed results in tests. Identification and tracking (with X-band radars and the like) seems to be pretty decent, but actual interception has been plagued with failures and troubles with the interceptor. It also leads to all sort of political issues, especially with Russia.
A lot of other (largely loony) ideas have been tried over the years as well. The looniest one lately was the Airborne Laser system, where Boeing stuck a gigantic chemical laser into a 747 for shooting down missiles in their vulnerable boost phase. It never made any sense (how do you get a 747 close enough to an enemy launch to be effective?) and was incredibly expensive. Ditto with plans for orbiting X-ray laser satellites from back during SDI. Strategic ABM systems have come a long way from that, and in principle it seems likely that the latest attempts will be relatively successful given enough money and time. I question, though, whether this is all an exercise in futility. It's a lot easier to make re-entry vehicles on ICBMs more difficult to kill than it is to make interceptors even better than they are. The costs start to get prohibitive as well - ICBMs are expensive, but you need quite a lot more (also expensive) interceptors to convincingly defend against a large salvo of ICBMs. The economics doesn't really work, and once a country achieves a relatively large arsenal of unconventional-tipped ICBMs, it's probably a waste of time to try for interceptions. Something like GMD might be useful if a country like Iran or NK ever gets hold of true ICBMs and gets smaller nuclear warhead designs - it's unlikely either country will have the resources to develop a large enough arsenal to overwhelm GMD. But the development costs are so prohibitive that's it's probably wiser just to make sure they never get such capabilities in the first place. As for near-peer competitors, though, strategic ABM is probably an expensive boondoggle that won't even work effectively.
To conclude, it's true that missile defense has become much more of a reality than the days of SDI. Technology has evolved to the point that the real limiting factor is one of economics rather than feasibility. Given these threats, though, it's not always clear that the cost/benefit analysis supports aggressive investment into various forms of missile defense. Point defense using cheap and flexible systems like Iron Dome (and to a lesser extent, Phalanx/Centurion) is appropriate for some applications, notably for protecting civilian areas and sensitive installations from nearby asymmetric threats. It's a useful technology that was rapidly brought from initial conception to operation, and marginal costs are low enough to make it viable for a number of different scenarios. Longer range rocket/cruise missile defense can be part of a multifunction air defense system for an entire country, but is still relatively untested and should only be pursued by countries with immediate needs in this area (SK, Israel) - this is such a niche need that it's probably a waste to spend much money on development outside of these particular countries. Theater ballistic missile defense is convincingly reaching feasibility now and seems like an eminently wise investment against a number of emerging threats - it may be expensive, but it gives policymakers a great deal of flexibility in dealing with unconventional threats from non-peer competitors. It also is not too expensive, and has given us convincingly working technology. I would caution, though, against funding theater ABM systems in perpetuity if a better and cheaper alternative exists. These programs have inertia, and I fail to see why THAAD wasn't cancelled a decade ago given the relative success of Aegis BMD and Arrow. Conversely, strategic ballistic missile defense is almost certainly too expensive and complex to work against any real threats for the foreseeable future. In times of budgetary austerity, it's a disgrace that we have spent over $30 billion on a technology that is so incredibly complex... and yet still not have anything approaching a working system. It's expensive, politically a mess, and probably useless against most real strategic threats. It's not being 'soft on defense' to suggest this is a waste of our resources, and I would urge policymakers to cut or curtail this program for the foreseeable future.
I'm sure that plenty of people will take the wrong lessons from the dramatic success of Iron Dome. It is not a panacea and should not be treated like that; it is a useful technological tool that gives policymakers more flexibility in their responses to a given situation. The lessons we should take are that missile defense technology is rapidly improving and is indeed capable of remarkable feats - yet it is still fundamentally limited by the calculations of economics and the extremely high intercept rates needed for unconventional threats. That being said, many missile defense systems indeed appear to be worth the investment. Others, though, are either lower priority (e.g. long range rocket interception) or not worth it (strategic ballistic missile defense).