We’ve had plenty of defence announcements so far this year. Many have rightly focused on Australia’s situation and vulnerabilities as an island trading nation, one whose daily survival depends on the fuel, fertiliser, pharmaceuticals and other essentials imported by sea.

This awakening has been backed by significant announcements for the program for nuclear-powered submarines, a historic increase in the Royal Australian Navy’s surface combatant fleet, and a capability acquisition plan that puts 38 percent of its money into the maritime domain.

From the glossy documents you could be forgiven for thinking that Australia’s naval capabilities have been strengthened in the short term—that our navy can defend maritime trade and project into the region to contribute to our strategy of deterrence by denial.

The perception of growing naval power is only half true, however.

Ten years from now, when Australia’s first Virginia-class nuclear-powered submarine will have arrived and new general-purpose frigates and Hunter class anti-submarine frigates are emerging from shipyards, the country will have its most potent navy in decades.

But before then its naval power will decline, without significant readjustment of its naval posture, and so will its ability to defend maritime trade. Despite the increased spending over the decade, the alarming fact is that for the next 10 years Australia will have the least capable surface fleet it has had in more than 50 years.

The surface combatant fleet, comprising the frigates and destroyers that constitute the backbone of the navy, will increase to 20 in the 2030s and early 2040s. But it will contract until the early 2030s.

Most reviews of the naval surface fleet over the past 50 years have found a need for 16-20 surface combatants. These assessments were conducted before the rapid militarisation of the region, before China became aggressive in the South China Sea, and before Australia was subjected to its economic coercion.

Australia now has 11 surface combatants. HMAS Anzac, one our eight frigates, will decommission this month, reducing the surface combatant fleet to 10. Then in 2026 the fleet will fall to nine, around half the strength recommended by 50 years of analysis—and this in a time of global tension that has been compared to the period before World War II.

How did we get here? It’s a long story that cannot simply be attributed to the current or former government. It is a story of devaluing naval capability, of delaying decisions and accepting risk where the risk was not fully understood.

But the pivotal question is what to do about the widening gap between needs and capability.

The answer is in naval structure, readiness and posture. We must be bold. For the next 10 years, we must maximise the potency, operational readiness and posture of our remaining ships.

This includes maximising their sea time, with alternate crewing and readiness models. Rather than one crew per ship, we may need multiple crews per ship. The Royal Australian Navy has previously trialled that, and the Royal Navy is doing it now with frigates it deploys to the Middle East.

But what about the implications for the challenged naval workforce? If alternate crewing matters enough, we can make it happen. Historic times call for historic approaches.

But we must go further. To maximise the utility of our diminishing naval assets, we should take a leaf out of the playbook of our AUKUS partners and permanently deploy one of the remaining surface combatants to Southeast Asia.

Basing a frigate in Singapore or the Philippines would increase Australia’s presence in the contested South China Sea, maximising our naval potency during the foreseeable capability gap.

While Australia is spending more on naval capability, during the next decade it will have its least capable naval surface fleet.

In times of historic global tension, the government and navy must take bold action to maximise the operational effectiveness of our remaining fleet. The lifelines of the Australian economy and our way of life may well depend on it.

A recent Strategist article by ASPI’s Marcus Hellyer highlighted issues becoming evident with the Hunter-class future frigate as the planned construction date slides another 18 months to mid-2024. A lot of the delay is put down to modifications to the British Type 26 design to accommodate Royal Australian Navy requirements. This has had negative impacts, including increasing the weight from 8,000 tonnes to 10,000 tonnes.

Contributors to The Strategist have previously discussed what makes a frigate a frigate or a destroyer a certain type of destroyer and whether they’re really cruisers. One writer implied that this might be an insider trick to improve the RAN’s capability without the ‘bean counters’ realising it’s happening.

But if we want to increase the RAN’s capability quickly, there already is a vessel meeting most of its frigate requirements—the Hobart-class air warfare destroyer. Simply continuing to build the AWD as the replacement frigate (or whatever you want to call it) may have many advantages over designing a separate platform.

At first glance, the Hunter frigate’s requirements are remarkably similar to the Hobart AWD’s specifications. They have similar missiles and guns, they’re both designed to use the MH-60 Romeo helicopter and both have hull-mounted sonars and towed arrays, so they are roughly equivalent for anti-submarine warfare.

Their dimensions are close. The Hobart class is 146.7 metres long with a beam of 18.6 metres, and the Hunter class is said to be 149.9 metres long with a beam of 20.8 metres. But the weight of the Hobart class is stated as 7,000 tonnes, far less than is suggested for the Hunter class.

The Hunter class is reported to have a range of over 7,000 nautical miles, compared with the Hobart’s 4,500 nautical miles. However, if the reported weight increases in the redesign of the Hunter are to be believed, that range might well decrease.

It’s also assumed that the RAN is comfortable with the AWD’s range for its surface warfare needs. The Arafura-class offshore patrol vessel is an apparent replacement for the Armidale-class patrol vessel, but it’s more like a corvette, so it could be argued that the RAN isn’t losing patrolling capability.

Requirements for the future frigates remain straightforward, but it seems that the implementation of some RAN specifications is causing considerable engineering challenges. The design reportedly needed significant modifications to enable the use of the Australian CEAFAR radar system, the Aegis combat system, US weapons already in Australian service and the Seahawk maritime combat helicopter. The contract also covered any other modifications that might be needed to meet Australian regulatory requirements. The Hobart-class AWD already meets those requirements except for the CEAFAR radar.

It’s not clear why this radar system is essential. The AWDs don’t use it as their air search and detection radar, yet they are considered excellent air warfare fighting platforms.

It can be assumed that the CEAFAR radar is included because it’s Australian made. Its design and installation in the Anzac-class frigates, as documented by ASPI, is a great advertisement to the world of Australia’s engineering capabilities. It fulfilled a dire need for a more effective air warfare system on the Anzacs and greatly enhanced their capabilities.

However, to insist that the system be used because it is Australian seems to be the equivalent of insisting that the South Australian car industry must remain open simply because it’s Australian, regardless of the cost.

The long-term survivability of a technologically advanced company such as CEA Technologies should be guarded as it makes a huge contribution to national security. However, since the future frigate requirements are very close to the current AWD’s capability barring the CEA radar, it seems appropriate to ask how crucial the CEA radar is.

Since the Hobart class is already operational and mission capable, we could start building more AWDs very quickly, potentially on the original timeline for the future frigates, starting at the end of 2022. The design is already done if we resist the temptation to tweak it. Rectifying a few system design or obsolescence issues would be orders of magnitude easier than making the Type 26 modifications.

Australia’s 2020 force structure plan provides $45.6 billion (in out-turned dollars) for the nine Hunter-class vessels. In constant, current-day dollars, that’s in the $30–35 billion range. The final cost of the AWD program will be less than $8.5 billion, but the ‘sail away’ cost of the third Hobart will be around $2 billion, possibly even less than that. Whichever way you look at it, any additional Hobarts will cost less than Hunters. Aside from being a mature design, they will be substantially smaller.

As has been clearly documented, despite a huge increase in defence spending, the navy’s capability is hardly going to increase in the next decade. However, with solid data from the AWD build, one could come up with accurate schedules for building more Hobarts. The RAN could have new vessels sooner, as long as the hard lessons from the original AWD build are learnt.

This would also introduce economies of scale.

Today’s decision to drop the Attack-class submarine and move to a nuclear vessel demonstrates the huge challenges with making major modifications to an existing design. It also acknowledges the difficulties with operating and maintaining unique platforms. Concerns have been rightly raised about the huge logistical cost of the RAN maintaining these unique AWDs. Increasing the number of Hobart-class vessels could reduce maintenance costs per vessel and make it easier to support them in the long term. The government and the RAN made a courageous decision in accepting these realities with the submarines; they should apply such thinking to other platforms.

In part 1, I discussed how the Anzac class of frigates is only halfway through its service life and the last of the class now needs to remain in service until 2042–43 as the new Hunter class is rolled out. The key question, then, is can the Anzacs remain a relevant frontline capability for another 24 years?

It’s useful to distinguish between the quality and quantity of capability. In terms of quality, the Anzac class is a long way from its original infamous ‘fitted for but not with’ configuration. After a series of upgrades (see the table below), the Anzac is now tonne for tonne among the best frigates in the world. And, as noted in the table, further upgrade projects are either planned or underway.

Central to its new capability is the anti-ship missile defence program that upgraded the excellent Saab 9LV combat management system and installed the world-leading CEA Technologies’ phased array radar to target the Evolved Sea Sparrow missile against incoming cruise missiles. The upgraded system has performed very well in trials. Australia is also participating in the international consortium that’s producing an enhanced version of the ESSM designed to counter modern supersonic  missiles. This will equip the Anzac, Hobart and Hunter classes. The replacement of the Anzac’s long-range search radar with a version of CEA’s phased array radar has begun.

Taken together, these improvements significantly enhance the Anzacs’ self-defence capability in air warfare and give them some ability to protect escorted vessels. But the lack of the Aegis combat management system and SM-2 long-range air defence missile means that the Anzacs don’t have a true area defence capability; for that, the navy will have to rely on its three Hobart-class air warfare destroyers until the Hunter class frigates come on line. Hypersonic missiles will likely become operational while the Anzac class is still in service and will certainly be challenging to counter, but they will threaten the survivability of any surface ship, not just the Anzacs.

In terms of anti-surface warfare, the Anzac’s five-inch gun is consistent with those of other major surface combatants, but its main anti-ship weapon, the Harpoon missile, is now dated. It has a shorter range than threat weapons and it has neither of the main attributes that contemporary missiles use to overcome ships’ defences—supersonic speed and stealth. The integrated investment program suggests that Defence will acquire a more capable weapon for the Hunters, so it’s possible that it could be installed on the Anzacs. The Seahawk Romeo helicopter provides a useful capability against swarming attacks by small vessels.

Strangely for the navy’s main anti-submarine warfare ship, the Anzac has mixed capabilities in this regard. The Anzacs were fitted for a towed array sonar but they never received it, and there’s no indication in the integrated investment plan that Defence intends to install such a system. Somewhat ironically, the air warfare destroyers do have a towed array, but there are only three of those ships. Again, the situation won’t be significantly improved until the Hunters arrive with their own towed arrays. This shortfall is partially offset by each Anzac’s Romeo, which has a capable ‘active dipping’ sonar. Two helicopters operating in coordination with dipping sonars can make life difficult for submarines. Unfortunately, each Anzac has only one hangar and consequently one helicopter.

The navy has been quietly but effectively progressing its ‘crawl, walk, run’ strategy for the adoption of uninhabited aerial systems. Since some smaller drones could fit alongside a Romeo in the Anzac’s single hanger, they offer the possibility of enhancing the ship’s aviation capability, in particular by providing greater endurance for surface surveillance. But without a second hangar or mission bay, the potential to integrate autonomous systems without giving up the Romeo that is central to the ship’s anti-submarine capability is limited.

The class has recently, or is currently undergoing, upgrades to other systems such as the electronic support and communications suite.

So, the good news is that overall the Anzacs compare well with other small frigates. The bad news is that it will be challenging to keep the capability relevant against likely threats for the next 24 years.

The upgrades have increased the ship’s weight from its original 3,600 tonnes to 3,900. It probably can’t get any heavier, and the air-search radar replacement may be the last substantial physical alteration made to the vessels. A towed array sonar could be an upgrade too far, even if funds were available. It probably won’t be possible to deploy new solutions like lasers on the Anzacs.

The ships’ upgrades have cost over $2.2 billion, though across a fleet of eight that looks like pretty good value for money given that each Hunter-class vessel will cost at least that much. But finding funds for further upgrades could be difficult as the future frigates and submarines absorb more of Defence’s capital budget.

Much of the cost of defence projects is in design and engineering, which remains the same regardless of whether you buy or upgrade one, eight or 108 systems. So as the Anzacs approach retirement, the value for money of a final round of upgrades for a shrinking subset of the fleet may start to look questionable. The last Anzacs may be left to serve out their time with many of the same systems they have now. Such is the challenge with extended transitions.

That trajectory may amount to a relatively graceful degradation, however. The substantial commonalities between the Anzac and Hunter classes, in their radars, combat management systems, missiles and helicopters, will help. The government’s decision to mandate the Saab 9LV combat management system across the navy’s entire surface fleet, for example, means that improvements to it will be fed back into the Anzacs. Similarly, as the potential of CEA’s radars to conduct tasks such as electronic warfare and secure communication is realised, they can also be used to enhance the Anzac.

But the issue of quantity may be more problematic than the issue of quality. I’ll look at that in part 3 of this series.

The rumour du jour in defence circles this week is that the planned date for beginning work on the future frigate program has been slipped a couple years, out to 2022. That’s despite a robust defence of the originally mooted 2020 date by the Minister for Defence Industry as recently as three weeks ago. Normally such a slippage would smack of poor project management and raise the spectre of delayed capability delivery to the ADF. But this seems to be a case of prudent engineering fundamentals reasserting themselves.

And it’s not entirely clear that the Department of Defence or Navy would regard this as a setback. A mid-2014 presentation by Navy shows a 2022 start to the build phase (see slide 4). The enthusiasm for a 2020 start seems to have been largely been driven by the politics surrounding workforce considerations down in the Adelaide shipyard, where the Air Warfare Destroyer (AWD) program is winding down and workers are being laid off.

Navy’s 2015 schematic shows the detailed design period extending to the end of 2021. As I’ll explain below, allowing time for careful design work makes sense. In fact, the 2015 slide assumes that the ‘concept, preliminary and detail design’ phase will extend for 7.5 years. From the outside it’s impossible to tell if that timetable’s on track, but there’d have to be a chance that time’s been lost. So even 2022 might result in a compressed process.

But there doesn’t seem to be much downside to a later kick-off, at least from a capability and project management perspective. For a start, the upgrade program to the Anzac frigates is only just winding up, and those vessels are well placed to sail on for a few years yet. So there’s no imperative from a frigate availability point of view. In fact, it will amortise the cost of the upgrade over a few more years of service, making it a more efficient use of taxpayer’s funds.

The more conspiratorially minded among the defence community suspect that the rumoured slip is an effort to play BAE’s Type 26 frigate more firmly into the mix. It’ll probably have that effect, as the other competitors have representative vessels already at sea while the British contender is only just entering production. But I don’t think that’s the main reason. Instead, it’s likely that the realities of the engineering task ahead are sinking in.

None of the competing ships are ‘off-the-shelf’ fits to Australia’s requirement. In fact, there’s probably no such thing, given our stringent maritime regulatory requirements. In the AWD program, changes to meet local regulations led to many changes in the engineering drawings. (See the ANAO’s 2014 report for more detail.) So there’s substantial work to be done to modify foreign designs even if we don’t insist on Australian-unique changes for capability reasons.

And in this case there are most certainly Australian-unique changes that will require rigorous systems engineering work. The indigenous (and very impressive) CEA phased array radar is formally described as a ‘preference’ for the future frigates, but is mandated to all intents and purposes. Given that Minister Pyne has been in Canada spruiking the system to the RCN, it’s not credible that the RAN will get anything else.

None of the three contenders have ever been fitted with that system, and history shows that the required modifications won’t be straightforward. Data from the ANAO and Defence’s own figures show that ‘Australianising’ systems is rarely easy. The CEA radar has been proven on the Anzac frigates, and so might be termed ‘military-off-the-shelf’ (MOTS). And the competing vessels could also be termed MOTS—if somewhat charitably given the likely regulatory changes noted above. But life in defence systems engineering is rarely that simple. Gumley’s second law tells us that MOTS + MOTS ≠ MOTS.

And there’s still some work to be done to sort out what combat system will be fitted to interface between the ships’ sensors and weapons. The Anzacs have the Saab 9LV system, which is very effective. Retaining the CEA/Saab combination would reduce the amount of integration work required. The competition for the combat system comes from the US-sourced Aegis system, which would have the advantage of commonality with the AWDs and with ballistic missile defence. The downside to Aegis would be yet another step of systems integration. Either way, these aren’t things to be rushed. Working slowly and systematically can actually speed things up later (and save money).

There’s a lot more to say about the Future Frigate program, including the pros and cons of each of the contenders. ASPI will air its views on those over the next few months, including the combat system choice. And James Mugg and I will be back on The Strategist next week with some thoughts about the ways in which the government might handle the delay. (Spoiler: a ballistic missile defence capable fourth AWD wouldn’t be silly in the current international environment.) But the rumoured ‘slip’ of a couple of years—if true—actually gives Australia a better chance of delivering an efficient build program when work finally commences.

Correction: a previous version of this post incorrectly dated the Navy presentation slides as early 2015. The error was the author’s.

Three weeks ago, the finance minister announced the third delay, and the fourth get-well initiative, for the beleaguered Air Warfare Destroyer program. In case you’ve lost track, delays and remediation initiatives were announced in 2011 and 2012, and a ‘reform strategy’ was announced in 2014. The vessels will now be delivered between 30 and 33 months late and cost ‘at least’ $1.2 billion more than planned.

According to the finance minister, ‘these ships are costing $3 billion a ship’ which implies a total project cost of $9 billion. Looking at the budget papers released just prior to the announcement, the pre-existing cost of the project was $7.9 billion. Adding $1.2 billion to $7.9 billion yields $9.1 billion—a 15% cost blow out. That’s bad enough, but those aggregate figures conceal the true state of affairs.

Under the alliance, Defence, Raytheon and ASC share the pain and gain relative to a Target Cost Estimate (TCE) for the alliance part of the project. If the target is met, industry participants receive a pre-determined ‘fee’ to cover profits and corporate overheads. If the project’s under-budget, participants share the unspent funds up to the amount of twice the fee. If the project’s over-budget, additional direct costs are shared by all three participants up to twice the fee. Thereafter, the Commonwealth pays all additional direct costs.

Consistent with the alliance contract, the current approved cost of $7.9 billion has four components:

• $5.1 billion corresponding to the TCE (including a management reserve)
• an undisclosed amount to cover ‘government furnished’ inputs including the Aegis combat system
• the ‘fee’ to be paid to the alliance partners if the TCE is met or improved upon (and a ‘procurement fee’ for Raytheon that’s not at risk)
• a further financial contingency.

Although the latter three amounts haven’t been disclosed, estimates are possible.

While there’s no hard and fast rule for how large contingency should be for a defence project, the Department of Finance suggests 10%-15% for non-developmental defence IT projects. With this as a benchmark, and noting the manifestly developmental components of the AWD project, contingency of around $1 billion seems reasonable—split between the management reserve in the TCE and the contingency held by DMO. Based on commercial margins, the fee for industry participants is probably around 10% of the TCE or $500 million, leaving around $1.6 billion to cover government furnished inputs.

Everything the government says points to cost blowouts in the alliance component of the project. If there are  problems elsewhere they have not been disclosed. This means that the alliance part of the project is in deep trouble. For the overall cost to have blown out by $1.2 billion, the $1.0 billion contingency has been depleted and the ‘at risk’ component of the $500 million originally set aside to pay the industry partners fees has been redirected to cover emergent costs.

Assuming one-third of the contingency was embedded in the TCE, and that Raytheon’s not-at-risk procurement fee was 20% of the overall industry fee, the TCE has been exceeded by an eye-watering $2.3 billion (= $1.2 billion + $0.66 billion + $0.4 billion).

Actually, the situation is probably not that bad. For reasons I’ll explain in a future post, a change to the indexation arrangements for all defence projects back in 2010 gave rise to a funding shortfall that’s embedded into the foregoing estimate. The indexation shortfall—which is beyond the project’s control—could be in the vicinity of $0.5 billion.

But even taking that into account, there’s still been a massive cost blowout, for which the taxpayer will bear 90% of the pain because ASC is government-owned. Every dollar spent directly on the project will continue to be recompensed 100% by the taxpayer, and Raytheon will still receive its ‘procurement fee’ irrespective of performance.

In theory, ‘liquidated damages’ are owed for late delivery. On the basis of the latest reschedule, the Commonwealth could seek damages of around $557 million. Bizarrely, however, under the alliance contract, Defence is liable for 50% of the liquidated damages and the rest is only recoverable against the already forgone ‘fee’. As elsewhere under the alliance, the taxpayer has once again been left carrying the can.

The forgoing analysis raises a number of questions.

First, why did it take so long to understand the extent of the problems? As recently as mid-2014, Defence assessed that there was ‘sufficient budget remaining for the project to complete against the agreed scope’. How can the project’s prospects have deteriorated so dramatically in less than 12 months? Serious questions need to be asked about the level of oversight exercised by Finance (as owner of ASC) and Defence (as both customer and alliance member).

Second, what’s going to motivate the industry partners now that the incentive fee is no more? The Commonwealth has neither carrots nor sticks. Moreover, what assurance do we have that the line between ‘direct costs’ and ‘fee’ has been properly maintained? Ultimately, the demarcation is an accounting exercise that’s open to interpretation.

Although the details of the AWD alliance’s finances aren’t publicly disclosed, we can get a feel for the sorts of big dollars that arise in the shipbuilding sector by looking at Defence’s handling of the future submarine IPT (Integrated Project Team). As explained in Chapter 7 of the 2015-16 ASPI Budget Brief, there are 35 separate contracts at an average annual per-capita cost of $513k for industry participants in the IPT. For example, the services of a combat system sonar engineer cost $778k a year and an industry and business analyst costs $703k. Either the profits and overheads charged by the firms are egregious, or the direct costs being attributed to personnel are egregious—or both. In any case, the payment of such astronomical fees erodes confidence in Defence’s commercial acumen.

Third, and most importantly, how can the taxpayers’ interests be better protected in future domestic shipbuilding projects? According to the Finance Minister, ‘the Government will release an enterprise-level naval shipbuilding plan later this year, which will provide for the long-term future of the Australian naval shipbuilding industry’. Before we double down on domestic shipbuilding, we need something better than the approach taken for the AWD—which, among other shortcomings, made provision for Defence to sue itself.