Insight Economics Pty Ltd

Submarines for Australia

March 2020

Australia’s Future Submarine

Do we need a Plan B?

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Insight Economics Pty Ltd

ACN:141 097 565

ABN: 29 627 712 906

Cover: Four Collins class submarines off HMAS Stirling with the Los Angeles class attack submarine, USS Santa Fe. Photo: Licence: Alamy

Disclaimer: While Insight Economics endeavours to provide reliable analysis and believes the material it presents is accurate, it will not be liable for any claim by any party acting on such information.

! i! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

CONTENTS

Glossary of acronyms iii!

Acknowledgements 4!

Foreword 5!

Executive Summary 7!

Chapter 1 11! Emerging problems with SEA 1000 11! 1.1! Elimination of competition 12! 1.2! Ambitious capability requirement 12! 1.3! Structure of the report 15!

Chapter 2 16! Problems arising from lack of competition 16! 2.1! Excessive cost of design and construction 16! 2.2! Collins class life of type extension 21! 2.3! Sovereign capability – local content and access to IP 24! 2.4! Technical risks 29!

Chapter 3 33! Will the Attack class be fit for purpose? 33! 3.1! What do Australian submarines do? 33! 3.2! Submarine operations ‘up threat’ in the Indo Pacific 34! 3.3! Strategic capability requirement for the Attack class 39! 3.4! Operational and tactical capability requirements 41! 3.5! Would the Attack class be suitable for a different role? 48! 3.6! Will the Attack class be cost-effective? 49! 3.7! Conclusion 49!

Chapter 4 51! The way forward: a risk mitigation strategy 51! 4.1! Collins 2.0 preliminary design study 51! 4.2! Review of submarine technology requirements 56!

! ii! Glossary of acronyms

A2/AD Anti-access and area denial ADF Australian Defence Force AEW&C Airborne early warning and control aircraft AIP Air independent propulsion AO Area of operations ANAO Australian National Audit Office ARPANSA Australian Radiation Protection and Nuclear Safety Agency ASC Australian Submarine Corporation ASW Anti submarine warfare ASuW Anti surface ship warfare AUV Autonomous underwater vehicle AWD Air warfare destroyer ASPI Australian Strategic Policy Institute C3I Command, control, communications and Intelligence CAD/CAM Computer assisted design/computer assisted manufacture CAPEX Capital expenditure CASG Capability, Acquisitions and Sustainment Group, Department of Defence CDRE Commodore CEP Competitive evaluation process CMS Combat management system DCNS Direction des Constructions Navales Services (predecessor of Naval Group) DWP Defence White Paper EEZ Exclusive economic zone FCD Full cycle docking FOC First of class FSM Future submarine IIP Integrated Investment program IP Intellectual property ISR Intelligence, surveillance and reconnaissance IUSS Integrated Undersea Surveillance System LOTE Life of type extension MOTS Military off-the-shelf MPA Maritime patrol aircraft NATO North Atlantic Treaty Organisation ORBAT Order of battle PLA People’s Liberation Army (China) PLAN PLA Navy RAAF Royal Australian Air Force RADM Rear Admiral RAN Royal Australian Navy RN Royal Navy (UK) SF Special forces SLOC Sea lines of communication SOA Speed of advance SSBN Nuclear powered ballistic missile submarine SSK Conventionally powered attack submarine SSN Nuclear powered attack submarine TASS Towed array sonar system T&E Test and evaluation TKMS thyssenkrupp Marine Systems TLAM Tomahawk land attack missile UAV Uninhabited aerial vehicle UUC Usage and upkeep cycle UUV Uninhabited underwater vehicle USN United States Navy VLS Vertical launch system

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Acknowledgements This independent report by Insight Economics has been commissioned by Submarines for Australia, an entity with a website owned and operated by Gary Johnston, founder and CEO of Jaycar Electronics Pty Ltd. Mr Johnston has no commercial interest in the SEA 1000 program or Defence projects more generally. Over the last three years, Mr Johnston has supported significant research by Insight Economics and others into Australia’s future submarine project. We are extremely grateful for Mr Johnston’s sponsorship of our work. We are also keen to acknowledge the contributions from the members of an expert reference group that has developed around this project. These individuals include distinguished Australians with strong relevant experience who have contributed their time and expertise in the national interest. Inter alia, this group includes: • Dr Michael Keating AC, former Secretary of the Department of Prime Minister and Cabinet and head of the Australian Public Service • Professor Hugh White AO, Emeritus Professor of Strategic Studies at the Australian National University and former Deputy Secretary of the Defence Department • Rear Admiral RAN (Retired), Peter Briggs AO, former commanding officer of Oberon class submarines and CO of the Submarine Force, Director of Submarine Warfare and Head of the Submarine Capability Team • Rear Admiral RN (Retired), Chris Stanford CB, former anti-submarine warfare expert and NATO squadron commander • Commodore RAN (Retired), Paul Greenfield AM, former Engineering Officer in the Submarine Force, and a principal of the Coles Review into Collins class sustainment • Commodore RAN (Retired), Terence Roach AM, former commanding officer of two Oberon class submarines, former Director Submarine Policy and Warfare, Director General Naval Warfare and Director General Maritime Development • Dr Hans J Ohff, Visiting Research Fellow at the University of Adelaide, Managing Director and CEO of the Australian Submarine Corporation from 1993 to 2002 • Dr Andrew Davies, Senior Fellow, Australian Strategic Policy Institute. While these individuals have been extremely generous in contributing their expertise, they may not all agree with some of the findings of this report. The responsibility for the material in this report lies with Insight Economics and Submarines for Australia. A large part of Chapter 4 of this report relies heavily on the work done by the retired naval officers of flag rank in the context of the proposal they sent to the Prime Minister in December 2018. This has been of very considerable assistance. Finally, I would also like to acknowledge the contribution made by Senator Rex Patrick who, before he entered Parliament, kindly sent me hard copies of a large number of articles he had written on the future submarine. These were extremely helpful as I began my long and arduous course of education on the subject. Jon Stanford Director, Insight Economics. March 2020 4!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Foreword by Gary Johnston (Submarines for Australia)!

Over a century ago, the Royal Australian Navy acquired its first submarines. Built in the UK, AE1 and AE2 were state-of-the-art platforms, embodying the world’s best contemporary technology. They were powered by diesel engines and lead acid batteries. Their challenging voyage to Australia was undertaken on the surface and mainly under tow. Forty years later, USS Nautilus, the world’s first nuclear-powered submarine, put to sea for the first time. It had a similar revolutionary effect as the entry into service of HMS Dreadnought in 1905. At least at the highest end of offensive operations, diesel submarines were arguably rendered obsolete. In April 2016, 101 years after AE2 entered the Dardanelles on a wartime mission and 62 years after the Nautilus’s maiden voyage, Prime Minister Turnbull announced that the French government-owned Naval Group had been selected to design Australia’s future submarine, now the Attack class. Like AE1, it would have diesel engines and lead acid batteries. If all went well, the first submarine would enter service in 2035. The program will cost an eye-watering $50 billion in constant 2018 prices. Australia already holds the record for the most expensive surface warships of their size ever built with the air- warfare destroyers. Why not see if we can extend our Gold Medal performance to the underwater domain? In December 2019, the US Navy ordered nine Virginia class nuclear powered attack submarines at a contract price of US$22 billion (or around A$33 billion). This was said at the time to be the largest warship building deal ever (watch this space). These will be more than twice the size of Australia’s Attack class, with a significantly more potent offensive capability and unlimited endurance. Over their 30-year life they will never need refuelling. They will all enter service in this decade, the ninth boat being delivered in 2029, when we will still be waiting for the first Attack class boat. When I heard Malcolm Turnbull’s announcement about the future submarine, I literally couldn’t believe it. At the time, the French proposed to convert their nuclear powered Barracuda design to diesel-electric propulsion (now they acknowledge it will be a new design). Together with the elimination of any further competition solely on the basis of an early concept design, everything about the project – particularly the cost, delivery, technology and extremely high risk – suggested this was a dud idea. Simply put, it involved dumbing down a nuclear submarine by removing the whole basis of its superior capability, and then charging at least twice as much for a far less capable submarine. When the first of class is delivered it will likely be obsolete, with the final submarine due to be delivered 100 years after the Nautilus put to sea. When the Super Seasprite helicopter was cancelled after outlaying a cool $1.4 billion in return for not one single helicopter that the Navy could use, I wondered what the Defence department could possibly do for an encore. We soon found out. But were this massive project to fail, the consequences for our national security are on a completely different scale and are simply unthinkable.

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

My first reaction, together with Dick Smith, was to place an advertisement in a national newspaper at significant cost. Neither Defence nor the government took any notice. Clearly this would be a hard nut to crack. I then commissioned Insight Economics to undertake a substantial amount of research and come up with an alternative way forward. Now working with an expert reference group, I am delighted with this new Insight Economics report. It shows at least that if the government takes action now it is not too late to change course. But the most important message I draw from the report is that Australian submarines are required to operate at the highest level of intensity, even in peacetime, in an increasingly contested and congested theatre where four nations deploy nuclear submarines and where the potential adversary is pursuing a strategy of anti-access and area denial. With both the size and capability of the PLA Navy increasing at a prodigious rate, by the time the Attack class are due to enter service the intensity of submarine operations ‘up threat’ will be even greater. Not only will a diesel submarine have a lesser effectiveness in our area of operations than the American nuclear submarines with which we partner, but its lack of stealth while snorting and its low sustainable speed if detected will give rise to an increasing threat to its survivability. In my view, one of the most shameful episodes in our military history occurred in 1941- 42 when we sent brave young Australians, with predictable results, to fight the advanced Japanese Zero fighters in obsolete aircraft. Never again should the nation abrogate its duty of care towards its service men and women in this manner. We are a wealthy country and have a moral obligation to provide ADF personnel with the best possible military platforms when they are sent into harm’s way. I conclude, therefore, that if the government wants to continue undertaking submarine operations at the highest level of intensity, it should acquire a fleet of nuclear-powered attack submarines, complemented by autonomous underwater vehicles. If the government decides it is unwilling or unable to acquire nuclear submarines, it should consider withdrawing from operations of this nature. But submarines are an offensive weapons system and currently provide the ADF’s only substantial power projection capability. If they can no longer be deployed in this capacity, the question then arises of whether we should continue to operate submarines at all. With Australia’s vast coastline to defend, there is little value in deploying a few conventional submarines to chug around the archipelago. The answer may well be that two squadrons of advanced bomber aircraft would provide a superior offensive capability, delivered much earlier than the future submarine at a considerably lower cost. But perhaps in light of the significant ongoing deterioration in Australia’s strategic circumstances, in the future we will need both nuclear submarines and long-range stealth bomber aircraft to provide a credible deterrent to any potential adversary. Finally, I recognise the challenges involved in enhancing Australia’s capabilities in nuclear science. Should the government decide to embark on the long and difficult road to acquiring SSNs, I will commit to endowing a Chair in nuclear engineering in an Australian university.

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Executive Summary*

The selection in April 2016 of the French government-owned company Naval Group as Australia’s design partner for the Navy’s future submarine was the outcome of a novel competitive evaluation process (CEP) between French, German and Japanese companies. The Turnbull government accepted Defence’s recommendation of Naval Group despite the fact that Ministers did not know how much the submarine would cost, its level of Australian industry content (AIC) and, with a very late delivery, the implications of a capability gap. The government rejected an attractive German offer – fixed price of $20bn; 12 submarines to be built locally in a new digital shipyard at same cost as in Kiel; 70 per cent local content; and delivery in 2028. A major difficulty was that with Defence intending to work in an integrated way with Naval Group on the submarine design, the SEA 1000 team felt unable to manage any further competitive process. One lesson from previous acquisitions was the need to invite two or more proponents to prepare preliminary designs and then provide a tender covering price, delivery and AIC. Instead, further competition was eliminated. Naval Group were gifted a monopoly position on the basis of only a very early concept design for the Attack class and with no commitments on cost, schedule and Australian industry content. The second problem was around capability. The Defence department was attracted by the capability offered by the French during the CEP, but perhaps took insufficient account of the risk that it could not all be delivered. As occurred with the original combat system for the Collins class, it is possible that Defence’s ambitions for the submarine’s capability are inconsistent with the limits of contemporary technologies. A previous Defence Minister had stated that the Navy needed a conventional submarine (SSK) with the capability of a nuclear submarine (SSN) – effectively an oxymoron. The French offered the RAN just that, a diesel-electric platform based on their Suffren SSN. Yet the French concept was the riskiest possible – an ab initio design with an extended delivery schedule, with the first Attack class submarine due to enter service after the last Collins boat reached the end of its design life. By emphasising capability over both cost and delivery, Defence may have forgotten the maxim: “the last 10 per cent of the performance sought generates one-third of the cost and two-thirds of the problems”. In a recent report, the Auditor General noted the high risks in the SEA 1000 program even at this very early stage. The ANAO also revealed that the government-appointed Naval Shipbuilding Advisory Board, including US Admirals who had previously endorsed the French proposal, advised the government to consider terminating the arrangement with Naval Group. In March 2020, a member of that Board told the Senate that the main problem with the program was that it lacked a contingency ‘Plan B’.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! *No!references!are!provided!in!the!Summary.!These!can!be!found!in!the!body!of!the!report.! 7!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Emerging problems

Excessive cost The government should be deeply concerned at the cost of its naval acquisitions. The air warfare destroyers, for example, are the most expensive warships of their size ever built and cost nearly three times the benchmark global price. The budget for the Hunter class frigates suggests that RAN will pay at least twice as much as the Royal Navy for those ships. Even before a preliminary design has been produced, the cost of the Attack program has escalated from an already extremely high level: • The design of the submarines will cost at least $4bn, compared to the ~$1bn that is a contemporary benchmark for complex submarine designs • Since the announcement, the acquisition budget has been increased by 60 per cent to $80bn in out-turned dollars (confirmed in the recent ANAO report) – an outlandish cost for a conventional submarine • Defence has confirmed that the whole of life cost for the 12 submarines will be $225bn out-turned, compared to ~$90bn out-turned for the TKMS bid. Inadequate Australian industry content After experience with the cost of maintaining Oberon submarines and the unreliability of supply from UK, the principal reason for building submarines in Australia is to establish a sovereign capability to sustain them. Sustainment of Collins is now achieved very efficiently with ~ 90 per cent AIC. With Naval Group having an economic and political incentive to maximise French content, the outcome in terms of AIC is unlikely to be satisfactory. The first CEO of Naval Group Australia promised 90 per cent AIC, which Minister Pyne said he would ensure was a commitment. In February 2020, Naval Group flagged less than 50 per cent Australian content, possibly much less if the overseas content in work by local subsidiaries of French companies is taken into account. The recent commitment to 60 per cent is not yet contracted and, with a definition of local content that includes Australian resorts and other services, may not meet the requirement to ensure the advanced industrial capability required for future submarine sustainment.

Dangerous capability gap The Collins class reaches the end of its design life between 2026 and 2034. The first Attack class submarine is expected to be operational by 2035-36. This implies a very substantial capability gap. Since ab initio submarine projects almost inevitably run late or fail, it is possible that the first Attack will not enter service until 2040. For the first time, Defence now acknowledges the possibility of a capability gap. Beyond the obvious impact on national security, the implications of a gap in submarine capability are far reaching. In particular, the effect on the availability of trained submarine personnel can persist for a long time. Due to an early lack of operational availability of the Collins class, the transition from the Oberons was a fraught experience in regard to the impact on recruitment and retention, from which, over 15 years later, the Submarine Force has only recently recovered. It is inevitable that all six Collins submarines will need a life extension (LOTE). This will be a high risk exercise, including replacement of diesels, the main motor and other major systems, with estimates of cost ranging between $6 and $15bn. The ANAO report 8!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? states that planning for this is at an early stage. No budget for the LOTE separate from that of SEA 1000 has been announced. Taken together with an increase in activity in Attack class construction at the same time, in the late 2020s this will make a major impact on Defence’s Integrated Investment Program (IIP). The plan is also to undertake the LOTE within the 10-year full cycle docking major maintenance schedule that take two years to complete. There is a high risk that the LOTE will not be able to be completed within this timeframe. This would exacerbate the likely capability gap. According to industry experts, the most efficient way of incorporating the installation of new propulsion systems would be to build a replacement hull section for the after end of the submarines. This would require expertise beyond repair and maintenance, including in design and construction. The high risks in this suggest that Saab Kockums, the designer and builder of Collins, should be engaged by the Design Authority, ASC, to assist in what will be a very complex task.

Low cost-effectiveness and questionable strategic value Apart from the vital considerations of cost, delivery, local content and the risk of a capability gap, we have a more fundamental concern in regard to the Attack class. The main role for RAN submarines is to operate with the US Navy in intelligence gathering missions in the South China Sea, 3,500 NM from their base. The long distance to their primary area of operations is a major challenge – no other conventional submarine in the world undertakes similar operations – and means that nearly half of a RAN submarine’s 70-day operation is spent in transit. Even with 12 submarines by 2050, in peacetime generally only one submarine will be able to be on station at any time. By the mid-2030s, half the world’s submarines will be operating in the South China Sea, many of them nuclear powered. The cost-effectiveness of outlaying $225 billion in whole of life costs in order to deploy one conventional submarine to such a congested theatre at any time must be in doubt. With the ongoing increase in the intensity of undersea operations in the South China Sea, the operational effectiveness and survivability of conventional submarines become ever more challenged. Developing its anti-submarine capability is now a high priority for the PLA Navy, with the aim of creating an ‘underwater great wall’ and denying access to foreign submarines. The first tactical problem for a diesel submarine is the need to ‘snort’ near the surface every few days so as to recharge its batteries. This makes them far more liable to detection. The second tactical problem is their inability to sustain a high speed for any length of time. This limits their ability to track nuclear submarines and other high value targets. Together with limited endurance, it also compromises their ability to break contact if detected and escape at high speed, thereby reducing survivability. The operations undertaken by RAN submarines can be at the highest end of the intensity scale, similar to NATO operations during the Cold War over 30 years ago where ultimately all the parties used SSNs. As technologies continue to develop, intensity is only likely to increase over time. Taking account also of the very long transits for Australian submarines, we do not believe that a conventionally powered submarine will be fit for the purpose of operating in this theatre in the 2030s and beyond. Compared to the SSNs operated by the US Navy, a SSK will not only be less effective but it will also have a lower level of survivability.

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Proposed risk mitigation strategy: Plan B In responding to the ANAO report, Defence acknowledged, at the highest level, there was a need to mitigate the very high level of risk in the SEA 1000 program.

Re-introducing competition Our proposal for mitigating the risks around SEA 1000 would not involve delays to the present program with the Attack class, and a very small increase in cost that we are confident would be recouped many times over. Fundamentally, it would introduce a parallel pathway that would give rise to valuable competitive pressure. The alternative pathway is for an evolved Collins class submarine – Collins 2.0. This recognises that evolving a new submarine from an existing successful platform – and the Collins class is now performing extremely well – is the approach followed by all other countries that build and operate submarines. The main advantage of this approach, which the Americans advised Defence to follow in the first place, is that it involves a significantly lower risk. But there are also benefits in terms of lower costs and a shorter delivery schedule, as well as being able to utilise an existing supply chain. On the basis of expert professional advice, we consider that an evolved Collins 2.0 submarine, with a comparable capability to Attack, could be delivered at least five years earlier, at a much lower cost and with 70 per cent local content. There would also be benefits from significant synergies with the Collins class LOTE. We propose that in parallel with the Attack class program, the government should commission a preliminary design of a Collins 2.0 option. The estimated cost of the design is less than 0.2 per cent of the future submarine budget, money that would be recovered many times over even if only by the effects of eliminating the current monopoly. In 2022-23, the government could then choose between the Attack class and Collins 2.0 on the basis of fixed price tenders for the first six submarines, also including firm proposals for delivery, AIC and intellectual property rights. This task could be undertaken by an independent expert review team reporting to the Minister for Defence through the CDF and the Secretary of Defence.

Review of submarine technologies Operations ‘up threat’ in an offensive, power projection role using conventional submarines will be highly challenging by the 2030s. Only nuclear powered submarines have limitless dived endurance on an operation and the speed required to put more submarines on station – perhaps three times as many as with diesel boats – in the face of RAN submarines’ long transits. Their sustainable high speed also offers major advantages in terms of operational effectiveness and survivability. We recognise it would be difficult for Australia to acquire SSNs, however, and would take 15-20 years. The Navy would require 12 conventional submarines first (including six life extended Collins boats) in order to build up numbers of trained personnel. The 2016 Defence White Paper flagged a review of future submarine technologies in the late 2020s. We believe this review should be brought forward and undertaken as a matter of urgency. It should include an analysis of whether Australia should explore the option of acquiring nuclear-powered submarines. The future role of autonomous underwater vehicles (AUVs) should also be investigated in detail.

10!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Chapter 1

Emerging problems with SEA 1000

In April 2016, the Turnbull government announced that the French company Naval Group (then DCNS) had been selected as Australia’s strategic partner to design the Royal Australian Navy’s future submarine. Naval Group is majority owned by the Government of France with major French defence companies as minority shareholders. Now, nearly four years after the announcement, some significant problems are beginning to emerge, as detailed in a report by the Auditor General released in January 2020.1 These include: • A lengthy delay in concluding the Strategic Partnership Agreement (SPA) between the Australian and French governments • A 60 per cent increase in the acquisition budget, confirmed by the ANAO, for the submarine to $80 billion in out-turned (inflated) dollars • A delay of nine months in the completion of the preliminary design so that the project has already missed two milestones • A delivery schedule that requires all six Collins class submarines to have a ten- year life extension (LOTE), but with little preparation having been made for the LOTE even though the need for it was understood years ago • A poor relationship has developed between Naval Group and Defence due to cultural differences and disagreements about Australian industry content • Very high risks around the SEA 1000 project, as also acknowledged by the Defence Department • Before the SPA was agreed, the government was advised by the Naval Shipbuilding Advisory Board, which included three retired American admirals, that the risks were so high that Australia should consider walking away from the project even were the SPA to be concluded satisfactorily. Abstracting from the individual problems around issues such as cost and risk, there are two main categories of issues that require the government to intervene and establish a ‘Plan B’. The first problem, that underlies most of the current issues between the parties, is the premature elimination of any competition long before a preliminary design has been completed, and the serious issues that are arising as the consequence of the gifting of monopoly power to Naval Group. The second is a much more fundamental problem that will be even more difficult to address. This is the likely pursuit of an unrealistic capability requirement for the Attack class in light of the unique role that Australia requires its conventional submarines to undertake. These two issues are discussed below.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 1!ANAO$(2020),$Future&Submarine&Program&–&Transition&to&Design,$Auditor0General$Report$No.$22$2019020,$para.11,$ https://www.anao.gov.au/sites/default/files/Auditor0General_Report_201902020_22.pdf! 11!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

1.1 Elimination of competition

The elimination of competition beyond the concept design stage is giving rise to problems that any economist would expect when a large contractor is gifted with a monopoly position. The competitive evaluation process (CEP) that led to this decision was controversial. In the largest ever Australian capital project, for the government to commit to a single partner on the basis of a simple concept design seemed risky. It went against all the lessons learned from previous unsuccessful acquisitions about maintaining competitive pressure for as long as possible. Without negotiating any bankable commitments while having stated that the budget for the project was $50 billion, the government had gifted Naval Group with a monopoly position. In doing so it has severely compromised its negotiating position on every aspect of the project – cost, delivery, Australian industry content and rights to intellectual property. Even if the government was determined to down select to a single company so very early in the program, this need not have occurred. Although ultimately unsuccessful, the German group, thyssenkrupp Marine Systems (TKMS), had made a firm lump sum offer. They proposed a fixed price of $20 billion for building 12 submarines, with the first being delivered in 2028 with 70 per cent Australian industry content. They would all be built in Adelaide at the same cost as in Kiel. This was in accord with the then best practice global cost of a conventional submarine of about $400,000 to $450,000 per tonne.2 This was not a will of the wisp proposal. The German principals needed to be convinced that Australian industry could achieve this while delivering an acceptable profit and at the same time maintaining a contingency margin. We are not arguing that the German offer should necessarily have been accepted, but it did establish a strong benchmark for acquiring a submarine of the same size as the French design with a similar need to satisfy Defence’s ambitious capability requirement. As was their right, the government selected the French on the basis of Defence’s recommendation that the capability offered by Naval Group’s concept design was markedly superior. But Ministers could at least have used the German proposal as a benchmark and made acceptance of the Naval Group proposal contingent on the negotiation of a comparable agreement, no doubt with some flexibility included. They could also have nailed down the degree that Naval Group would guarantee that the highly advanced capability they offered could be delivered. Indeed, it is surprising that the three coordinating departments and their Ministers (Prime Minister and Cabinet, Treasury and Finance) did not insist on a much tighter approach along the lines suggested.

1.2 Ambitious capability requirement

The greatest underlying problem with the Attack class program is that the evidence suggests that Defence is seeking to develop an unprecedented level of capability in a conventionally powered submarine (SSK). After the RAN began operating submarines again in the late 1970s, it was only later on that an important role for the Oberon class became to undertake intelligence, surveillance and reconnaissance (ISR) operations in waters far to Australia’s north. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2!Unsurprisingly,!since!the!estimates!were!nonGbinding,!all!contenders!submitted!a!rough!order!of!magnitude! cost!within!10!per!cent!of!each!other.! 12!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

These operations involved very long transits. But they were conducted mainly on the surface. This was commonplace for conventional submarines at that time operating in a strategic environment with a low level of intensity and a fairly rudimentary anti- submarine warfare (ASW) capability on the part of potential adversaries. As submarine detection technologies improved and satellites became more numerous, operations undertaken mainly on the surface became less tenable. The Collins class submarines that replaced the Oberons with a similar role were designed to operate dived throughout. In order to maintain the same speed of advance (SOA) as the Oberons in transit, this required frequent recourse to snorting, close to the surface, so as to recharge batteries. On a 3,500nm transit, frequent snorting imposes a degree of stress both on a submarine’s crew and on its machinery. It also exposes the submarine to a significantly greater risk of detection and interdiction, albeit a lower risk than when travelling on the surface. At the same time, high intensity blue water submarine operations in other parts of the world, particularly the Atlantic theatre, increasingly became the preserve of nuclear- powered attack submarines (SSNs). These operations were highly intense during the Cold War, entering the dangerous ‘grey zone’ between peace and war. Both their speed and their ability to remain fully dived throughout for a 90-day operation gave SSNs a higher level of effectiveness. High speed and virtually unlimited dived endurance also provided SSNs with a higher level of survivability because of their ability to break contact when detected and escape at very high speed. By the end of the last century, the US, Britain and France had discarded all their conventional submarines. Because only six countries currently deploy SSNs – the US, Russia, China, Britain, France and India – these blue water, high intensity operations ‘up threat’ now are effectively limited to great powers – with a few exceptions. Countries that operate conventional submarines generally undertake short operations – perhaps of around 21 days maximum – and operate dived throughout, with a minimal need to snort due to their use of air-independent propulsion (AIP) systems and/or modern batteries with a much greater level of energy intensity. Japanese Soryu class submarines, for example, undertake covert ISR operations in liaison with the US Submarine Force in the East China Sea, but we understand these are generally of two weeks duration undertaken fully dived throughout. Their use of AIP, or more recently Li-Ion batteries, allows them to achieve this without the need to snort. Australia is the only country that requires its conventional submarines to undertake 70- day operations involving 3,500nm transits and then a long patrol of 30+ days far from base. When they reach their primary area of operations (AO) they also work in an extremely high intensity operational and tactical environment. Given that Australian governments from both sides of politics have rejected the use of nuclear propulsion for Australian submarines while still seeking to undertake the same, increasingly challenging, operations as in the past, this gives rise to a capability requirement for a unique conventional submarine. The then Defence Minister, David Johnston, said in 2015 that Australia required a submarine with the capability of a nuclear powered boat but with diesel-electric propulsion. Given that a SSN derives its capability from the advantages in terms of speed and endurance offered by a nuclear reactor, this is an oxymoron. It is probable

13!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? that the RAN’s capability requirement may well test the boundaries of what conventional submarines can reasonably be expected to do. We believe the desire to design and build what is essentially a unicorn lies at the heart of most of what is wrong with the Attack class program. The major issue is that the Naval Group won the SEA 1000 CEP not on the basis of cost, delivery or local content but rather on the claimed superior capability it offered. But just as the CEP failed to lock in any commitments on cost, schedule and local content, it also failed to lock in commitments around capability. The decision by the German group TKMS to offer a fixed price well below the budget for the future submarine appeared a smart move at the time. But in retrospect it may have been an own goal. It meant that unlike their rivals, they couldn’t afford to offer capability that would be risky and perhaps impossible to deliver. But for other proponents who had made no commitments around price or delivery, the sky was the limit. Defence’s statements about the superiority of the capability offered by the French suggest they apparently accepted all Naval Group’s claims without undertaking a serious risk assessment of whether the capability could actually be delivered. This can be illustrated by the much spruiked proposal by Naval Group to use a pump-jet propulsor on the Attack class. Naval Group said at the time of the announcement that pump-jets had rendered propellers ‘obsolete’ for submarines. Yet the physics of pump- jets tell us that while they can provide a superior acoustic solution for power-rich SSNs, they are too inefficient for use on conventional submarines. Indeed, four years later it appears that none of Naval Group’s designs for SSKs have adopted pump jets. More tellingly, within 18 months of winning the contest, the French head of the Australian program suggested that propellers might be used in the Attack class after all. This drew a response by Defence insisting on pump-jet propulsion as demonstrated during the CEP process. The second straw in the wind regarding the onerous nature of the capability requirement is how the design schedule is blowing out, with Naval Group already requesting a delay of 15 months in the preliminary design. If there is a disconnect between the design partners on what level of capability can be expected to be delivered, this substantially increases the risks around the project. A former Secretary of the US Navy, Norman Augustine, proposed what he called ‘the law of insatiable appetites’ in defence procurement. There was a strong basis in past experience to support Augustine’s proposition that “the last 10 per cent of the performance sought generates one-third of the cost and two-thirds of the problems”. In Australia’s case, we may be looking at an even higher level than 10 per cent of the required capability that is at risk. With the exception of local content, this fundamental tension between the Navy’s required capability and what is presently achievable dominates every issue around the project. Perhaps most importantly, it explains the lack of competition, which underlies current issues around cost, delivery, local content and IP. Defence stated at the outset that it only had the resources to manage one project proposal. This may well be correct, although in any normal project it shouldn’t be. The reason is that in selecting a design partner, they meant exactly what they said; Defence is the other partner in designing

14!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? the submarine and a very active one. Early in the design process, there were more Australian designers than French on the team. Defence, has been involved in designing an ideal submarine since before the 2009 Defence White Paper (DWP). It intends to make a major contribution to the design with Naval Group. This, no doubt, is why Naval Group publicly refers to Defence as a very demanding customer. Also, if Defence’s requirements for capability are regularly extending further into the wish list than was understood at the time of the CEP, this is also probably why the design has encountered a significant delay. There is no way Defence could manage two competing design proposals and run a neutral competition when it is working on a complex design in an embedded way with a single partner. But this also reflects fundamental governance issues, when the customer is also part of the team designing the submarine in liaison with the Prime contractor. Should the design fail, this fundamental conflict could well be a major cause. But then the same question arises as with the air warfare destroyers where all the parties to the partnership arrangement seemed to be in bed together. If it the project fails, who does the Australian government sue? This also reflects another problem – the equivalent of “don’t ever let a fighter pilot design the air force’s next fighter”. Without Ministers driving the project, the emphasis is likely to be on capability with a lesser consideration for cost, schedule, local content, ownership of IP or, particularly, deliverability. This is always inappropriate. It is particularly inappropriate when Australia’s increasingly threatening strategic circumstances suggest we need new submarine capability as soon as possible and when there is a financial constraint in terms of the various new programs, across all three services, that are making growing demands on the defence budget.

1.3 Structure of the report

The remainder of the report is structured as follows:

• In Chapter 2 we examine problems around cost, delivery, AIC and IP protocols and discuss how these have been exacerbated by the lack of competition in the process

• The significant challenges around what Australian submarines do and what technologies are required to support their role are discussed in Chapter 3

• A proposed ‘Plan B’ risk mitigation strategy is presented in Chapter 4. !

15!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Chapter 2

Problems arising from lack of competition

The Competitive Evaluation Process (CEP) to select a design partner for the FSM was a novel initiative for the Defence department. Given that it effectively led to the selection of a platform, based only on a design concept, a major source of risk source of risk to the FSM project was the government’s decision to eliminate any further competition following the selection of Naval Group as design partner. As the Auditor-General pointed out in his initial (2017) audit of the CEP for the FSM, this is a high-risk program. The implications of compounding these risks by eliminating competition were identified clearly by the ANAO:

The approach taken by Defence for the Future Submarine program removes competition in the design phase, and removes incentives for the international partner (DCNS) to produce a more economical and efficient build. This places the onus on Defence to ensure that its approach to the Future Submarine’s design and build phases, where final costs and schedules will be determined, returns value-for- money to the Commonwealth in the absence of a competitive process.3 We are especially concerned that the elimination of competition at a very early stage of this project exposes the government and the taxpayer to a number of very substantial risks and a contingent liability that is enormous both in financial terms and also in respect of the delivery of what is a vital defence capability. Having been gifted a monopoly position at the very start of the project, a rational approach by Naval Group would be to maximise the commercial benefit it will derive from the project. We can point to some emerging evidence, even at this very early stage, that significant problems are becoming apparent as a result of abandoning competition so early in the acquisition process. We discuss below the emerging effects of lack of competition on risks around: • The cost of designing and building the Attack class • The timely delivery of the submarine and the capability gap • Australian industry participation and sovereign capability • Australian access to and ownership of critical intellectual property • Technical specifications of the Attack class. These are considered below.

2.1 Excessive cost of design and construction

Among several issues apparent with SEA1000, excessive cost is perhaps the most obvious. It has serious and broad implications. The budget breakdown has not been explained but it appears to us to be far above what we should pay for the capability sought. The problem of embarking on major acquisition programs much too late and then paying much too much for them is not confined to the SEA1000 program. Australia paid between two and three times the benchmark price for the AWD and we look like paying much more for the frigates than we should. This is unacceptable,

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 3$ANAO$(2017),$Future&Submarine&—&Competitive&Evaluation&Process,$Performance$Audit$Report,$27$April,$para$3.31,$ https://www.anao.gov.au/work/performance0audit/future0submarine0competitive0evaluation0process$! 16!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? particularly after the successful and cost-effective Anzac frigate program. We note too that the first AWD was delivered 16 years after the last ship of the class it was replacing was decommissioned.

Design cost and risks Currently, approved expenditure on SEA 1000 is $5,959 million, including outlays to date of $779 million. Defence had spent $396 million on design to 30 September 2019.4 But in addition, the Director of SEA 1000 told the Senate that the budget to complete the design process would be ‘in the order of $4.5 billion’.5 Our research, including discussions with Australian shipbuilders, suggests that a complete design of a contemporary submarine should not cost more than $1 billion. Admittedly, however, this cost represents what you would expect to pay for an evolved design, which is what most submarine builders around the world seek to undertake because of the much lower risk. In many cases, around 60 to 70 per cent of systems used in these designs are already proven in existing submarines. If indeed the designers of the Attack class are starting with a clean sheet of paper, however, it seems clear that the process would be more resource-intensive and take a much longer time. It would also be much more risky. This is another significant problem with selecting an ab initio design. It now appears that the Shortfin Barracuda, initially presented as a diesel-electric version of the French Barracuda class SSN and then an evolved version of the Scorpène (of which details of its performance were leaked to the world’s media), is in fact rather more of a “new” design than were the other two contenders. Even so, to pay $5 billion for designing a SSK seems excessive. Given that Defence told us that the rough order of magnitude project price presented by the three contenders were fairly close, it seems extraordinary that design costs could account for around 25 per cent of the overall budget for a program to deliver 12 very large submarines. Perhaps the simple answer to these imponderables, as one retired senior naval officer told us in regard to the proposal to use pump-jet propulsion, is that “our guys just had the wool pulled over their eyes”. Another major risk in terms of the design task concerns the possibility that the specifications may be subject to significant change. As discussed later in this submission, the current design excludes air independent propulsion (AIP) and modern batteries as well as including power-hungry pump-jet propulsion. Should the requirements change in regard to any of these features, perhaps as a result of land- based testing, a very substantial re-design would be required mainly because of significant changes is weight distribution. This could give rise to a considerable delay. In addition, with lead acid batteries embodied in the first ‘flight’ of three or four submarines, it is almost inevitable that much more energy-dense batteries would be incorporated in the second flight. This would also require a major change in the design. For example, lead acid batteries play a second role in providing the submarine with the required ballast, but lithium-based batteries are much lighter. If, however, a different technology such as the emerging nickel zinc design, were to be adopted, this would

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 4$ANAO$(2020),$Future&Submarine&Program&–&Transition&to&Design,$Auditor0General$Report$No.$22$2019020,$para.11,$ https://www.anao.gov.au/sites/default/files/Auditor0General_Report_201902020_22.pdf$ 5!Hellyer,!Marcus!(2019),!“The!cost!of!defence,!ASPI’s!Defence!budget!brief,!2019G20”,!ASPI,!June,!pages!82G83,! https://s3GapGsoutheastG2.amazonaws.com/adGaspi/2019G05/Cost%20of%20Defence%202019G2020.pdf! 17!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? simplify things if only because the weight of a nickel zinc battery is similar to lead-acid, although the energy density is much greater. In a report released in January 2020, the Auditor-General found that Naval Group had requested an extension of 15 months for the preliminary design process. In response, Defence agreed to a nine months extension:

The program is currently experiencing a nine-month delay in the design phase against Defence’s pre-design contract estimates, and two major contracted milestones were extended. As a result, Defence cannot demonstrate that its expenditure of $396 million on design of the Future Submarine has been fully effective in achieving the program’s two major design milestones to date.6

This delay before even the preliminary design has been completed is a major concern. It suggests some early difficulties, perhaps with delivering aspects of the advanced capability that Defence requires.

Production cost Under the CEP, each contender was required to provide a non-binding, rough order of magnitude of their submarine’s overall cost. We understand these came in within about ten per cent of each other – not surprisingly because the estimates were non-binding and the three contenders, as well as Defence, would all have known the approximate benchmark cost per tonne of building a submarine. Although TKMS publicly offered a guaranteed fixed price, the others may have pitched at a similar level, being confident they could quote a much higher price later if successful when monopoly status had been achieved. A rational monopolist would be loath to leave money on the table, particularly in a rare instance where the customer had thoughtfully published a program budget that would have appeared generous to the point of extravagance. When the initial budget for the FSM of $50 billion (excluding the combat system) was announced, it caused many eyebrows to be raised. Compared to what other nations spent on acquiring conventional submarines, this seemed excessive, even after accounting for the large size of Australia’s requirement. Yet that budget was framed in terms of what Defence calls future or out-turned dollars, meaning that it allows for inflation at an assumed annual rate of 2.5 per cent until the completion of the program in the mid- 2050s. In June 2018, however, Defence announced that while the headline figure for the Attack class budget remained the same, it was now expressed in constant dollars. This implied that the previous $50 billion in out-turned dollars had become $80 billion overnight. Only a couple of years after the decision on the CEP and before even a concept design of the submarine had been produced, the budget for SEA 1000 had increased by nearly 60 per cent. While the then Minister denied there had been any increase – “it was always $50 billion” – the recent ANAO report confirmed the increase from $50bn to $80bn in out-turned dollars.7 In evaluating whether the submarines will provide value for money, the key question is: how much should submarines cost? This is a challenging exercise because comparable and verifiable data are difficult to find. Credible analysis undertaken over several years in Australia by people closely involved in the submarine enterprise has produced

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 6$ANAO$(2020),$op.&cit.,¶.$11.$ 7!Ibid,&page!7.& 18!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? estimates of a global western benchmark cost for a conventional submarine. 8 Despite all the hype and misinformation at the time, this study found that the average cost per submarine for the Collins class was just under the contemporary benchmark. This was quite a remarkable achievement for submarines built by an inexperienced work force in a new shipyard built on a green field site and in a country that had never built submarines before. On the same basis, we are advised that the global benchmark cost per tonne at the time of the CEP was around $450,000 per tonne. By 2018, this had risen to $500,000/tonne.9 Using this parametric, we could accept that the average cost of the Attack class at around 4,270 tonnes (the publicly available surfaced displacement) should be in the order of $2,135 billion per submarine, This suggests an acquisition cost for 12 submarines should cost $25.62 billion. Defence has stated that the budget for SEA 1000 is $50 billion in constant 2018 dollars (equivalent to $80 billion in out-turned dollars). Other statements lead us to understand that this does not include the combat system (an additional $6 billion), the cost of extending the life of the Collins class submarines or sustainment costs for the Attack class. What then is the remaining $24.38b buying? Design fees, project management costs, spare parts, integrated logistic support, new infrastructure including a shipyard, training and education systems such as the Naval Shipbuilding College, weapons, provision for project contingency and other foreseeable items would all quite reasonably be a part of the $50 billion total. But based on what those items should cost, it’s very hard to see anything like another $24.38b. We note too that the unsuccessful German proposal was to build 12 Type 216 submarines in Australia with a high local content for a fixed price not to exceed $20 billion, or a maximum of $1,667 million per boat at 2016 prices. At 4,200 tonnes surfaced displacement, this amounts to $400,000 per tonne, equivalent to eleven per cent below the estimated global benchmark cost at that time. Although there is insufficient information available on which to make a definitive judgement about the difference between that bid and the current SEA 1000 budget, the size of the difference must raise very serious questions. Even if we allow $10 billion for the ancillary items identified above, which seems generous, at $40 billion the Attack class would still be twice as expensive as the German Type 216. In the absence of any official public breakdown of the budget for SEA 1000, this analysis suggests that we are at risk of paying a great deal more for the Attack class than we should. If SEA 1000 is costing even a few billion dollars more than we should pay, the opportunity cost to the rest of the Defence IIP and to national security are very serious indeed. The problem is that with Naval Group knowing the size of the budget, it gives them a free hit. Like any rational monopolist, they will set a price consistent with the budget, or may even ask for more. Already the cost of the design is excessive and in a couple of years or so when the construction contract will need to be signed, Defence will have

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 8!Pacey,!Brice!(2012),!Sub&Judice:&Australia’s&Future&Submarine,!Kokoda!Papers,!No.!17,!January,!page!50,! https://www.regionalsecurity.org.au/Resources/Documents/KP17.pdf.!! 9!More!recent!estimates!are!based!on!correspondence!with!CDRE!(Rtd)!Paul!Greenfield!RAN,!who!contributed! to!the!Kokoda!report.!The!estimates!are!expressed!in!constant!2018!dollars,!tonnage!being!the!submarine’s! displacement!on!the!surface,!not!when!dived.!!This!cost!benchmark!figure!includes!the!combat!system!but!not! the!weapons,!which!are!usually!costed!and!procured!separately.! 19!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? little capacity to resist. As Bergmann suggests in the passage quoted above, “without commercial leverage the Commonwealth will either have to pay whatever the designer wants—or start the entire process again”. When that decision needs to be taken, around six years after the 21016 announcement, the cancellation option would be untenable in terms of the time it would take to ramp up a new design. This situation argues for more than just the scrutiny of a review. Bringing pressure to bear on the prime contractor by re-introducing competition into the project is probably the only way in which the cost of the SEA 1000 project could be reduced to a more acceptable level. Equally critical and important is the absence of any information on funding for the Collins LOTE, which we estimate will cost at least $1 billion per submarine and possibly much more. If the LOTE funding must come from the existing SEA 1000 budget, what is being traded off? If it is new money, this will be an additional opportunity cost to the Defence capital budget, which the government has not suggested will be increased. This could mean that when both the construction of Attack and the LOTE are commencing, annual outlays on the submarines may amount to $2 billion or more. Already, pressures on the annual Defence budget are causing some projects to be delayed. Without a significant increase in the budget, perhaps to 2.5 per cent or even 3 per cent of GDP, some major decisions on spending priorities may be required. It is difficult to see the Army of RAAF agreeing to cut their programs in the IIP because RAN expenditure is higher than expected. It may well be that Navy programs, including SEA 1000, may be reduced or curtailed. Based on discussions with Australian shipbuilders, we also strongly disagree with the RAND group’s assessment that the cost of having naval platforms built in Australia exceeds the global benchmark by over 100 per cent. The RAND study reflected the experience with the only substantial platform being built in Australia at the time RAND undertook its analysis, the Hobart class air warfare destroyers under SEA 4000. There were several particular reasons as to why the acquisition process for those three ships was highly inefficient, to the extent that they are the most expensive warships of their size ever constructed in the world. But the preceding two substantial RAN platforms – Anzac frigates and Collins class submarines – were built at a cost that was roughly on a par with the global benchmark and with a local content probably nearly double that of the Hobart class. In addition, a commercially driven company like TKMS would not have offered the same fixed price to build the Type 216 in Osborne as in Kiel had they not made a very detailed study of the capability and costs of the Australian naval shipbuilding industry. Interestingly, one of the reasons given for rejecting the German bid was that Defence didn’t believe they could deliver the submarines for this price. Never mind that TKMS was offering a fixed price guarantee as well as significant technology transfer with high local content. Never mind that accelerated delivery of the Type 216 was designed to obviate the need for a costly and risky LOTE for the Collins class. Given that Defence recognises the very high risks around the Naval Shipbuilding Plan, any by a contractor to share some of these risks surely should have been seen as a positive.10

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 10!In!reporting!the!German!proposal!for!the!FSM!at!length,!we!are!not!necessarily!endorsing!it.!But!given!the! detail!around!the!bid!and!the!excellent!reputation!the!Germans!have!for!building!technologically!advanced! conventional!submarines!the!value!for!money!on!offer!from!tkMS!appears!to!be!substantially!greater!than!that! 20!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Perhaps the very high risk of paying an excessive price for the future submarine is the strongest of the many arguments for the need for competitive pressure.

2.2 Collins class life of type extension

According to the current delivery schedule, the 12 submarines of the Attack class will enter service with the RAN between 2035 and the mid-2050s. This is if all goes well. But all is very unlikely to go well with an ab initio design of a submarine. The experience of other countries shows they are always delivered late. The Suffren, first of the French nuclear powered Barracuda class, which is the reference design for the Attack class, is already running five years late, although reportedly with reactor problems. Even Naval Group’s program to deliver six Scorpène submarines, a well-established design, to India is running five years late.11 With successive governments delaying taking a decision on the Collins replacement program before deciding to take up the riskiest possible ab initio proposal, the likelihood of a gap in the RAN’s submarine capability has increased over time. Only in the last year has Defence acknowledged that the selection of Naval Group under the CEP means that action needs to be taken to avoid a capability gap. With the Collins class reaching the end of their design life between 2026 and 2034, before the first Attack class boat enters service, all six submarines will need a ten-year LOTE to reduce the extent of the capability gap. This needs to be added to the cost of SEA 1000. As yet the government has not approved the LOTE and no budget for it has been included in the Integrated Investment Plan. Estimates of the cost of the LOTE range between $6 billion and $15 billion, but no announcement has yet been made on how extensive it will be. As noted earlier, the 2020 ANAO report stated that “Defence’s planning for a life-of-type extension for its Collins class submarines, to manage the risk of a capability gap, remains at an early stage”.

Critical importance of the Collins class LOTE Defence’s plans for the Collins LOTE were laid out in limited detail in answers to questions at Senate Estimates. The objective is to undertake the LOTE within the two- year maintenance schedule during the full cycle docking (FCD) process. Replacing diesel generators, the main motor, elements of the power distribution system and periscopes were all included in the LOTE. This is a very major scope of work that will be complex, costly and risky. It will involve substantial design effort and very invasive industrial activity. Industry experts have suggested that building the replacement components into a new after end hull section, constructed before the submarine enters the dockyard to begin the LOTE, would improve the likelihood of completing the work within the FCD window, thereby minimising the impact on submarine availability. Conceptually, such an approach makes sense but it moves the LOTE from being largely a repair and maintenance activity to one with a heavy emphasis on shipbuilding, which would require specialised infrastructure and industry skills. There would be competition for the same resources required for the Attack construction program. The !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! of!the!French!bid.!Perhaps!the!French!submarine,!coming!from!a!blue!water!pedigree!of!nuclear!submarines,! offered!a!substantially!greater!capability!than!the!Type!216.!But!if!so,!we!have!yet!to!understand!in!what! respects!that!may!have!been!the!case.!Defence!has!never!justified!its!selection!in!any!significant!detail!but!has! spruiked!the!Attack’s!substantial!advantages!in!terms!of!capability.! 11!https://www.navyrecognition.com/index.php/news/defenceGnews/2019/june/7192GprojectG75GscorpeneG subsGofGindianGnavyGtoGgetGdelayedGdueGtoGseriousGdefects.html!! 21!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? effects of this would be reduced if the FCDs were being undertaken in Western Australia at that time. Defence has experience of comprehensive LOTE programs, not all of it happy. The FFG Upgrade Program is a worthwhile case study in how challenging it can be, particularly for platforms, like Collins, with systems originally designed way back into the last century. The capability provided by the LOTE did not meet the original requirement, it was way over budget and only four of the six ships received the upgrade. Worst of all was the fact that it ran so far behind schedule that the fleet had no area air defence capability for a long period of time. Regardless of how thorough the preparatory work may be, unseen deterioration of the vessel and unforeseeable problems always emerge once the shipyard starts pulling things apart. Some of these will be major and even if they are not, they can rarely be ignored, especially in a submarine. It will consume time and effort to rectify such things. All sorts of things are discovered too when the shipyard starts putting it all back together and trying to make it work. Equally important, sensors, combat, communication and weapon systems will also need to be upgraded to maintain the submarines’ relative capability in the context of potential threats evolving in the region. Some of this work is scoped, funded and underway or planning is advanced, but some has not yet even been identified. There is a high risk that the totality of this work will result in the submarines’ availability being constrained, which will have a flow on impact on Navy’s ability to generate the increased number of qualified and experienced submariners it needs to provide crews for a much larger Submarine Force. If the Collins LOTE is to be completed during scheduled FCD deep maintenance and docking periods, ASC will need additional resources as well. This is one of the main reasons that the choice of location for future Collins Full Cycle Dockings and the LOTE is so important. Getting this right, so submarine availability is not reduced, will be critical if the risk to the recruitment and retention of more submariners is to be minimised. The design effort required for the major changes to the submarine mentioned above will be significant and getting that work right will be vital. This will be especially important if the LOTE entails major changes that alter the overall weight distribution and therefore balance of the submarine. This is very likely for the scope mentioned above. Yet, while ASC has substantial experience in cutting the hull of the submarines for maintenance purposes, the design of a new after end and integrating the new complex systems with the platform is no simple task. It is a submarine construction and systems integration project rather than maintenance and repair. ASC has not built a submarine for nearly 20 years and does not currently have the recent experience in a significant submarine construction task to undertake such a major activity without support. In this case, and notwithstanding ASC’s very useful design capability, expertise and capacity, as a substantial risk mitigation measure we believe that ASC, as the Design Authority, would be most unwise not to engage the services of an experienced design partner. ASC’s capability partner, US submarine designer General Dynamics Electric Boat would provide valuable but limited assistance. The original designer, now Saab Kockums, is the company best equipped to provide such services, having designed the submarine in the first place. Strict limitations on access to Collins intellectual property also restrict the choice to Saab Kockums.

22!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Even if the ten-year life extension is successful, the last Collins class submarine would be retired around 2045-6. If the Attack class does run late, there may very well be a significant capability gap from the mid-2030s (Exhibit 2.1).

EXHIBIT 2.1: POTENTIAL GAP IN SUBMARINE CAPABILITY

Total submarines required: "Force 2030"

Number of Collins Class submarines in service

12 Number of Attack Class submarines in service

Number of Collins Class submarines in service with LOTE of 10 years

5 year delay in Attack Class submarines

6 Number of submarinesNumber of

0 2020 2025 2030 2035 2040 2045 2050

Source: Insight Economics

If all goes well with the LOTE, life extended Collins submarines would be delivered every two years between 2028 and 2038, with the first extended boat being retired in 2038. With the Attack class intended to be commissioned on a two-year drumbeat from 2036, in principle a six-submarine force would continue until 2046 when it would begin to build up to 12 thereafter. In terms of delivering on the 2009 government decision to double the size of the Submarine Force to 12 boats under the 2009, Force 2030 DWP, this is a clear fail. A full 37 years later, there will still only be six to seven submarines in the Force.12 The recent ANAO report showed that Defence had, for the first time, acknowledged the possibility of a capability gap:

Defence has identified that a delay in the Future Submarine Program of more than three years will create a gap in Navy’s submarine capability. Defence’s planning for a life-of-type extension for its Collins class submarines, to manage the risk of a capability gap, remains at an early stage.13 The worst possible outcome would be if the LOTE were unsuccessful and in addition the Attack class were delayed. The capability gap would then be total and would last for several years. This would have major implications not only for national security but also for the ability of the Submarine Force to sustain its highly trained cadre of people who need extended periods at sea in order to sustain their skills. In a situation where the

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 12!While!Defence!makes!a!convenient!target!for!delays!and!failures!in!military!programs,!Ministers!should!not! be!let!off!the!hook.!Planning!for!a!followGon!submarine!to!Collins!should!have!begun!under!the!Howard! government.!It!did!not.!While!the!Rudd!and!Gillard!governments!proposed!to!double!the!size!of!the!submarine! fleet,!they!dropped!the!ball!in!terms!of!implementation.! 13!ANAO!(2020),!op.&cit,&page!10.& 23!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? population of the Submarine Force will need to more than double in order to provide crews for a fleet of 12 submarines, this is a very serious prospect. The looming problem could well be worse than the situation with the transition from the Oberon to the Collins class, which resulted in problems in recruiting, training and crewing that took the Submarine Force 15 years to unwind. That occurred in a far more benign strategic situation than Australia is likely to face in the 2030s and 2040s. While it is now too late to avoid the need for the Collins LOTE, there is still an imperative to bring new submarines on stream as quickly as possible so as to minimise the capability gap. We note the German contender in the CEP, TKMS, provided not only a fixed price tender for their Type 216 submarine, but also guaranteed delivery of the first submarine in 2028 with the option of a much more rapid drumbeat than that currently planned for the Attack class. But without the ability to apply competitive pressure to Naval Group, there is no incentive for that company to raise performance levels so as to accelerate delivery of the Attack class. Finally, there is also a high risk that the LOTE will not provide sufficient advanced capability to enable the Collins class to be deployed in a high threat environment in the 2030s. It is unlikely, for example, that the LOTE will include equipping the boats with AIP or modern high energy density batteries that could significantly reduce their indiscretion ratio. With some other countries in the region already greatly increasing their capability in anti submarine warfare (ASW), the survivability of the extended Collins boats while operating ‘up threat’ would be called into question. This is discussed further in Chapter 3 below.

2.3 Sovereign capability – local content and access to IP

Australian industry content and the Naval Shipbuilding Plan An excessive amount of debate around the future submarine focuses on the politics of jobs in Adelaide rather than the best outcomes in terms of capability and national security. There seems to be a view that naval shipbuilding can substitute for the car industry in terms of bolstering economic activity and creating jobs. It cannot; the automotive industry, at its peak, gave rise directly and indirectly to perhaps 200,000 jobs.

EXHIBIT 2.2: NAVAL SHIPBUILDING – EMPLOYMENT IN MID TO LATE 2020’S

! Source:!Department of Defence (2017), “Naval Shipbuilding Plan”, May, page 68.!

24!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

The government’s own Naval Shipbuilding Plan did not support the view that the very large naval acquisition program would give rise to a jobs bonanza (Exhibit 2.3). As shown in Exhibit 2.3, Defence’s projections for Australian employment in the industry when naval shipbuilding is in full swing in the late 2020s shows a total of 15,000 jobs, including the acquisition workforce (around 6,000 jobs), the sustainment workforce (similarly about 6,000 jobs) and employment in the supply chain (about 3,000 jobs). An immediate observation is that many of these 15,000 jobs are already there, with ASC still engaged in new construction (carry over from the Hobart class to the OPVs) and large-scale, long-term work in sustaining both the Anzac and Collins class vessels, activities that have a very high level of local content. One implication of the data shown in Exhibit 2.3 is that only around 3,000 jobs will exist in the supply chain, presumably covering both acquisition and sustainment, many of which are already there as noted above. It may be, therefore, that only around 1,500 jobs will be in the acquisition supply chain. The implied job creation seems to represent a very poor return for a projected capital expenditure of over $90 billion on these acquisitions. It is clearly at odds with then Minister Pyne’s aspiration in 2016 to spend 90 per cent of the Defence acquisition budget in Australia.

Australian industry content and sovereign capability As background to the discussion on AIC as well as intellectual property rights (contained in the next section), we need to understand why the highly risky decision was taken in the 1980s to build the new class of submarines in Australia. Experience with the preceding Oberon class submarines, which were acquired off-the-shelf in the UK, showed the disadvantages of being on the end of a very long supply chain for spare and replacement parts. (Supply was cut off completely during the Falklands war.) Also, as proposed by Commodore Paul Greenfield RAN (Retd):

Another major factor was to reduce the in-service support costs. For example, with the Oberon submarines, one data point – OVENS’ 3rd refit - cost more than the asset value itself. And the last two refits at Garden Island were so excessive in costs that the Government cancelled the 3rd refits for both ORION and OTAMA. The Navy believed that by building all the Collins boats in-country, the cost of maintenance would be significantly reduced, through transfer of technology, establishment of infrastructure, and establishment of the supply chain and technical support networks, in-country, from the word “go”. The take-away from all this is that the major driver behind building all the Collins boats in Australia was to reduce the cost of maintenance.14

It was decided in the interests of national security that Australia needed a sovereign submarine capability so as to ensure the efficient and cost-effective sustainment and the required level of availability for the Submarine Force. One criterion underlying SEA 1000, therefore, is the need to build such a sovereign capability. An obvious question is why we need to build sovereign capability all over again with a new entrant with no experience of Australian industry and majority owned by a foreign government not allied to Australia. After the expenditure of great effort and substantial funds over three decades, with many disappointments along the way, we finally have achieved sovereign capability as embodied in the publicly owned ASC and its associated supply chain. ASC built the Collins class submarines to a high standard at a world competitive price – !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 14!Greenfield,!Paul!(2015),!“Engineering!considerations!of!sourcing!Australia’s!Future!Submarine!offGshore”,! Address!to!the!Royal!United!Services!Institute,!Submarine!Summit,!25!March,!page!3.! 25!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? remarkable for an ab initio design in a country that had never built a submarine before – and is now maintaining them with a very high level of local content and achieving world best practice availability. The risks around trying to replace this hard won capability with a new supplier do not seem to be generally understood. There is the definite prospect of experiencing déjà vu all over again, and not in a good way. In any case, in order to achieve these goals and self-reliantly sustain the submarines, a high level of local content is required. Despite the fact that Australian industry had no prior experience of submarine construction, the original Collins class were built with around 70 per cent local content. In sustaining the Collins submarines, over 90 per cent of the budget is spent in Australia. Ministers have stated that there will be a high level of local content in the SEA 1000 program. At the time of the announcement of the result of the CEP process, President Hollande suggested that thousands of jobs would be created in France as a result of the “historic contract”. According to an article in Le Monde, 4,000 jobs would be created in DCNS (now Naval Group) and its 200 sub-contractors, while 2,900 jobs would be created in Australia.15 Approximately $12 billion would be spent in France. This imbalance created some consternation in Australia. Minister Pyne immediately stated that all the submarines would be built in Adelaide. Soon after, Sean Costello, the then CEO of Naval Group Australia, promised ‘at least 90 per cent’ Australian content. He left Naval Group shortly afterwards. That commitment dropped to ‘at least 60 per cent’ from the next Australian CEO, Brent Clark. He left too. In January 2020, the current CEO said he could not guarantee even 50 per cent local content because of the limited capability of Australian industry and the fact that Australia was a demanding customer. Following this downbeat assessment, the Defence Minister has extracted a commitment to 60 per cent from her French counterpart. But what does this really mean? In May 2016, Minister Pyne said he would obtain a commitment from Naval Group for 90 per cent AIC. Now out of office he says the government should “hold Naval Group’s feet to the fire”. Industry remains unimpressed for two reasons. First, Naval Group will not guarantee 60 per cent for at least two years when the construction contract is scheduled to be signed. Secondly, local content is very broadly defined, and seems to include such expenditures as lunches at the Novotel Barossa Resort, just up the road from Osborne, and also revenues to Australian subsidiaries of French companies. Transfer payments within global companies constitute a thriving business for the accounting firms these days. In this context it is interesting that the latest estimate of Australian jobs created in constructing the Attack class is 2,800, slightly less than the 2016 Le Monde estimate based on many more jobs in France. 16 Had the SEA 1000 team been genuinely focussing on the relative attributes of overseas design partners/shipbuilders during the CEP rather than the potential capabilities of pre-concept design platforms, they would surely have come to understand the risks in selecting a partner owned by a foreign government with a record of a highly

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 15!Gallois,!Dominique!et!Taïx!,!Caroline!(2016)“SousGmarins!vendus!par!DCNS!à!l’Australie:!les!coulisses!d’un! contrat!‘historique’,!&Le&Monde,!26!April,!https://www.lemonde.fr/entreprises/article/2016/04/26/leG francaisGdcnsGremporteGunGmegacontratGdeGsousGmarinsGaG34GmilliardsGdGeurosGenG australie_4908510_1656994.html!! 16!Hellyer,!Marcus!(2019),!op.&cit.,&page!79.! 26!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? nationalistic approach to work sharing. On a massive project like this one, supposedly extending over four decades and more, it would seem obvious that Defence would have engaged closely with the proposed design partner and sought commitments from them in a large number of areas such as Australian industry participation. Even if they had undertaken any basic research, and talked to other friendly navies such as Britain’s Royal Navy, they would have understood that French Primes have substantial form in giving a clear preference to their own national defence industry. For example, this extract from a House of Commons report on the reasons why the UK abandoned a joint endeavour (Project Horizon) to design a frigate for the UK, France and Italy:

“The enforcement of a best value for money UK-style equipment selection decision was always going to be difficult in such a situation ... In the Horizon procurement, the position of the DCN—part of the French Ministry of Defence—meant that it was the customer (working with the DGA [French procurement agency]), a shareholder in the prime contractor organisation (the IJVC) and a subcontractor bidding ... its HEPICS combat management system. This, combined with the French Government's support for its national champion, had a significant impact on the prospects for a level playing-field competition. Differences in procurement procedures between the UK and the other partners were clearly at the very heart of the difficulties encountered with Horizon, and the other partner countries highlighted for us the difficulty of reconciling the UK's market approach with their own focus on work-shares. Such differences in approach were ultimately dictated by whether the relevant defence industries in each country were privately or publicly owned. With hindsight, it seems clear that the Italian and (in particular) the French governments were never going to be prepared to embrace the MoD's focus on competition, value for money and contracting for system performance. Indeed, France's DCN appears to have had every intention of buying certain equipment from sources close to home, whatever the requirement or the equipment competitions might have otherwise suggested.” 17

This is a highly critical assessment. The French government, represented by the State- owned company that is now Naval Group, engaged in a joint project with the UK and Italy but was apparently determined to prioritise the employment of French suppliers above the objectives of competition, value for money and technical performance. Currently, every sign is that the same approach is being followed here. Without Naval Group being subject to competitive pressure, they hold all the cards in any negotiation about Australian industry participation. Unless the Australian government intervenes decisively, this will not have a happy ending. The problem is that to be successful Australian industry needs to be involved throughout the detailed design stage. As the SEA 1000 programme proceeds and it becomes more and more difficult for the government to change course, Australian industry’s unease about the process continues to grow. The two parties involved in the negotiations clearly have divergent interests. Naturally enough, the priority of the French government, majority owner of Naval Group, is to maximise economic activity and employment in France as a result of the project. They appear to be playing hardball in pursuit of this objective and are now dismissing Australia’s industrial capability and looking to secure local content by means, for example, of frequenting French-owned resorts in the Barossa Valley. Yet if they succeed, that would contravene Australia’s requirement for a sovereign submarine capability. While companies are understandably reluctant to go on the record, some highly capable, innovative Australian firms report an ominous lack of interest in their capabilities on the part of Naval Group. This appears not to be so much about

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 17!House!of!Commons,!Select!Committee!on!Defence,!Eighth!Report!(1999),!paras!17!and!20,! https://publications.parliament.uk/pa/cm199899/cmselect/cmdfence/544/54402.htm!! 27!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Australian industry seeking protection but rather to be allowed to compete on a level playing field. A group of Australian companies has come together to put their case to government (Exhibit 2.3).

EXHIBIT 2.3: WHAT DO AUSTRALIAN COMPANIES IN THE SUPPLY CHAIN WANT? The Australian Industry Capability (AIC) policy framework established by the Coalition Government is good and has been warmly welcomed by Australian industry but lacks teeth and rigour compared to more developed and mature policies such as the Canadian model which has proven to be best in class. The reality on-the-ground does not reflect the ambition of the AIC policy setting. Foreign original equipment manufacturers (OEMs) will always default to their own indigenous supply chains and state that time and investment to qualify for alternative options are a major risk. Early engagement with Australian industry is key to negate this position. A pre-qualified, credible Australian supply chain with merit is key to this initiative. Non-resident majority owned foreign companies must be contractually obliged to engage Australian Industry to engineer the platforms to enable integration of Australian industry. Australian companies must be included in early engagement in the design and supply chains for the development program, to ensure they will inevitably be engaged on full rate production, through life support and sovereign capability. Australian industry is not after a hand-out, we are after the opportunity to quote for work on a level playing field, from the outset. Australian industry is creative, innovative and cost-competitive - but it needs strong Government support to access the Defence procurement programs on a fair and equal basis.

Source: Dr John White, communication to Insight Economics.

These local firms operate in a global industry corrupted in terms of conventional industry economics by governments and Eisenhower’s ‘military industrial complex’, providing highly favourable trading conditions to the domestic players in their defence industries. The French play this lucrative game particularly well, but are by no means alone in their endeavours. Australia already has built a sovereign capability, at great pain and cost, which is performing very well in the successful Collins class sustainment program. It now provides world-class availability of submarines, as illustrated in the photo on the front cover of this report (four out of six at sea). The Naval Group proposition is by no means sufficiently compelling to abandon this essential capability. Finally, the government needs to develop a policy for enforcing a level playing field approach on foreign Primes in respect of AIC so that competitive Australian firms are provided with appropriate opportunities. One model is the one used by the Canadian government in respect of defence procurement.18 Managed through an independent agency, the Canadian government imposes contractual obligations for local content and engagement with a points system tied and weighted to sovereign capability. If Australia followed that model, it could, for example, in each major naval acquisition project, mandate an expenditure level in Australia of say 70 per cent of the value of the contract. If local firms were unable to offer the full 70 per cent at a competitive cost, the Prime would be free to meet the target in another way. For example, the Prime could bring the most efficient local firms into their global supply chains or purchase Australian defence products unrelated to the particular acquisition, such as CEA radars or PMB batteries.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 18!See!http://www.ic.gc.ca/eic/site/086.nsf/eng/00006.html#itbpolicy$! 28!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Access to intellectual property The inadequacy of Australian rights to relevant IP was a thorn in the side of Collins class sustainment for a very long time. The failure to negotiate appropriate IP protocols in the first place was a result of Australia’s lack of experience in acquiring major complex platforms for which it had parent nation responsibility. It has now learned that lesson. Appropriate rights to IP are vital in terms of both sustaining the submarines and undertaking upgrades. As with local content, IP protocols would generally have been included in the SPA. But if so, we do not know whether an outcome satisfactory to Australia was achieved. It is clearly in the interests of the French to maximise the value of their IP. If Naval Group play hardball, in the absence of competitive pressure it is difficult to see how Defence will be able to negotiate an acceptable outcome that also protects Australian proprietary IP. 2.4 Technical risks

Although flagged as a ‘regionally superior’ submarine, even compared to other conventional submarines the Attack class may not embody advanced technologies. As ASPI’s Marcus Hellyer has pointed out: “even though the schedule seems to be extremely long compared to other conventional submarine projects, Defence has said that the project is adopting a low-risk approach to key technology, so only currently existing technologies will be used in the first batch. Therefore, Defence has said it will still use traditional lead-acid rather than higher density lithium-ion batteries, despite the latter being used on a recently launched Japanese submarine.”19 With the German navy also now trialling Li-Ion technology in a Type 212 submarine, there is a clear tension here between reducing risk in the design and construction of the submarine and the need to minimise operational risks and increase survivability. In many ways, the adoption of a conservative approach to new technology carries a significant risk of its own in that when the early Attack class submarines first go to sea in 15 years time, they may in some respects be obsolescent. In addition, although Defence has suggested that the Attack class may deploy AIP, the current presumption is that, because of the weight penalty, it will not. This is a controversial decision. While incorporating AIP may lead to penalties in terms of the speed of advance (SOA) while in transit, it also provides substantial advantages in terms of both effectiveness and survivability, particularly in a highly contested operating environment. Although AIP is valuable mainly while on patrol, it can also be of value in transits that are contested or while passing through areas where counterforce interdiction is more likely, such as the choke points in the Indonesian archipelago. Defence also seems determined to fit the submarines with pump-jet propulsion, which can provide some signature advantages and is increasingly used on nuclear submarines. If a fundamental reason Defence selected the French offering was the acoustically superior performance (under certain scenarios) of its pump-jet system, they may not have understood the possible downside of their choice. An obviously well informed article by Cameron Stewart suggests that superior stealth in a particular tactical scenario (possibly the need to break contact at high speed when detected) favoured the !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 19!Hellyer,!Marcus!(2019),!op.&cit.,!page!77.! 29!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

DCNS pre-concept design. According to Stewart, the Germans were told their design would have an unacceptable level of radiated noise on a particular frequency. The use of pump-jets combined with an electric drive system rather than turbines arguably give French nuclear submarines an acoustic advantage over those from other countries. Again according to Stewart, DCNS presented to the Defence team an impressive demonstration of the superior acoustic performance of a French nuclear submarine with propulsors compared to a submarine with propellers.20 When the decision in favour of the French submarine was announced, DCNS declared propellers on submarines to be “obsolete”. This was presented as a breakthrough technology on the Shortfin Barracuda. On the DCNS website at the time, it was stated that:

Pump jet propulsion means the Shortfin Barracuda can move more quietly than submarines with obsolete propeller technology. In a confrontation between two otherwise identical submarines, the one with pump jet propulsion always has the tactical advantage.”

The problem with this is that despite some experiments in the past using pump-jets on Russian and French conventional submarines, the downside is that a pump-jet propulsor is more power-hungry than a propeller, at low speeds significantly so. While this has no downside for a power-rich nuclear submarine and can even be a benefit, it is a critical issue on a conventional submarine with restricted power supplies. Other things being equal, pump-jets will increase the indiscretion ratio, perhaps considerably, as well as reducing range. Submarines for Australia has supported research in this area by Aidan Morrison, a physicist with expertise in marine propulsion systems. Morrison stated that:

Unless the French have achieved a quantum leap in pump-jet technologies in the past few years and none of the previous physics or experimental results apply, it seems that the choice of a pump jet is curious indeed. Exactly what kind of advantage would justify accepting such a penalty in terms of dived range, dived endurance, indiscretion ratio and overall range is quite hard to imagine when building a ‘regionally superior’ submarine.21

Despite numerous examples in Senate committee hearings of ongoing support by Defence for pump-jets, we believe there are substantial concerns about using them on a conventional submarine with limited power resources. In October 2017, a very senior executive in Naval Group said that the Attack class might use a propeller, only to be instantly refuted by Defence where RADM Sammut stated that they wanted the pump- jet system demonstrated to them by the French during the CEP process. He also claimed that pump-jets could be more efficient through the whole speed range, which is quite contrary to our sponsored research. But as Hellyer recently pointed out, “Defence also has not been able to provide a compelling plain English explanation to counter the arguments of some analysts who have suggested that expecting a conventionally

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 20!Stewart,!Cameron!(2016),!“The!sound!of!silence!–!why!Germany!lost!its!subs!bid”,!The&Australian,!30!May,! https://www.theaustralian.com.au/nation/defence/theGsoundGofGsilenceGGwhyGgermanyGlostGitsGsubsG bid/newsGstory/4b3d69b49a8371e9837ed59e4f0faac2!! 21!Morrison,!Aidan!(2018G2),!“A!pump!jet!for!the!future!submarine?!Not!so!fast!(or!slow)”,!The&Strategist,!14!February,! https://www.aspistrategist.org.au/pumpjetGfutureGsubmarineGnotGfastGslow/! 30!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? powered submarine to perform better with a pump-jet than with a traditional propeller defies the laws of physics”.22 With the Attack class presently being designed for lead acid batteries and no AIP, the use of pump-jets could convert an already uncompetitive indiscretion ratio to one that constitutes a threat to both operational effectiveness and survivability. Ultimately, probably following land-based testing, this may well be understood and the design will need to be changed so as to accommodate a propeller. But because of weight distribution issues, this will be no easy task, possibly requiring changes amounting to one third of the submarine’s design. The risk here is that the already lengthy delivery schedule for the Attack class will be further delayed, possibly significantly, and depending on contractual arrangements, the cost may well increase. If all three of these issues – lead acid batteries, no AIP and pump-jet propulsors – play out as presently appears likely, the Attack class will have an indiscretion ratio that is likely to be no better than that of Collins and may be worse. It will also be higher than most submarines operating in the Asia-Pacific in the 2030s, which will either be nuclear powered or, if not, will incorporate AIP and/or modern batteries. ASPI’s Marcus Hellyer also raises the question of “how the design is being ‘future- proofed in order to adopt emergent technologies and capabilities, such as unmanned and autonomous systems, that are likely to mature before 2035 … does the future submarine’s low-risk approach to technology, while good for project management, push risk into the resulting capability delivered by the submarines?”23 We understand, for example, that the Attack class design will only be able to launch small autonomous underwater vehicles (AUV), limited to the size of its torpedo tubes and the cavity between the pressure hull and the outer skin. Indeed, one former Collins class commanding officer told Insight Economics that if the Attack class does not at least embody modern batteries, the first submarine “will be obsolete before it hits the water”. Incorporating major new technology into the Attack class, such as new battery technology that would probably significantly alter the submarine’s balance, will take a significant amount of time. How much time will depend on many factors, such as the maturity level of the new technology to be incorporated and therefore how much research and development work is needed to bring it to readiness for service. Also, the more the new technology impacts the original design, especially the overall weight balance and margins (space, weight, reserve buoyancy, power, and cooling) the longer it will take to do the necessary design work and the greater the cost and schedule implications will be. In practical terms, even if we started preparing significant changes to the Attack design immediately, those changes would probably not be delivered in a new submarine until well into the 2030s – Attack class submarine number three, four or possibly even five. Defence has given no indication that any such capability and design development process is underway or even yet contemplated. It seems likely, therefore, that obsolete battery technology is already locked into the first flight of the Attack class.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 22!Hellyer,!Marcus!(2019),!op.&cit.,!Footnote!99.! 23!Hellyer,!Marcus!(2019),!op.&cit.,!page!77.! 31!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

In conclusion, these risks are significant and generally reflect choices made by Defence. The introduction of contestability could have a beneficial effect on the technological choices made in the course of the acquisition program. As a Senate committee noted:

This reluctance of overseas countries to relinquish their most advanced technology means that Australia needs to apply competitive pressure to ensure that it is best placed to obtain the cutting edge technology it is seeking to acquire. Again this need to test and encourage designers and builders to devise the best option for the future submarine is another compelling reason for having a competitive tender.24

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 24!The!Senate!Economic!References!Committee!(2014),!Future&of&Australia’s&naval&shipbuilding&industry:&Future& submarines,&November,!page!42.! 32!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Chapter 3

Will the Attack class be fit for purpose?

In Chapter 1 we suggested that the main criterion for success under the CEP was the ability to propose an advanced capability – namely a ‘regionally superior’ submarine. We don’t know what that means. How can a SSK of any type claim a superior capability ‘up threat’ in the Indo Pacific when it will be operating in the same battlespace as nuclear-powered submarines from at least four other countries? How can the Attack class be superior to other SSKs when it is unlikely to deploy either AIP or, at this stage, modern, energy dense batteries? Perhaps what Defence means by ‘regionally superior’ is the ability to do what other conventional submarines generally do not attempt to do. Unlike other SSKs, Australia’s submarines undertake 70-day operations with very long transits conducted at a challengingly high SOA. They then undertake patrols in AOs where the operational and tactical intensity is of the very highest order and where the opposition is growing, both absolutely and relatively, in power and capability by the month. But the important question is not whether the Attack class will be regionally superior. It is whether, in terms both of effectiveness and survivability, it will be fit for the purpose for which it is being designed, not now but in the 2030s and beyond.

3.1 What do Australian submarines do?

The primary operational requirement for the RAN Submarine Force is to undertake anti-submarine warfare (ASW) in wartime and intelligence gathering, surveillance and reconnaissance (ISR) more generally. As discussed below, this involves them operating in high intensity areas ‘up threat’. These are inherently risky operations even in peacetime. The key question addressed in this Chapter is whether the Attack class is likely to be fit for this purpose when it begins to come on stream in the mid-2030s. In order to evaluate this issue, we need to understand, in some detail, the nature of the operations our submarines undertake and how those operations may need to change over time as the strategic circumstances change and new technologies are developed that disrupt existing operational and tactical doctrine. This requires considerable research. Both the Australian and the British defence departments have a longstanding policy of not commenting on submarine operations. ADF officers are also commendably scrupulous about not revealing classified information. The Americans, however, are generally more open about military operations. As Chris Uhlmann wrote recently, “ask any journalist working in a theatre where the US and Australian forces are working together and they will attest to the fact that the fastest way to find out what the Australians are up to is to ask the Americans.”25 We didn’t ask any Americans but we did look at the Internet, which provides a rich source of authoritative material on US military operations. The Commander of the US Submarine Force, for example, periodically publishes a statement of his intent, including how the Force works with allied submarines in their main areas of operations. In addition, since the break-up of the Soviet Union, a great deal of material has been !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 25!Uhlmann,!Chris!(2019),!“We!Australians!are!wowsers”,!The&Age,!14!August,!page!19.! 33!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? published about NATO submarine operations in the north Atlantic during the Cold War. Much of this doctrine remains the same today in other theatres, although advances in technology have had a considerable impact on its likely effectiveness These insights provide valuable pointers as to how submarine operations are currently developing in the South China Sea. Indeed, as the 2016 Defence White Paper made clear the ADF’s force structure is increasingly designed to undertake operations in close liaison with US forces or, as the Minister put it in a recent speech, “operating in high threat environments with coalition partners”.26 Our research suggests that this is a good summation of the role of the RAN’s Submarine Force, both currently with the Collins class and what is planned for the future with the Attack class submarines.

3.2 Submarine operations ‘up threat’ in the Indo Pacific

Under RAN doctrine, submarines should be employed in an offensive role where their covert qualities and, particularly for a middle power, their value as a force multiplier can be used to greatest effect in prosecuting asymmetric warfare. Since submarines are difficult to detect both in the commons of the open ocean and around the multitude of target zones where they may loiter, ASW (and, in practice, ISR) operations are most effective when undertaken close to an adversary’s base. These doctrinal points suggest that Australian submarines operate ‘up threat’ in the waters far to our north. Indeed, various Parliamentary inquiries have been given evidence that the main area of operations (AO) for our submarines is up to 3,500 nautical miles from Fleet Base West at Fremantle. Although Defence is not specific, it is clear that the main AO for RAN submarines is in the South China Sea. The PLA Navy (PLAN) has a major base at Hainan Island, where its submarines are able to emerge and return to base covertly via a tunnel. As a former commander of the Submarine Force, Rear Admiral RAN (retired) Peter Briggs, stated in 2014:

Our submarine force must be capable of operating and surviving north of the archipelago, throughout the South China Sea, able to observe, report and if necessary strike. … This is the high payoff area, where their impact is greatest and unique amongst ADF assets. To be able to exploit the initiative gained from their stealth, Australia’s submarines must be able to covertly reach sensitive areas throughout our region with sufficient mobility, endurance and payload for the long duration missions involved, frequently in or through demanding tropical waters. … Without being specific about the scenario, it’s therefore likely that Australia will wish to be capable of maintaining a deterrent submarine presence at very long ranges, say 3,500 nm.27

In peacetime, the principal AO for the submarine force remains the same, up threat in the South China Sea, with the emphasis being on ISR. This has been an important role for the submarine force and one that, as well as producing intelligence for the ADF, also provides Australia with valuable ‘coin’ in the Five Eyes intelligence network.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 26!Reynolds,!Senator!the!Hon!Linda!(2019),!address!to!the!RAN!Sea!Power!Conference,!Sydney,!8!October,! https://www.minister.defence.gov.au/minister/lreynolds/speeches/royalGaustralianGnavyGseaGpowerG conferenceGinternationalGconvention! 27!Briggs,!Peter!(2014),!“How!many!submarines?!(Part!1)”,!The&Strategist,!8!January,! https://www.aspistrategist.org.au/howGmanyGsubmarinesGpartG1/! ! 34!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

It would be idle to pretend that RAN submarines will operate alone in this theatre. As the Defence Minister said in October 2019, “by 2035 we expect around 300 submarines to be active in our region.”28 Indeed, the South China Sea has now become centre stage for the strategic contest between the military forces of the United States and China. The media reports, however, focus almost exclusively on the deployments on and above the surface of the South China Sea. Here the contest is highly visible. China’s strategy appears clear: it regards the South China Sea as its Mare Nostrum, to be militarised so as to be able to enforce a strategy of anti-access and area denial (A2/AD) directed particularly towards the US Navy. The militarisation of the disputed reefs and islands with air bases and batteries of advanced anti-ship missiles, including DF-21Ds and DF- 26Bs, the so-called ‘carrier killer’ ballistic missiles, plays an important role in this A2/AD strategy. The US strategy, on the other hand, is to contest these initiatives, some of which have no basis in international law. As well as conducting freedom of navigation operations (FONOPS) within China’s newly claimed territorial waters in the disputed island chain, the US Navy is unlikely to be fundamentally deterred by vulnerable PLA military installations located a very long way from major bases on the mainland. The USN maintains its powerful and highly visible Seventh Fleet in this theatre, operating out of Guam. Indeed, it was the deployment of two American carrier battle groups in 1996 that led to China backing down from its moves against Taiwan at that time. It is worthy of note, however, that it is over a decade since a US aircraft carrier was deployed to the Taiwan Strait.

Underwater contest in the South China Sea But it is likely that the main arena where the contest in the South China Sea plays out is not in the surface or above surface domains. To be sure, tension will be heightened there from time to time by FONOPS or by some other interaction between opposing forces, perhaps over competing territorial claims. But the history of the Cold War in the northern hemisphere suggests that the main contest will be played out in the undersea domain. Submarines, particularly nuclear-powered submarines, are now the capital ships of great power navies and this is the theatre where they can be deployed to greatest strategic effect. The US Navy has developed its ASW doctrine over a long period during the Cold War as a consequence of high intensity operations to counter the threat from the then Soviet Union’s powerful submarine force in the north Atlantic. The details of the strategic contest in the north Atlantic during the Cold War were largely declassified in the 1990s and extensive material is now available on how NATO submarines (principally American and British SSNs) operated in what was a grey zone positioned somewhere between a cold and a hot war.29 During the Cold War, a major strategic objective of US forces was to locate the position of every Soviet submarine in real time. As the Soviets developed a credible second-strike capability, based on submarine launched ballistic missiles (SLBM), the particular focus was to detect and track every Soviet ballistic missile submarine (SSBN). One objective early on was to confine the SSBNs to defensive ‘coastal bastions’, where their ballistic !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 28!Reynolds,!Senator!the!Hon!Linda!(2019),!op.!cit.! 29!For!a!comprehensive!description!of!submarine!operations!in!the!north!Atlantic!during!the!Cold!War!see! Hennessy,!Peter!and!James!Links!(2015),!The&Silent&Deep,!Penguin!Books,!London,!2016.! 35!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? missiles would be out of range of the continental United States. Later, as the range of Soviet SLBMs increased, the objective was to detect the Soviet SSBNs when they set out on patrol and then track them closely so as to be able to eliminate them immediately at the outset of any war. So successful were the NATO forces that increasingly desperate Soviet countermeasures were said to reflect the Soviet High Command’s realisation at one point that American and British submarines “could sink Soviet ballistic missile submarines at will in the northern Atlantic and Pacific seas”.30 This objective was pursued assiduously by NATO using a multi-layered approach to detection, including human intelligence (HUMINT), signals intelligence (SIGINT), satellites (to detect sub-surface anomalies), maritime patrol aircraft (MPA), helicopters with advanced radar and sonobuoys, surface ships with towed arrays and onboard sonar and, not least, submarines, also with towed arrays and other sensors. As a vital backdrop to all these assets, the Americans installed extensive seabed arrays of passive sensors (SOSUS) in parts of both the Atlantic and Pacific oceans. Dating from 1954, SOSUS is now part of the Integrated Undersea Surveillance System (IUSS). The IUSS was constantly upgraded over time and incorporated data generated not only by submarines but by many other layers in the network, including satellites; MPA, helicopters; towed array and shipboard sonars on surface ships; and sonobuoys. This complex system all required the development of a highly advanced and very costly Command, Control, Communications and Intelligence (C3I) network to share information in real time and to ‘cue’ allied assets onto tracking the adversary’s submarines and high value surface ships. “At its cold war peak, SOSUS employed around 4,000 personnel working at 20 shore stations.”31 The Americans also needed to know the position of allied submarines and so a sophisticated system of water space management was created within the C3I network. Each NATO submarine was given a course to its AO and was required to keep within a virtual ‘box’ so that other assets knew where it was at any time. If the submarine could not keep within its box, perhaps because of mechanical problems or interdiction, it needed to notify the controller and the progress of the box was changed. When a Soviet submarine was detected, its coordinates were reported to the controller who cued other assets, or the asset that detected it, onto tracking the submarine and maintaining contact. With the rise of China, the Americans have focussed much more on submarine detection in the Indo Pacific and have transferred their ASW priorities from the north Atlantic to the Pacific (while still maintaining a close weather eye on Russian submarine activity). In 2005, the Pentagon announced that 60 per cent of USN submarines would be deployed to Asia (a very substantial shift).32 In terms of doctrine, it seems clear that successful American ASW tactics, developed and proven during the Cold War, have been transferred to the Indo Pacific. The very sophisticated and expensive infrastructure that underpinned these operations in the Atlantic now operates in that theatre. Technology has moved on substantially in the last 30 years, however, with the enormous increase in computing power allowing a mountain of complex data generated by the IUSS to be processed in real time. Existing systems like the Surveillance Towed

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 30!Hennessy,!Peter!and!James!Links!(2015),!The&Silent&Deep,!op.&cit.,!page!552.! 31!Save&the&Royal&Navy,!“Listening!to!the!oceans!–!the!secretive!enablers!in!the!underwater!battle”,!page!2,! https://www.savetheroyalnavy.org/listeningGtoGtheGoceanGtheGsecretiveGenablersGinGtheGunderwaterGbattle/! 32!Ross,!Robert!(2012).!"The!Problem!with!the!Pivot:!Obama's!New!Asia!Policy!Is!Unnecessary!and! Counterproductive".!Foreign&Affairs.!91(6):!70–82.! 36!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Array Sensor System (SURTASS) have also been complemented by new technologies such as the Deep Reliable Acoustic Path Exploitation System (DRAPES):

As China and Russia have asserted themselves anew as “pacing competitors,” as described by Undersecretary of Defense Robert Work, the U.S. Navy has taken a renewed interest in its traditional Cold War antisubmarine warfare mission. Together, DRAPES and SURTASS promise to provide a persistent, long-range ability to detect adversary submarines around the globe. Using cueing data from those platforms, improved local anti-submarine assets like the P-8 Poseidon sub hunter aircraft (which replaces the 50 year-old P-3 Orion) and surface combatants with new, improved towed sonar arrays of their own, like the Multi- Function Towed Array, can then close on a target, and track or engage it as needed.33

SURTASS ships are generally crewed by civilian mariners. “Following the ‘Asia Pivot’, SURTASS vessels now operate mainly around the Chinese coast and the western Pacific, but the USN is also in the process of fitting all its destroyers and cruisers with a new Multi-Function Towed Array sonar system.”34 In a pivot away from a primary focus on terrorism and the littoral back to blue water great power conflicts, it has also revived the production of the Arleigh Bourke destroyers (all being fitted with towed array sonar) and is in the process of procuring a new class of large frigates with a focus on ASW. The task of American submarines in their various theatres was set out clearly by the force commander in March 2018: The main role of our Submarine Force is to hold the adversary’s strategic assets at risk from the undersea, the same list we want to protect: their homeland, SSBNs on patrol, carriers, critical undersea infrastructure, cyberspace, strategic ports and chokepoints, and submarines.35

In pursuing these objectives, the Submarine Force will:

• “Conduct undetected operations such as strategic deterrent patrols, intelligence collection, Special Operations Forces support, non-provocative transits, and repositioning • Penetrate adversary defensive perimeters to deny safe haven, reduce defenses, and exploit opportunities created by being inside their fence line.”36 We can reasonably assume that this is the current strategy being pursued by the US Submarine Force in the South China Sea. The emphasis on ‘SSBNs on patrol’ is significant. Until recently, the PLA has not possessed a credible second strike nuclear capability, leaving open the theoretical possibility that the US could apply nuclear blackmail in any crisis between the two nations. Now it appears China is deploying operational SSBNs, each armed with 12 submarine-launched ballistic missiles (SLBMs) with multiple independent re-entry vehicles. “In its most explicit assessment so far of this Chinese capability, the Pentagon in its latest annual report on China’s military,

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 33 Stashwick,!Steven!(2016),!“US!Navy!Upgrading!Undersea!SubGDetecting!Sensor!Network”,!The&Diplomat,! November,!https://thediplomat.com/2016/11/usGnavyGupgradingGunderseaGsubGdetectingGsensorGnetwork/! 34!Save&the&Royal&Navy,&op.&cit.,!page!3.! 35!Commander’s&Intent&for&the&United&States&Submarine&Force&and&Supporting&Organisations,&March!2018,!,! https://www.public.navy.mil/subfor/hq/Documents/Commanders%20Intent%20March%202018.pdf.!pages! 1G2! 36!Ibid,!page!4.! 37!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? published in August [2018], said that Beijing now has a “credible” and “viable” sea- based nuclear deterrent.”37 This is a game changer in the Asia Pacific. When the PLA Navy’s bomber submarines are fully operational, which may occur in the next couple of years, the doctrine of mutual assured destruction will apply and America will lose its theoretical ability to blackmail China with the threat of a first strike nuclear attack. If the American strategy is the same as that pursued against the Soviet SSBNs, the US will aim to locate and track every PLA Navy bomber submarine with the objective of eliminating the adversary’s SSBNs immediately if war were to break out.38 While the US may have a well-developed strategy for prosecuting the contest with the PLA in the subsea environment in the South China Sea, its operational capacity to do so is less clear. The ratio of US military resources, particularly submarines, relative to those ranged against it has become less advantageous:

The demands on the United States Submarine Force and supporting organizations continue to grow at an unprecedented rate. The threat continues to evolve and expand rapidly in at least three world regions, each posing substantially different operational demands for both high-end warfighting and peacetime operations. As a result of decisions made many years ago, the number of [USN] submarines has begun a long sustained reduction that will increase these demands on each ship and crew.39

EXHIBIT 3.1: PROJECTED FLEET OF US NAVY ATTACK SUBMARINES

Source: Colley, Christopher K. (2019), “Is America Now Directly Arming Against China?”, The Diplomat, 2 August, https://thediplomat.com/2019/08/is-america-now-directly-arming-against-china/

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 37!Torode,!Greg!and!David!Lague!(2019),!“China’s!furtive!fleet!of!missileGladen!submarines!tests!the!Pentagon”,! Japan&Times,&May,!https://www.japantimes.co.jp/news/2019/05/07/asiaGpacific/chinasGfurtiveGfleetGnuclearG missileGladenGsubmarinesGtestsGpentagon/#.XS9H2K2B2Rs!! 38!CarnegieGTsinghua!(2018),!“The!survivability!of!China’s!SSBNs!and!strategic!stability”,!24!October,! https://carnegietsinghua.org/2018/10/24/survivabilityGofGchinaGsGssbnsGandGstrategicGstabilityGpubG77494!! 39!Commander’s&Intent&for&the&United&States&Submarine&Force,!op.&cit.,!pages!1–2.! 38!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

The scheduled decline in US submarine numbers is significant. From a high point of 98 SSNs in 1987, the fleet declined to 51 at end-2018 and, with the retirement of the Los Angeles class will fall to a low point of 42 in 2028. From there, new construction of Virginia class boats, which is currently under way at a higher rate of two per year, will see an increase to 56 submarines in 2036 (Figure 3.1). In order to maximise their deployed submarine capability in a period where the number of submarine platforms in the US fleet is declining, it makes very good sense for the Americans to engage to the fullest extent possible the submarine forces of trusted allies such as Britain and, in our region, Australia and Japan. In the words of the US Submarine Force Commander: ‘We also integrate with our U.S., partner, and allied Navy undersea forces – including ASW aircraft and ships – that have vital undersea warfare roles.’40 In fact, one of his objectives as stated in March 2018 was to “formalize US/Japan/Australia TASW [theatre anti-submarine warfare] initiatives”.41 It seems clear from this US statement, therefore, that when on patrol in their principal AO, Australia’s Collins class submarines operate in liaison with American submarines and other assets. Indeed, in practical terms it would be impossible for Australian submarines to operate effectively in that theatre without being part of a much larger, multi-layered force or without access to the highly sophisticated American IUSS and C3I networks. In order for the controller to know their position, they would also need to operate within the American water space management system, including when in transit. It is no coincidence that the AN/BYG-1 combat management system and associated communications suite installed in the Collins class (and specified for the Attack class) are nearly identical to those employed on American nuclear submarines. The jointly developed Mk48 CBASS torpedo is the primary ASW weapon in both nations’ submarine armoury. These operations are occurring in a situation where the level of intensity has increased significantly since the 2016 DWP. As a senior ADF officer is reported to have said recently, the waters where the ADF operates are becoming “more congested and more contested”.42 With 300 submarines, many more of them nuclear-powered, projected to be operating in these waters when the Attack class starts to come on stream in the mid 2030s, and with the PLA Navy increasing its capacity to enforce an A2/AD strategy, we can expect the intensity of these operations to continue to increase. In examining whether the Attack class will be fit for purpose in these highly challenging operations up threat, we need to analyse the issue in terms of the capability it will offer in strategic, operational and tactical terms. This is addressed below.

3.3 Strategic capability requirement for the Attack class

As the only significant power projection capability in the ADF’s order of battle, the RAN Submarine Force plays an important strategic role in Australia’s defence. Because of the covert nature of their operations and the power embodied principally in their Mk 48 CBASS torpedoes, Australian submarines have the potential to conduct asymmetric warfare successfully against a much stronger adversary. Ideally, this could offer a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 40!Ibid,!pages!1G2.! 41!Ibid,!page!9.! 42!Quoted!in!Tillett,!Andrew!(2019),!“China!puts!a!rocket!up!our!defence!dollar!challenge”,!Australian&Financial& Review,!5!October,!page!47.! 39!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? considerable deterrent effect and can provide Australia with the military capability to punch above our weight. But in analysing the strategic value of this capability, we need to be cognisant of the current and future relative weight between the RAN and the PLA Navy. In strategic terms, while the Collins class submarines are effective SSKs, Australia does not possess a sufficiently powerful submarine force to deter the armed forces of a great power, particularly one whose naval capability is developing very quickly. At least in terms of the number of naval platforms, the PLA Navy is now larger than the US Navy and it is building warships and submarines at a much more rapid rate. Despite patronising remarks from some observers in the US and British navies, it would also be foolish to underestimate the level of technology embodied in the PLA Navy’s latest assets or the professionalism of its crews. In the words of the commander of the US Submarine Force, “China is aggressively pursuing the most capable and well-funded military in the world, after our own”.43 As one RAN Admiral told us: “Make no mistake. They are mariners.” Australia uses its submarines differently from other nations that operate conventionally powered boats. No country other than Australia requires diesel-electric submarines to undertake high intensity missions 3,500 nautical miles from base. While 70-day operations are the established standard for RAN submarines, European submarines, generally with air-independent propulsion, undertake much shorter missions, dived throughout and without needing to snort. Japan’s Soryu submarines are understood mainly to undertake two-week missions also dived throughout. The major strategic problem for Australian conventional submarines is the long transits, undertaken dived throughout, where the submarines are expected to achieve a high average SOA of around ten knots. This requires frequent snorting, which together with the high SOA can take its toll on both machinery and personnel. While this SOA may have been readily achievable for the Oberons, which undertook transits on the surface, it is far more difficult for a modern SSK operating dived throughout. With adverse currents in some choke points and high temperatures in tropical waters, as well as the possibility of being interdicted en route, overall transit times can blow out to nearly 40 days, equivalent to half of the operation. The relatively short time available for patrolling on station, a clear result of the tyranny of distance, is a major contributor to the main strategic problem confronted by Australia’s current six-boat Submarine Force.44 Submarine availability is now at least consistent with world best practice, with four submarines available and three operational at any time. Yet both because of the very long transits and the RAN’s workplace agreements, in peacetime only one submarine can be on station in its AO for around half the time in any year. By doubling the size of the Force to 12 submarines with the new Attack class, this will allow one submarine to be operating on station at any time. But this will not occur until the 2050s, over thirty years from now, when both the strategic situation may be very different from what it is today and there will inevitably have been profound and unforseen developments in technology. In addition,

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 43!Commander’s&Intent&for&the&United&States&Submarine&Force,!op.&cit.,!pages!1–2.! 44!There!are,!of!course,!offsetting!benefits!of!the!tyranny!of!distance!in!that!it!makes!an!invasion!of!Australia!a! very!difficult!proposition!for!any!aggressor.! 40!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? in 30 years time the power of the PLA Navy relative to the RAN will likely be much greater than it is now. Of course, the Navy’s workplace policies would change in an emergency so as to allow a ‘surge’ in submarine numbers to the AO. But it would still be several weeks, perhaps months, before at least two submarines could be guaranteed to be on station at any time. This also takes no account of attrition. The length of RAN submarines’ transits means that the deployed capability, in terms of time spent on patrol in the AO, is lacking and its deterrent effect must consequently be relatively low. Even with 12 submarines, the value of maintaining one submarine on patrol in its AO at any time must be assessed in terms of the opposing force that confronts it. One conventional submarine is very unlikely to deter a PLA Navy that will have well over 100 submarines, many of them nuclear powered, when HMAS Attack makes its first patrol. Another very important consideration is that as currently conceived without modern batteries or AIP, far from being ‘regionally superior’, the Attack class will not be a state-of-the-art conventional submarine in that its indiscretion ratio will make it more vulnerable when operating in a heavily congested and contested battle space. 45 This again reduces the deterrent effect, since the PLA will fancy its chances of eliminating any Australian submarine that seeks to breach its ‘underwater Great Wall’ in the South China Sea. Of course the Attack class could be designed to incorporate AIP. While this would bring a significant benefit in terms of underwater endurance and stealth while on patrol, the AIP system could be used only for a quarter of an 70-day patrol. The value of AIP, therefore, needs to be balanced against the addition of considerable weight that, other things being equal, would further increase transit times and thereby reduce patrol time on station. Aside from the fact that the intention is eventually to acquire twice as many of them as Collins, it is not clear that the Attack class will provide a more credible deterrent than currently exists. By the time that the Submarine Force has doubled in size in the 2050s, the opposing forces will have much more than doubled. It is difficult to see that the Attacks will enjoy shorter transit times than Collins and therefore their patrols will be similarly limited. One conventional submarine on station will not be a credible force in the 2030s when, according to the Minister, it will be contesting the theatre with 300 0ther submarines. The PLA Navy is forecast to have 100 submarines by 2030, but this may well be conservative. A new facility to construct nuclear powered submarines has just been completed at Hainan Island and, once the PLA Navy is happy with the design, we can expect to see new SSNs and SSBNs coming off the production line at a rapid drumbeat. The quality of PLA Navy SSNs can be expected to improve significantly, particularly now that the relationship between China and Russia has become closer in terms of sharing military technologies.

3.4 Operational and tactical capability requirements

With its fleet of diesel-electric submarines, Australia brings a different capability to the allied multi-layered force in the South China Sea and one that provides a complement to the American SSNs. Advanced SSKs like the Collins class have some advantage in terms !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 45!A!submarine’s!indiscretion!ratio!refers!to!the!time!it!spends!‘snorting’,!or!running!its!diesel!generators!at! periscope!depth!in!order!to!recharge!its!batteries.!A!submarine!is!far!more!vulnerable!to!detection!when! snorting.!RAN!submarines!have!a!high!indiscretion!ratio!in!transit!because!of!their!rapid!SOA.! 41!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? of stealth and, being less than half the size of a US Virginia class submarine, have a more modest physical footprint. Other things being equal, they are generally more difficult than SSNs to detect by passive sensors such as seabed hydrophones and towed arrays, particularly when operating at low speed, although modern SSNs are much quieter than earlier models and are closing the gap with SSKs in terms of stealth. Nevertheless, a smaller SSK is still probably better suited to operate in the shallower waters of the littoral than the large American submarines. Given our broad understanding of the role of Australian submarines play in liaison with the US Navy in the South China Sea – with a focus on ASW and ISR missions – how well suited in terms of operational and tactical capability will the Attack class be in prosecuting this task?

Counterforce ASW capability46 Any task such as intelligence gathering or detecting submarines in the South China Sea will increasingly pose a significant challenge to Australian submarines because of the rapid developments in ASW technologies and the increase in the PLA’s investment in this capability. Once they reach their AO in the South China Sea, RAN submarines will be operating in the most heavily contested, high intensity subsea environment in the world. China’s strategy of seeking to enforce A2/AD on and above the surface of the South and East China Seas, with the militarisation of disputed islands and the installation of advanced anti-shipping missile batteries, is now being matched in the sub-surface domain. Only recently, the PLA’s capability in ASW was rated as relatively poor. To some degree, this may have allowed some complacency to develop on the part of allied submarines operating in the South and East China Seas. But it would be remiss for a nation to design a new submarine capability to be deployed in 2036-2050 on the basis of an assessment of inferior ASW capability on the part of the adversary at this point in time. ASW capability is now an evident priority for the PLA and appears to be developing rapidly. The Chinese media highlights the development of an ‘Underwater Great Wall’ in the South China Sea aimed at denying access to allied submarines. Seabed sensors are being installed, along the lines of the American SOSUS network. Having developed advanced technological solutions to submarine detection, new large destroyers and frigates, together with small corvettes, all equipped with advanced ‘two- tailed’ sonars, including linear towed-arrays (TASS) and towed variable depth sonar (VDS) as well as hull mounted sonar, are joining the fleet at a rapid drumbeat. Deploying two-tailed sonars on small corvettes (1,500 tonnes) is innovative as the ships can be turned out in quantity and absorb a higher degree of attrition with less impact on overall capability.47 The large platforms deploy new heavy helicopters, one model apparently reverse engineered from the American M-60 Seahawk, with chin radars optimised for detecting snort masts at ranges up to 35 nautical miles. In terms of anti-

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 46!The!information!in!this!section!is!sourced!primarily!from:!Joe,!Rick!(2018a),!“The!Chinese!Navy’s!growing! antiGsubmarine!warfare!capabilities”,!The&Diplomat,!September!12,!https://thediplomat.com/2018/09/theG chineseGsurfaceGfleetsGgrowingGantiGsubmarineGwarfareGcapabilities/,!and:!Joe,!Rick!(2018b),!“Chinese!antiG submarine!warfare:!aviation!platforms,!strategy!and!doctrine”,!The&Diplomat,!October!16,! https://thediplomat.com/2018/10/chineseGantiGsubmarineGwarfareGaviationGplatformsGstrategyGandG doctrine/!!! 47!Australia!has!chosen!a!different!philosophy!by!focussing!its!ASW!capability!in!the!surface!fleet!on!a!smaller! number!of!large!(7,000!tonnes)!and!very!expensive!Type!26!frigates.! 42!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? submarine weaponry, PLA Navy warships deploy light torpedoes as well as a rocket system, suspiciously similar to the American ASROC, that reputedly can engage a submarine 30 nautical miles away, a safe distance from a possible counterattack. 48 China is also deploying more civilian manned ships with TASS, a direct copy of the American SURTASS platforms, although currently they employ them in waters close to home rather than up threat. While China still lags behind the west in terms of ASW helicopters and MPAs, the sensors being deployed on new platforms appear to be sophisticated and this state of affairs is unlikely to be the case in the late 2030s by the time the Attacks are to be deployed. The PLA appears to be investing significantly in developing submarine detection and tracking capabilities, in particular using advanced new technologies. Chinese universities are heavily involved in this research. For example, as part of the ‘Sea Watcher’ programme, it has been reported that the Shanghai Institute of Optics and Fine Mechanics has recently tested a laser device that can detect objects more than 160 metres underwater.49 Looking forward to the 2030s, ASW technologies are developing rapidly, including the use of satellites and electromagnetic sensors for submarine detection. In addition, while the USA leads the way in the use of uninhabited underwater vehicles (UUVs), including their employment in an ASW counterforce role, both Russia and China are investing heavily in this capability, According to Sylvia Mishra of the Institute for International Science and Technology Policy at George Washington University:

UUVs increasingly play a critical role in antisubmarine warfare (ASW) and perform missions such as placing and monitoring sensors on the sea floor to track enemy submarines. They can gather intelligence on opponents, detect and neutralise mines, hunt submarines and chart the ocean floor. They could, potentially, detonate warheads. And they could take part in a coordinated attack on an enemy submarine in conjunction with ‘friendly’ submarines and surface vessels. The United States, Russia and China are investing in this technology to bolster their ASW capability and it’s evident that UUVs will be deployed in the near future in combat operations.50

There is one important area in ASW where PLA Navy still falls short of western capabilities, namely the use of attack submarines to operate up threat and seek and destroy other submarines. This probably reflects the fact that the focus of naval capability development has been on enforcing A2/AD in waters closer to home but also that designing a competitive SSN has taken some time. This gap in capability is unlikely to remain as the PLA Navy acquires more nuclear attack submarines and begins to flex its muscles on a larger blue water canvas. Long before the Attack class enters service, it would be no surprise to discover a PLA Navy submarine loitering around the HMAS

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 48!Goldstein,!Lyle!J.!(2015),!“A!frightening!thought:!China!erodes!America’s!submarine!advantage”,!The&National& Interest,!17!August,!http://nationalinterest.org/feature/frighteningGthoughtGchinaGerodesGamericasG submarineG13592?page=1! 49!Peck,!Michael!(2019),!“China!is!trying!to!use!lasers!and!magnets!to!‘unstealth’!US!Navy!submarines”,!The& National&Interest,!9!October,!https://nationalinterest.org/blog/buzz/chinaGtryingGuseGlasersGandGmagnetsG unstealthGusGnavyGsubmarinesG86846!! 50!Mishra,!Sylvia!(2019),!“Could!unmanned!underwater!vehicles!undermine!nuclear!deterrence?”,!The& Strategist,&8!May,!https://www.aspistrategist.org.au/couldGunmannedGunderwaterGvehiclesGundermineG nuclearG deterrence/?utm_medium=email&utm_campaign=Daily%20The%20Strategist&utm_content=Daily%20The%2 0Strategist+CID_30e4b5ced72f9500c81e225dfd28896e&utm_source=CampaignMonitor&utm_term=Could%2 0unmanned%20underwater%20vehicles%20undermine%20nuclear%20deterrence& 43!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Stirling base near Fremantle ready to track an Australian submarine setting off on an operation. Australian submarines in transit to their AO are also liable to be interdicted by an adversary’s attack submarines as they pass through choke points, particularly if they need to snort.

Growing risks to large conventional submarines As discussed above, technological advances in ASW will make operations in contested waters more dangerous for all populated submarines, not just diesel-electric boats. Andrew Davies concluded five years ago that:

“The risks of operating large submarines in contested spaces will likely be higher— possibly much higher—in the future than is the case today … It seems to me that the combination of high-power processing and robotic systems is likely to make submarine operations—at least in the form of sneaking large platforms into contested spaces—much more difficult.”51 Davies concluded that future submarines will need to: 1. “Operate away from chokepoints and contested spaces but be able to project influence into them 2. Have a low indiscretion rate 3. Be a hub for a suite of long-range sensor and weapon systems 4. Be networked with other units, including electronic warfare platforms and systems 5. Be able to manoeuvre quickly in response to a rapidly changing threat environment.” 52

Davies’s analysis is of critical importance for the Attack class. He states that “the design of the future submarine has to be cognisant of these trends, which will make penetration of adversary space or operations in contested chokepoints by the submarine itself very much harder. Basing our investment around traditional ideas of submarine operations isn’t likely to be a winning strategy a couple of decades from now.” Australia needs “to decide whether our subs are going to play in the highest end operations. If we decide we need to, we’re necessarily going up the risk reward curve for a conventional boat.” The other option would be to “temper our ambitions and settle for a fleet that can still operate effectively in less than the most challenging situations”.53 If we look at Davies’s five conditions for successful submarine operations above, neither the Collins class nor, on the basis of what we understand about its design, the Attack class, rate highly overall. In terms of the first criterion, the raison d’être for RAN submarines is to transit through chokepoints so as “to play in the highest end operations” in their increasingly challenging AO. Yet without flexibility in its design so as to enable adequate provision for deploying autonomous systems (Criterion 3) in the future, the Attack class will be unable to act as an effective hub for launching UUVs into heavily contested spaces where it may not choose to venture itself. As Malcolm Davis of ASPI wrote recently: The navy is currently focusing on small, tube-launched unmanned underwater vehicles that can undertake basic mine-clearance tasks or surveillance. But that emphasis may prove to be misplaced if such systems are overtaken by much more capable platforms that can operate entirely independently from !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 51!Davies,!Andrew!(2014),!“Trends!in!submarine!and!antiGsubmarine!warfare”,!ASPI,! https://www.aspistrategist.org.au/wpGcontent/uploads/2014/04/ASPIGsubmarineGconferenceG2014G Davies.pdf! 52!Ibid.! 53!Ibid.! 44!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

manned submarines. Fully autonomous unmanned underwater and surface vehicles with global range, extended time on station, and sophisticated sensor and weapons capabilities are likely by the mid-2030s, at relatively low cost and without the personnel challenges that manned naval vessels bring.54

Indeed, it is understood that the Attack class design allows only for the launch of a necessarily small UUV via a standard 533mm torpedo tube. In regard to Criterion 4, which prescribes a multi-layered and networked approach to their operations, Australian submarines currently satisfy this condition but only when they operate within the American C3I network. If the RAN were required to become self-reliant in defending Australia and no such network were available, the nature of our submarine operations may have to change in a fundamental way. More recently, Andrew Davies has updated his earlier jeremiads about the future prospects for manned submarines more generally and conventional boats in particular:

The proliferation of unmanned air, surface and underwater systems that we’re already seeing will pose a much greater risk to covert conventional submarines. Aircraft in the class of the MQ-4 Triton fly higher, faster and for longer (up to 30 hours) than the manned patrol aircraft of the past and they have much better sensors, being more capable of detecting submarines and far less predictable than satellites. Unmanned surface vessels currently in development have essentially unlimited endurance, and high-endurance unmanned underwater systems are being trialled in several countries today. Worse, from the submarine’s viewpoint, is that all of those platforms, perhaps augmented by fixed acoustic arrays, can be networked together to produce a persistent wide area system from which it will be difficult to hide—or even detect. A submarine commander keen to recharge the boat’s batteries has little chance of counter-detecting silent passive sensors.55

But perhaps the main risks to the future operational and tactical effectiveness of the Attack class submarines are in regard to Davies’ earlier Criteria 2 and 5, namely their indiscretion ratio and sustainable speed. These are addressed in more detail below.

Problem 1: Inferior indiscretion ratio While in the past conventional submarines, like Australia’s Oberon class, undertook patrols extending over many months, they operated in a more benign strategic environment where they were able to operate mainly on the surface. Since then, substantial advances in submarine detection technologies have occurred which, in turn, has fostered countervailing advances in stealth. In order to reduce or even eliminate a submarine’s indiscretion ratio while on patrol, AIP or, more recently, modern energy- dense batteries have become widely adopted in contemporary SSKs. We understand that Japanese submarines, which generally operate in the East China Sea, undertake short patrols of about two weeks in which they operate dived throughout. Australian submarines cannot do this. Indeed the ship’s company will arrive in what is now a highly contested AO after more than two weeks dived at sea having endured

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 54!Davis,!Malcolm!(2019),!“Unmanned!systems!are!the!future,!and!Australia’s!navy!needs!to!get!on!board”,!The& Strategist,!16!October,!https://www.aspistrategist.org.au/unmannedGsystemsGareGtheGfutureGandGaustraliasG navyGneedsGtoGgetGonGboard/! 55!Davies,!Andrew!(2019),!“Australia!should!bring!forward!planned!submarine!technology!review”,!The& Strategist,!22!October,!https://www.aspistrategist.org.au/australiaGshouldGbringGforwardGplannedGsubmarineG technologyGreview/! ! 45!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? regular snorts along the way. We would do well not to underestimate the intensity of the covert ISR missions they are then tasked to undertake. A significant indiscretion ratio, and the associated need to snort, has two important implications in terms of Australian submarines’ operational and tactical effectiveness. First of all, if engaged in an intelligence gathering operation, a conventional submarine submarine may have to abandon the mission in order to find a safe place to snort. This clearly reduces its effectiveness. The second implication of a high indiscretion ratio is that the submarine becomes much more liable to be detected and, in wartime at least, attacked. When snorting, the substantially greater noise signature of diesel generators relative to the main motor means that the submarine becomes more exposed to detection from seabed hydrophone sensors and sonar employed by surface ships, other submarines, sonobuoys dropped from MPA and helicopters. The snort mast is also liable to be detected by radar deployed on MPAs, surface ships and helicopters. All these detection technologies have improved significantly in recent years. The same can be said for magnetic anomaly detection, undertaken mainly by MPA, a technology that is more effective when the adversary’s submarine is operating close to the surface. Finally, wave anomaly detection is also a contemporary ASW technology deployed by satellites that is becoming more effective, again particularly when a submarine is snorting close to the surface. While all these significant threats are occurring, a submarine also loses much of its ability to detect an attacker because its own sonar becomes less effective due to self-noise generation. With operations in the South China Sea becoming increasingly intense, a rapidly developing ASW capability on the part of the PLA Navy will make life much more difficult for allied submarines, particularly those with limited dived endurance and a significant indiscretion ratio. It is difficult to see how the weakness of Australian conventional submarines on patrol in the South China Sea – their requirement to snort in a restricted and highly contested battle space – can be countered by the support of the US Navy. Although a RAN submarine will have access to available intelligence on the whereabouts of the adversary’s naval platforms when they snort, they could be detected by a MPA, helicopter or seabed sonar array and would then be effectively on their own. It could well be argued that operational conditions in the South China Sea have already reached a level of intensity that make it more challenging to continue to deploy Collins class submarines in their principal AO. It is notable that even in 2008 the Submarines Institute of Australia (SIA) was becoming concerned about the survivability of RAN submarines operating up threat without AIP: There is a rising risk from deploying Collins, without Air Independent Propulsion, into high threat, sensitive areas as regional ASW capabilities rise. Given the lead- time required to overcome this shortcoming, urgent resolution is required to assess when this risk becomes unacceptable and to determine the best option for avoiding the resultant capability gap.56 Twelve years later, not only has the perceived ‘shortcoming’ not been resolved, urgently or otherwise, but the intensity of operations in the South China Sea has increased !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 56!Submarine!Institute!of!Australia!(2008),!Submission!to!Defence!White!Paper!Community!Consultation!Panel,! Australia’s&Strategic&Sting,!(August),! http://www.defence.gov.au/whitepaper/2009/submissions/14_Various/SIA.pdf,!page!5.! 46!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? considerably while the PLA has developed a much more sophisticated counterforce ASW capability deployed on many more platforms. Yet in the mid-2030s, when the Attack class enter service, we must assume that the intensity of operations in their principal AO will only have increased much further and the risk of attrition will be greater still. As presently specified, the Attack class will have neither AIP nor modern, energy-intense batteries that provide superior dived endurance. They will also have an inefficient pump-jet propulsion system, employed for stealth reasons on power-rich nuclear submarines but not used on any other operational SSK. Their indiscretion ratio will likely be unacceptably high, both in terms of their operational effectiveness and their survivability.

Problem 2: Low sustainable speed The second problem, not only with Australian submarines but with SSKs in general, is their limited sustainable speed. While they may be able to achieve a maximum burst dived speed of around 20 knots, even with modern, energy dense batteries, this cannot be maintained for very long without running down the battery and needing to snort. Indeed, a SSK’s motor will only run at its highest power setting for a limited time, usually about an hour, after which, if unabated, it may suffer catastrophic failure. Although submarines using AIP can have a lengthy dived endurance of up to three weeks (shorter in the warm tropical waters in which RAN boats operate), their maximum speed when running only on AIP is around five knots. Of course they can go much faster, but this depletes the main battery very quickly. There are two broad reasons, one offensive and one defensive, why a submarine undertaking ASW and ISR missions in a high intensity environment like the South China Sea requires a high sustainable speed. In offensive terms, by the time the Attack class becomes available, many of the PLA Navy’s submarines will be nuclear-powered. A conventional submarine such as a Collins or Attack class submarine would be ineffective in trailing a nuclear submarine because of its low sustainable speed, as well as its need to snort. The SSK would also be unsuitable for pursuing high value surface targets, because of its inability to keep up. In addition, navies that operate SSNs increasingly allocate one or two to a task force or battle group. An SSK cannot operate effectively with a battle group based on surface units with a much higher SOA. Turning to the defensive scenario, while passive sensors (sonar and hydrophones) positioned on the seabed or embodied in military assets provide the ever-present system designed to detect submarines, they do not provide an exact location. But ASW submarines, ships or aircraft can follow-up the detection of a submarine on passive sensors by using their active sonars to provide an exact fix, thereby allowing them to deploy their weapons so as to destroy the submarine. The difficulty is that if located by active sonar, the crew of the submarine will be aware of this immediately as a result of the sound of a ‘ping’ against its hull. In defensive terms, the ability to break contact when detected and exit the area at sustained high speed means that a fast submarine, which effectively means a nuclear powered boat, would have a greater chance when detected of escaping. A submarine leaving the danger zone at a speed exceeding 30 knots, despite also making a lot more noise, would also have a greater chance of evading torpedoes or other missiles launched against it. Other things being equal, the bottom line is that in these circumstances of high intensity operations a nuclear powered

47!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? submarine, as well as being more effective, would offer a higher level of survivability than a conventional submarine when detected and attacked.

3.5 Would the Attack class be suitable for a different role?

A major risk for the Attack class is that if it is being designed for a specific purpose, namely coalition operations up threat in the South China Sea, what would happen if that purpose were no longer relevant and the submarines were required to take on a different role? When Attack class submarines join the fleet in 15 years time, Australia’s strategic circumstances may be very different from how they look today. In his latest book, Professor Hugh White states that in the future “Australia would be very unwise to assume that America will play the role in stabilising Asia and defending Australia that our defence policy still assumes it will”.57 Many Australian strategic analysts agree with him, while others do not. In looking at the future role of the Attack class we do not need to take sides in this debate, but Defence planning needs to take account of the possibility that White may well be correct in his judgement. In that case, what are the implications for the Attack class were the US to withdraw its military forces east of Hawaii? The first implication is that without the benefit of the US multi-layered ASW networks, including the extensive range of assets, the C3I system and water space management regime, Australian submarines would be unable to operate effectively up threat in the South China Sea. Indeed, absent the support of US forces in the theatre, the survivability of a lone Australian submarine would be in very serious doubt. This means that even if the bare bones of the network remained in place, confronted with a far stronger opposing force, a lone Australian conventional submarine could not responsibly be deployed to that theatre. The other implication of an effective US withdrawal is that even if we forged new alliances and ententes in the region, Australia would need to become more self-reliant in its own defence. The main strategy would likely be to seek to deter an attack on Australia by enforcing A2/AD in the air-sea gap to our north. Would a force of 12 Attack class submarines be effective in helping to prosecute this strategy? In operational terms, the short answer to this is “No”. Contemporary doctrine among the major naval powers has nuclear-powered attack submarines operating with surface task forces and battle groups, although generally well removed in terms of distance. Any SSK such as the Attack class would be too slow for this role as they could not keep up with warships for any length of time. In addition, were RAN submarines to be tasked with interdicting and destroying an adversary’s fleet attacking Australia via a choke point strategy, there would also be significant problems. First, with a force of perhaps three or four submarines on station, the available force would be insufficient to cover the large number of chokepoints available to a marauding fleet. Secondly, the Attack class would again be too slow to be able to respond to evolving and developing threats – perhaps a sudden change of course by an attacking force – or to be able to pursue high value targets such as nuclear powered submarines or aircraft carriers. Tactically, SSKs

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 57!White,!Hugh!(2019),!How&to&defend&Australia,&La!Trobe!University!Press,!July,!page!15.! 48!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? like Attack would also be vulnerable to an adversary’s SSN loitering silently around a choke point waiting for a RAN submarine to snort. In order to prosecute such a strategy, the RAN would either need a force of much faster submarines, namely SSNs, or a lot more (24 to 36) SSKs. The latter solution would be very difficult in terms of shipbuilding, maintenance and crewing, while there would still be significant questions around operational and tactical effectiveness as well as survivability if detected..

3.6 Will the Attack class be cost-effective?

The whole-of-life cost of acquiring and operating 12 Attack class submarines is estimated to be $225 billion in out-turned dollars. In return for this, in peacetime the Navy will be able to deploy one diesel-electric submarine at all times to operate in coalition with the US Navy in the South China Sea. In strategic terms it is very difficult to see how deploying this capability can be cost-effective, either in terms of providing a deterrent in its own right or in the weight of its contribution to American operations up threat. The opportunity cost of this capability, in terms of alternative means of power projection, such as long-range stealth bombers like the B-21 Raider, merits detailed consideration. The Attack class’s effectiveness in absolute terms in high intensity AOs will be limited because it will be operating in what is already a highly contested and congested space. By the late-2030s, HMAS Attack will be one of three hundred submarines (even if only half of them are available for operations) expected to be in the Asian theatre, many of them deployed by the PLA Navy and many of them nuclear powered. Within fifteen years, it is also likely that UUVs will be playing a major role, not only in anti-submarine warfare but also in force projection more generally. Developments in ASW suggest that all manned submarines will be more vulnerable to destruction in 20 years time than they are now and at much greater risk when operating in congested spaces subject to numerous, sophisticated arrays of sensors. Conventional submarines will be at greater risk because of their periodic need to snort and their low sustainable speed when they need to break contact with an adversary. By the time the Attack class comes on stream, the majority of conventional submarines operating in the South China Sea will embody modern energy dense batteries and/or AIP. Most or all of their operations in heavily contested waters will be undertaken dived throughout. On the basis of its current design, the Attack class will not deploy AIP or modern batteries. While a very large conventional platform, it is not being designed with the redundancy necessary to operate as a hub for new generation autonomous vehicles except small and relatively ineffective UUVs. With a large footprint, a significant indiscretion ratio and a low sustainable speed, it will be significantly less effective than nuclear powered submarines, with a lower level of survivability and consequently a heightened risk of attrition.

3.7 Conclusion

Australia is the only nation operating conventionally powered submarines that sends its submarines across oceans to get to the operating area. All other countries that face that challenge deploy nuclear attack submarines (SSNs). There are serious questions over whether a diesel-electric submarine relying on lead-acid battery technology (at least at 49!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? the beginning), will be fit for purpose in terms of effectiveness and survivability against the assets deployed by the potential adversaries Australia will face when it is delivered. This needs urgent and detailed examination. The decisions taken so far on such things as battery technology are not the issue. The question is more whether any conventional submarine, regardless of technology, will have sufficient utility for Australia, including the issue of survivability as well as effectiveness, to make the massive planned investment worthwhile at all. Dr Andrew Davies of ASPI has been a persistent critic of the Navy’s future submarine program. His recent conclusion should give us pause for thought:

All of the technological trends are against us, as is our unique operational requirement of having to negotiate choke points and transit long distances across open oceans to reach far-flung mission areas. We are investing many billions of dollars to get small, incremental improvements in stealth, range and endurance while the counter technologies are on the cusp of massive, and potentially relatively cheap, increases in performance. Based on all that, I’m betting now that the Attack program sees major changes or comes to a premature end long before the planned delivery of boat 12 in 2050.58 That aside and even if Australia continues down the current path, there is no evidence of any process either in place or yet contemplated for introducing new technology into the Attack design during the decades-long build. This is a critical component for the success of a continuous build program. Without such a process, including detailed planning to introduce selected new technology and capabilities, we are exposed to a risk of equipment and capability obsolescence as the build program progresses. This risk will increase over time and grow very quickly in some technologies. The review in the late 2020s of submarine technology mentioned in the 2016 Defence White Paper should be started immediately. Such a review also needs to factor in the Collins LOTE.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 58!Davies!(2019),!op.&cit.& 50!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Chapter 4

The way forward: a risk mitigation strategy

The analysis presented in this report has identified some very substantial risks around the SEA 1000 future submarine acquisition program • The main problems with the Attack class program are that delivery will be far too late, perhaps leading to a major capability gap, and that it will be much too expensive, leading to significant pressures on the Defence budget. But a number of other problems are emerging, such as the lack of Australian industry content, which cannot be easily resolved in the ongoing absence of competition. • Not only will the proposed Attack class fail to satisfy the value for money test in terms of cost-effectiveness test, but by the 2030s it may not be fit for purpose in the role for which it is being designed. We propose two studies that should begin urgently: • A preliminary design study for an evolved Collins class submarine • A review of future technology requirements for Australian submarines. In combination, we believe these studies would help clarify options to mitigate the risks with the current program.

4.1 Collins 2.0 preliminary design study

This proposed study would re-introduce competition that would help bring issues with the Attack program to a head. These include cost, schedule, local content and rights to IP. It would also seek to exploit the full breadth of the nation’s hard-won experience with the Collins class submarines, which in our view was abandoned prematurely. It takes account of important facts: • The service lives of all six Collins boats must be extended by a full operating cycle of ten years, at least, to avoid a capability gap. • These submarines and the Australian industry supporting them are performing well after a troubled start. Australia has learned a great deal from that experience, lessons we now face learning again with a supplier with whom we have no experience apart from a rocky beginning, which does not augur well. • The Swedish designer of the Collins class, Kockums, with whom we have long experience, is now in a healthy state. It has undertaken complex life extension work, including inserting new hull extension sections into the older submarines in Sweden’s own Navy and is building the first of a new class of submarines for Sweden. It is also offering an evolution of the Collins design in competition with French and German bids to replace the current Dutch submarine force of four Walrus class submarines, which itself was a contender for the Collins acquisition. • Importantly, after a long period of German ownership, Kockums is now owned by Swedish company Saab, with whom Australia has developed a very sound and

51!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

productive long-term relationship and with whom we are contractually engaged for major future naval acquisitions stretching decades into the future.

Evolving Collins – Collins 2.0 Creating a new submarine design by evolving the existing, in-service platform, thereby building on the lessons learned from building, operating, sustaining, and modifying it, is the approach followed by all countries that are successful in building submarines for their own use. Overall, it is an approach that involves lower risk and cost. This may sound like what Australia has done with the Attack class, but it is not. The design is an evolution of France’s submarine design expertise, not their experience with operating a SSK (the Marine National only deploys nuclear-powered submarines) or ours with an existing design. Because France exports but does not operate conventional submarines, the Attack class will incorporate France’s lessons predominantly from its export customers. It will not incorporate Australia’s, except to the extent that Australia has specified the performance requirements. There is evidence and analysis to suggest that a design evolved from Collins (Collins 2.0) and building on the additional experience that will be gained from the Collins LOTE would provide a comparable capability to Attack but could be delivered years earlier at lower cost than Attack and with much higher local content from the outset. There would also be significant and valuable synergies between Collins LOTE and Collins 2.0 in such areas as submarine and logistics systems and component commonality that would benefit both industry and Navy.

The proposed study We propose that without interrupting or occasioning any delays to the Attack class program, the government should commission a parallel preliminary design of a Collins 2.0 option. We estimate the cost would less than $100m or around 0.2 per cent of the future submarine budget. This could be recovered many times over even if only by the introduction of competition to the current unpalatable reality. The government could then choose between the Attack class and Collins 2.0 in 2022-23. In addition to capability, the decision should be based on fixed price tenders for the first six submarines, including firm proposals for delivery, local content and intellectual property rights as well. The viability of this proposal necessarily depends on Saab Kockums having the capacity to devote the necessary capability to the task. Our assessment is that there is a window of opportunity now but there is little spare capacity in the global submarine business at present and other commercial realities may intrude. The window may be short-lived, particularly if Saab Kockums win the Dutch submarine program.

Why Collins 2.0? Given that it was designed specifically for the RAN, the Collins class is the only existing conventionally powered submarine design that comes close to satisfying Australia’s operational requirements, which are driven mostly by geography. Most importantly perhaps, after many years sorting out problems, it is now accepted that the Collins class is most assuredly not a “Dud Sub”. It is a highly advanced, blue water conventional submarine.

52!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Early in the Future Submarine Program, Australia was encouraged by the US and others to build on the experience gained from 30 years of the Collins program. Overseas practice reflects this approach because an evolutionary program is cheaper, less risky and more quickly achieved than ‘starting from scratch’, as we have effectively chosen to do with SEA 1000. Nations that are long-term parents of their own submarine capability run their programs in this way, evolving what came before. With the Collins program and our unique submarine, Australia took on these parental obligations whether it understood that at the time or not. Undertaking a Collins 2.0 preliminary design would necessarily involve Saab Kockums. The synergies that could be achieved with Collins LOTE as a result are significant. As Defence Minister, Christopher Pyne stated that evolving the Collins design was considered and dismissed by Defence some years ago, but this is not the whole story. Defence did indeed study updating the Collins design but this was done prior to the competitive evaluation process (CEP) that Australia used to select between France, Germany and Japan as the designer of its new submarines. Sweden was not invited to submit a proposal during the CEP. The publicly stated reasons for this decision may have been relevant then (although arguably they were not) but there is plenty of evidence showing the situation is very different now. Importantly too, agreement on intellectual property (IP) rights was negotiated with Sweden before any work started on investigating an evolved Collins design. This agreement is presumably still valid and even if it is not, it provides a very sound basis for a new agreement. Collins 2.0 is a good Plan B, if we are to get as close as possible to meeting the capability requirement. We would argue it’s the only viable option. • German and Japanese designs were evaluated during the CEP and rejected; there is no reason to revisit them • Government has ruled out ‘off the shelf’ conventional submarine designs offered in the market (by France, Germany and Sweden, plus Spain and South Korea if their designs were to be available commercially) because they are much less able than Collins to meet the capability requirements. • With nuclear powered submarines not being considered, there are no alternatives available from the UK or USA. Other government policy objectives, such as a high level of Australian industry involvement as an enabler of greater sovereign control over the capability and control over through life costs, are of fundamental importance and underpin the analysis contained in this proposal. Building on over two decades of experience in sustaining the Collins class, with a local industry content of over 90 per cent being achieved today, Australia and the existing national supply chain brings with it a deep understanding of those submarines, as well as operating and supporting them. This knowledge is a valuable sovereign asset, the benefits of which are maximised by the approach proposed here. They should not be abandoned lightly although we are at risk of doing just that. Early experience with France is showing just how hard it will be to rebuild all that from a zero base with a country very different culturally from us and with whom we have no experience working on submarines.

53!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

There are other reasons for examining Collins 2.0. Defence has done a worthwhile body of work on it already, having engaged Kockums in 2013 to investigate Collins 2.0 through contracts valued at just under $10m. This was in addition to work Kockums undertook to evaluate extending the Collins class service life. The products and the collaboration are reported to have been good. The close integration that would be possible between Collins LOTE and Collins 2.0 in a single program approach with a single design partner could reasonably be expected to be simpler and therefore lower risk than the current arrangements because it would entail much less commercial complexity (and therefore risk and cost) than exists now. Cost should be lower overall and a quicker delivery schedule should be achievable. Greater Australian industry involvement would be achieved sooner and there would be greater commonality between the old and new submarines plus substantially greater opportunities to ‘carry over’ systems and lessons from the Collins LOTE to Collins 2.0, again reducing costs. LOTE could also be used to test systems intended for the new submarines, thereby further reducing future risk. This is in contrast to the current situation that will inevitably create tension between two separate submarine designers and builders involved in upgrading and sustaining Collins on the one hand, and building the Attack class on the other. Both will be competing for the same pool of manpower and resources but with very little ability to ‘cross-level’ those resources. In addition to the benefits we foresee, the challenges and costs for Navy that are associated with transition between classes, particularly logistics and training system requirements, would be reduced. Lower sustainment costs would also be achievable from necessarily less complex supply chains. The longer-term result would be reduced total cost of ownership. We should not forget that acquisition cost (already an eye- watering amount) accounts for only around 30% of the total capability cost through life. The preliminary design study (PDS) we propose would put competitive pressure on Naval Group for a better outcome, in Australia’s interests, whichever choice government was ultimately to make. We should have no concerns about doing so, given our experience to date. Furthermore, Naval Group should have no grounds for concern about the re-introduction of competition if their submarine truly offers Australia the best capability for the price – value for money.

This proposal This proposal is for a study. We do not propose that the government should make a decision now on the future of the Attack program, quite the contrary. The preliminary design study (PDS) proposed is a low-cost, risk mitigation activity. The objective is to test whether what we believe is the only viable alternative to the Attack class is competitive and deliverable and if so, how attractive it may be, so that when the government must commit to the massive investment involved in a construction contract it can do so on the basis of much better information and therefore with greater confidence than it could have today. This would be greatly reinforced were the government to require both Naval Group and Saab Kockums/ASC to provide fixed price tenders for the first flight of submarines covering cost, schedule, Australian industry content and IP protocols. With the existing situation, there will be three choices available when deciding whether or not to proceed to a construction contract: take the French design, modify it, or leave 54!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? it. Leaving it would delay replacing the Collins boats for a long time and probably mean the end of Australia’s submarine capability, because even if Collins LOTE was delivered successfully and as soon as possible, in all likelihood those submarines could not be sustained cost effectively for long enough to allow another new submarine program to be executed. Modifying Naval Group’s proposal would necessarily involve more delay and possibly also cost and/or capability trade-offs. Taking it as it is at that time would only be viable if it’s the right overall package, to be delivered cost-effectively on an acceptable schedule with a high level of Australian industry content. In order to implement this proposal, we suggest the following actions:

• Establish immediately a small team to oversee conduct of a PDS by Saab Kockums and ASC for a Collins 2.0 designed to meet the original CEP capability requirements. The primary objective of the PDS would be to understand the capability such a submarine would provide, when delivery could begin and at what rate, and the average price we might expect to pay per submarine based, in the first instance, on a six-boat build. Other selected Australian companies in the existing Collins supply chain should be involved, to ensure the lessons of long experience from building, operating and supporting the Collins class are fed back into the design. This is a valuable way to keep manufacture, assembly and through life support costs down. • ASC engage Saab Kockums to assist with planning and executing the Collins LOTE and to maximise the potential for carry over between the two activities. An additional objective would be to understand in greater detail the benefits and challenges of running Collins LOTE and a new build program of Swedish design origin in parallel. Defence’s comments at Senate Estimates suggest that they do not have the skilled workforce to run more than one design development program. In any case, we consider that the urgency, size and importance of this program are such that diverting resources from current Defence submarine activities or causing any delay or distraction to industry or Defence should be avoided. We consider the proposed PDS needs only a small team, between 15 and 20 FTE, to set up and manage the necessary contracts. A few APS administrative support people would be required but the required specialist submarine skills could be contracted from among experienced Australians who are no longer working in Navy, CASG, Naval Group or ASC. One or two overseas experts may also be needed. This team should be separate and not draw people from any current submarine activity, in order to avoid any impact on them. Given that Defence is now heavily embedded in the design of the Attack class, evaluation of the output of the PDS and comparison of the two tender proposals would also need to be managed very carefully to avoid conflict of interest and bias, conscious or unconscious. We are confident that the cost of the PDS would be under A$100m, or about 0.2 per cent of the budgeted cost for Attack acquisition. If started very soon an interim report could be expected in 2021-22, potentially to coincide with the preliminary design review for the Attack class. A final report in 2022-23 would be available before Government faces the critical decision point of committing to a construction contract on the Attack class.

55!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

4.2 Review of submarine technology requirements

In the 2016 Defence White Paper, the government flagged a review of future submarine technologies in the late 2020s. We believe there is a pressing need to start this review now. No nation that faces our geostrategic situation uses conventionally powered submarines for the missions that Australian submarines undertake. It is true that some regional nations in similar circumstances to ours have conventional submarines in service, but those nations also operate SSNs. For the missions that we undertake, they would use their SSNs first. Those nations with only conventional submarines all operate relatively close to home because their geography means that’s all they need to do, while still operating offensively. Only SSNs have limitless dived endurance on patrol and can transit to the operating area at much higher sustained speed than a diesel boat. This capacity for high-speed transit without the need to charge batteries dramatically reduces the risk of detection and would translate for Australia into an ability to put more submarines on station – perhaps three times as many as with a diesel boat during a period demanding a surge in operations. Their endurance and capacity for sustained high speed also offers major advantages in survivability and operational effectiveness once on station. There is no other technology today or in prospect that can come close to delivering these performance characteristics. Exotic batteries and air independent propulsion cannot now and never will come close, either singly or in combination. As analysed at length in Chapter 3, the overriding problem with the Attack class is that it is being designed to undertake a role for which it not only fails the cost-effectiveness test but for which it may not be fit for purpose. It is not cost-effective because at a whole of life cost of $225 billion in out-turned dollars, the long transits to its principal AO means that the RAN will only have one submarine on station at any time. There are two main reasons why it is not fit for purpose. First, operating in a theatre where the intensity of the undersea contest is now greater than anywhere else in the world, the Attack class’s high indiscretion ratio will be an increasing problem. It means not only that it will be relatively ineffective in certain missions when it needs to break contact to snort, but at a time when advanced ASW technologies are developing and being deployed at a rapid rate, it will be more vulnerable to detection, reducing the level of survivability and increasing likely attrition, perhaps to an unacceptable level. Secondly, with a low sustainable speed, the Attack class will not be able to work with task forces/battle groups or track and pursue faster, high value targets. Its survivability will also be compromised because it will also have a much lower ability than a nuclear powered submarine to break contact when detected and escape to fight another day. In light of these findings, there is a stark choice. If the government wants a submarine force that can operate effectively in the high intensity environment of the South China Sea in the 2030s and beyond it needs to evaluate the acquisition of nuclear powered submarines, increasingly complemented by AUVs. The alternative is to accept that the Submarine Force can no longer operate in high intensity environments. This would lead either to a change in the role of RAN submarines or, conceivably, a view that a powerful submarine force would no longer be cost-effective, and that in order to retain a credible

56!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? power projection capability, Australia needs to acquire alternative assets, such as long range strike aircraft and missiles. In terms of the first option, were the RAN to acquire twelve SSNs, the productivity and strategic effectiveness of the Submarine Force would be transformed. With transit times more than halved, there would be at least a 50 per cent increase in the number of submarines that could be on station ‘up threat’ in the South China Sea at any time. A force of that size could provide a significant deterrent, with the potential to create havoc around an adversary’s naval bases, not just by attacking hostile submarines, including SSNs, but surface ships as well. Also, the deterrent effect of twelve SSNs compared to twelve SSKs would be far greater than is implied by the larger number of submarines deployed on station, because of their much greater operational effectiveness conferred by their high speed and unlimited dived endurance in any operation. In the context of the strategic contest between the US and China for hegemony in the Asia-Pacific region, as Hugh White and others have suggested, there is a material possibility that in the next two decades the US may withdraw from the regional contest and focus on sustaining its strategic position globally. If that were the case, Australia may be required to focus on a self-reliant strategy that would involve, inter alia, pursuing an anti-access and area denial (A2/AD) policy in the extensive waters to our north. The Attack class would be sub-optimal in this role because of the submarines’ limited sustainable speed that reduces its flexibility and removes its ability to operate with a task force. In order successfully to pursue high value targets and be able to interdict a seaborne attack on Australia by a fleet that may feint its advance in one direction and then change course to another, a fleet of SSNs would be required because their much higher speed. A fleet of SSNs would constitute a far more powerful force than a similar sized fleet of SSKs. Given Australia’s deteriorating strategic circumstances and the long transits to and from our submarines’ AOs, there is a clear strategic need for a powerful force of fast submarines, which effectively means a requirement for SSNs. We need to be able to put a greater submarine capability on station for a substantially greater length of time. There is also a strong operational requirement for a submarine that can undertake missions in high intensity tactical environments without having to expose itself to detection by snorting and with the speed not only to be more effective in offensive missions but also with a better ability to disengage when required and withdraw in good order. Finally, the issues around survivability and a reduced risk of attrition are of critical importance. These operational imperatives also point to a requirement for SSNs. Yet acquiring nuclear-powered submarines would not be easy. Even if we started the process now, it would take between 15 and 20 years to see the first SSN in the water. We would need to acquire a minimum of six new SSKs first to build the Submarine Force up to 12 boats including six life-extended Collins. Twelve nuclear submarines may seem excessive, but expert analysis suggests that a minimum of 10-12 is required to develop the necessary critical mass. The number of personnel in the Submarine Force would need to grow by a factor of at least three, possibly four. The recruitment and training task would be prodigious, with all submarine personnel needing to be nuclear qualified to a greater or lesser degree. In addition, the through life support of nuclear submarines in Australia, which would be essential if a sovereign submarine capability was to be sustained, would require a whole new infrastructure designed to ensure nuclear

57!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B? engineering of the highest order supported by world class safety protocols. Australia’s engineering capability is often underrated, but very substantial enhancements in the area of nuclear engineering and nuclear science and safety would be required. Australia’s lack of a nuclear industry is sometimes cited as a reason why we cannot acquire nuclear powered submarines. But this can be exaggerated. Australia does have a nuclear industry, if small and underdeveloped. As we have seen, Australia is a major exporter of uranium. The Australian government owns and operates a small experimental nuclear reactor in the leafy southern Sydney suburb of Lucas Heights. It seems to be well accepted by the local community and is now moving into a new commercial market, namely the production of medical isotopes for domestic and export markets. Australia is one of only 11 countries in the world that produce the isotopes that are vital components of certain medical processes. In addition, Australia’s nuclear regulator ARPANSA is well credentialed and provides a solid base for future development.

EXHIBIT 4.1: UK ASTUTE CLASS SUBMARINE, AUDACIOUS, UNDER CONSTRUCTION

Licence: Alamy

The issue of whether Australia should acquire nuclear powered submarines is fraught. Apart from political and foreign policy challenges, there are issues of cost and implications for the force structure, which could be distorted if a large investment was made in SSNs. We would have by some distance the smallest economy of any countries that currently operate nuclear powered submarines. It would be very unlikely that the US Navy would agree to supply Australia with nuclear submarines and in any case the Virginia class and its likely successor are too big for Australian requirements. The French Suffren is a good size for the RAN and, unlike other SSNs, uses low enriched uranium for its fuel, making it an easier acquisition for the RAN under the terms of the

58!

! A USTRALIA’ S F UTURE S UBMARINE: D O W E N EED A P LAN B?

Nuclear Non-Proliferation Treaty (NPT). The French government has offered in the past to supply Australia with SSNs. If the government decided to cancel the Attack class, it would be tempting to minimise contractual break clauses by moving the design contract to an Australianised version of the follow-on class to the Suffren class, which is the reference design for the Attack class. As we have seen with the unhappy experiences with SEA 1000, however, the benefits of competition are very compelling and may well dwarf any cost savings derived from switching the current design contract from Attack to the next version of the Suffren. We would therefore need to consider at least two designs. In the next few years, Britain will be considering the successor to its current Astute class SSN. Under the 1958 US–UK Mutual Defence Agreement, the US has the right to block the sale of submarine nuclear reactors by the United Kingdom to any third party. Yet this right was waived by President Reagan in the 1980s when Canada was considering acquiring Trafalgar class SSNs, although the end of the Cold War overtook this aspiration. Under the current strategic circumstances, this right might very well be waived again to Australia’s benefit. In commercial terms, another customer for British SSNs with an order book quite possibly larger than the Royal Navy’s would be attractive and would allow Australia to have some influence over the design. In the context of Brexit and the UK’s desire to deepen trade links with Australia, this could also be an attractive proposition from the political and industrial perspective. While it may not be appropriate or even possible for Australia to acquire SSNs, the issue needs to be evaluated in detail before the nation expends a vast amount of money acquiring a submarine that may not be able to discharge its primary functions effectively. In the context of the current examination of force structure that the Minister has stated is occurring, the review of future submarine technologies announced in the 2016 DWP should be brought forward by ten years and undertaken now. The justification for this is the: • Deterioration in Australia’s strategic circumstances that has occurred in the last three years • Substantial increase in the intensity of the contest and the congestion in our submarines’ principal area of operations • Increase in investment in ASW activities on the part of the PLAN and the roll out of advanced anti-submarine technologies The backdrop to all this is the rapid growth in submarine activity generally in the Indo Pacific, with perhaps 150 submarines available for deployment in the theatre when the first Attack boat is due to enter service in 15 years time.59 These will include many nuclear-powered boats, perhaps with substantially better stealth capabilities to current designs, and conventional boats with AIP and/or modern batteries. Not only will the Attack class face a serious challenge with the numbers ranged against it, but also the quality of many of the other submarines may well pose a significant challenge to the Attack class’s claim of ‘regional superiority’.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 59!Although!it!is!regularly!stated!that!300!submarines!will!be!operating!in!the!Indo!Pacific!by!the!midG2030s,!we! assume!this!refers!to!the!total!number!in!service.!If!that!is!the!case,!regular!maintenance!schedules!mean!that! only!around!half!of!them!will!be!available!for!deployment.!! 59!