Replacing Polaris

uk 75

ACCESS: Above Top Secret
Senior Member
Joined
27 September 2006
Messages
6,049
Reaction score
6,148
The long road to replacing Polaris submarines in the 90s with Trident began under the Callaghan government in the late 70s.
The new US Tomahawk cruise missile was considered by Dr David Owen amongst others as a more affordable alternative. But the Trident required fewer submarines and was more likely to get through to its targets in the Soviet Union. It also ensures that RN and USN missile launches would be hard to tell apart.
Mrs Thatcher was elected in 1979 but it was Callaghan with his good relations with President Carter who had opened the way to Trident.

But if Thatcher had not won there were alternative paths.

An inconclusive win by Callaghan might have seen him needing the support of the Liberal Party. The Liberals favoured a smaller deterrent using cruise missiles.

Callaghan was replaced as Labour Party leader by Michael Foot. Foot was a passionate supporter of nuclear disarmament. Had Foot replaced Callaghan in Number Ten the deterrent and US nuclear bases in UK might well have been removed.

A fourth alternative was that faced with a poor economy a new Conservative government might have opted to retain the newly modernised Polaris and then phase out the deterrent to save money. Former PM Macmillan was sceptical at best about the so called independent British deterrent.
 
(OP: on my screen SPF brings up your 3 similar previous pertinent threads. What AH do you wish to explore?)

The 1947 origin of UK incurring the cost of AW was fear of one former Ally retreating to isolationism, as the other pushed towards Antwerp. Other Allies had the industrial capacity to comprehend US' 12/8/45 Smyth Report (AE for Mil.Purposes) - published by Truman in hope of dissuading Uncle Joe from diverting Reconstruction effort into AW. Canada, Oz, later FRG, Japan did exactly that: delegated that expense and accepted Uncle Sam's umbrella. Only France and UK chose to delay Reconstruction and buy some notional Independence.

But then 5 or so Nations chose to spend for that Independence: all started on IR/ICBMs, and some now have FBM/SSBNs - always, because, so far, only the Deep seems to offer certainty of Effect. The reason UK Liberal thoughts of a cheaper, good enough GLCM/ALCM did not find favour...can be seen in last week's Russian effort on Ukraine, where (15 of 18?) launches were intercepted.

The fallacy of UK CND - the Quaker thought, that my moral example (of no-AW) will shame Mr.Xi, Mr. Kim, Mr.Putin into sanity - lies in the arrogance of Brit Exceptionalism - the thought that our gesture would even be noticed in N.Korea. At each point where UK chose to double up on the investment, Ministers stared hard at the cost saving from letting this investment gently expire. Each time, Tory or Labour, they judged that...You Never Can Tell, and the cost was an affordable insurance premium.

Maybe we could explore Options if AI/Quantum/Cyber/Space...erode sub invulnerability.
 
Last edited:
The fallacy of UK CND - the Quaker thought, that my moral example (of no-AW) will shame Mr.Xi, Mr. Kim, Mr.Putin into sanity - lies in the arrogance of Brit Exceptionalism - the thought that our gesture would even be noticed in N.Korea.
Tellingly, something very like this logic was tried in the 1930s with the naval disarmament treaties, when the UK was still a top-tier power. The total effect of the UK trying to set a moral example for the rest of the world by limiting itself to 14-inch-gun battleships was.... that the UK's modern battleships were inferior to everyone else's.

In geopolitics, pragmatism is always the way to go over idealism.
 
a) Possibly but Cruise was never totally cost effective and destroying 40 cities was still the aim. Cruise couldn't guarantee that and at that stage nobody would have thought of packing out the Resolutions with 100x Tomahawks each. Trident was far more KABOOM for the money and was far more deadly given ABM fears of the time.

b) Foot was a principled man but then the day he becomes PM he gets the nuclear file and has to write the letters of last resort. Who knows whether that would change his mind? No PM yet has ever not been impressed by the enormous responsibility of the deterrent (except maybe in recent years, I shudder to think what Cameron, May, BoJo, Truss and Sunak have written/thought).

c) Any Conservative PM who isn't Thatcher but who still gets into Chicago School style of neoliberal economics is still likely to be best buds with Reagan and therefore might still get Trident at a good price.
Polaris was too old to further upgrade and the system was gone from USN service.
 
From practical point of view, Tomahawk would be a better solution for Britain. It could be launched from attack submarines - thus allowing to increase number of non-specialized units in service - and in non-nuclear version could be used for tactical purposes.
 
I found this one looking for the clip where Sir Humphrey says, "It's the nuclear missile that Harrods would sell you!"
View: https://www.youtube.com/watch?v=qVO85anasrA

About 3:15 into this one Sir Humphrey tells Hacker that Polaris isn't to deter the Russians, it's to deter the French.
View: https://www.youtube.com/watch?v=N6Et35vPYLg

Though it ends before he can says something along the lines of "For all we know we could be at war with France in another 20 years" and Hacker says something like, "France, that makes much more sense."
 
Last edited:
As the USN rearmed 12 of the Laffayette class with Trident I around 1980 was any thought given to buying some of the 192 redundant Poseidon missiles and installing them in the 4 R class SSBN?
 
a) Possibly but Cruise was never totally cost effective and destroying 40 cities was still the aim. Cruise couldn't guarantee that and at that stage nobody would have thought of packing out the Resolutions with 100x Tomahawks each. Trident was far more KABOOM for the money and was far more deadly given ABM fears of the time.
In fact, they did. Well, 80 Tomahawks, anyway. It was reckoned that four boats at sea (i.e. 320 missiles) were needed to obtain the requisite level of assurance, and the additional boats would wind up costing far more money than Trident.
As the USN rearmed 12 of the Laffayette class with Trident I around 1980 was any thought given to buying some of the 192 redundant Poseidon missiles and installing them in the 4 R class SSBN?
I believe that this, too, was considered, and rapidly dismissed - along with French missiles - in the Duff-Mason report. Which I know is available somewhere, but it isn't coming to hand right now.
 
As the USN rearmed 12 of the Laffayette class with Trident I around 1980 was any thought given to buying some of the 192 redundant Poseidon missiles and installing them in the 4 R class SSBN?
I believe that this, too, was considered, and rapidly dismissed - along with French missiles - in the Duff-Mason report. Which I know is available somewhere, but it isn't coming to hand right now.
Thanks. Is that Mason, Roy Mason of 1974-75 Defence Review fame?
 
Thanks. Is that Mason, Roy Mason of 1974-75 Defence Review fame?
The principal authors were Sir Antony Duff (Deputy Under-Secretary of State at the Foreign Office) and Sir Ronald Mason, (Chief Scientific Adviser to the Ministry of Defence). Having found the document, which was cunningly hidden in my files with the title 'The Duff-Mason Report', the options which were taken seriously enough to develop costings were:
  • An upgraded Polaris, which they called 'A4'
  • Trident C4 with a multiple (non-independent) reentry vehicle payload
  • Trident C4 with a multiple independent reentry vehicle payload
  • Tomahawk
The SLBM boats were costed with options for:
  • Four boats carrying sixteen missiles each, to maintain one on station
  • Five boats carrying sixteen missiles each, to maintain two on station
  • Five boats carrying twelve missiles each, to maintain two on station
  • Eight boats carrying sixteen missiles each, to maintain three on station
Dropping to only twelve tubes didn't save any significant amount, and developing an MRV capability for Trident C4 would cost more than using the US MIRV solution. It was seemingly only considered because there were thought to be potential political sensitivities to the UK deploying MIRVs.

The only option for Tomahawk was eleven boats (which they insisted on calling SSCNs) carrying 80 missiles, to maintain five (not four, as previously stated) at sea. This could only meet the least demanding targeting scenario, 3b, and at a cost of £12.4 million for 20 years of operation, with no redundancy whatsoever against Soviet ASW.

For comparison:
  • Four boats with MIRVed Trident C4 would cost £6.7 billion for 20 years, and could cover all targeting scenarios except the most demanding, Option 1, but would offer no insurance against the loss of a submarine on station to Soviet ASW.
  • Five boats with MIRVed Trident C4 were the minimum to cover targeting Option 1, and could meet all other options after the loss of one boat on patrol, at a cost of £7.88 billion.
  • Eight boats with Polaris A4 would be needed to cover Option 2, at the rather cost of £11.575 billion.
  • Eight boats with MIRVed Trident C4 would be able to assure coverage of the Option 1 target set even if one boat was lost, but would cost £11.3 billion.
Basically, the cheapest credible option was Trident C4. Anything else was more expensive for less capability.
 
For comparison:
  • Four boats with MIRVed Trident C4 would cost £6.7 billion for 20 years, and could cover all targeting scenarios except the most demanding, Option 1, but would offer no insurance against the loss of a submarine on station to Soviet ASW.
  • Five boats with MIRVed Trident C4 were the minimum to cover targeting Option 1, and could meet all other options after the loss of one boat on patrol, at a cost of £7.88 billion.
  • Eight boats with Polaris A4 would be needed to cover Option 2, at the rather cost of £11.575 billion.
  • Eight boats with MIRVed Trident C4 would be able to assure coverage of the Option 1 target set even if one boat was lost, but would cost £11.3 billion.
Basically, the cheapest credible option was Trident C4. Anything else was more expensive for less capability.
The above with nearly the same words in a different order.
  1. £6.70 billion over 20 years for four boats with MIRVed Trident C4 that could cover all target scenarios except the most demanding (Option 1) but would offer no insurance against the loss of a submarine on station to Soviet ASW.
  2. £7.88 billion over 20 years for five boats with MIRVed Trident C4 were the minimum to cover target Option 1, and could meet all other options after the loss of one boat on patrol.
  3. £11.575 billion over 20 years for eight boats with Polaris A4 that could cover Option 2.
  4. £11.300 billion over 20 years for eight boats with Trident C4 that could cover Option 1.
No. 2 cost £1.18 billion more (18%) than No. 1 but.
  • No. 2 met the requirements of Option 1 while No. 1 could only meet the requirements of Option 2.
  • No. 2 bought 25% more boats (5 instead of 4) but only cost 18% more.
  • No. 2 put two boats on station at all times instead of one (100% more) but only cost 18% more.
No. 4 cost marginally less (£275 million) than No. 3 and met the requirements of Option 1 while No. 3 only met the requirements of Option 2.

No. 4 cost £4.60 billion more (69%) than No. 1 but.
  • No. 4 would still meet the requirement of Option 1 if one boat was lost.
  • No. 4 bought 100% more boats (8 instead of 4) but only cost 69% more.
  • No. 4 put three boats on station at all times instead of one (200% more) but only cost 69% more.
Average costs per boat.
  • £1.675 billion - average cost of the 4 boats in No. 1.
  • £1,576 billion - average cost of the 5 boats in No. 2 - £0.099 billion (6%) less than No. 1.
  • £1.449 billion - average cost of the 8 boats in No. 3 - £0.228 billion (14%) less than No. 1.
  • £1.413 billion - average cost of the 8 boats in No. 4 - £0.262 billion (16%) less than No. 1.
No. 5 Ten Boats with MIRVed Trident C4

By extrapolating the cost of No. 4 ten boats with Trident C4 (to maintain four boats on station at all times) would cost less than £14.13 billion because cost per boat goes down the more boats that are built.

No. 5 cost £7.43 billion (111%) more than No. 1.
  • No. 5 would still meet the requirement of Option 1 if two boats were lost.
  • No. 5 bought 150% more boats (10 instead of 4) but only cost 111% more.
  • No. 5 put four boats on station at all times instead of one (300% more) but only cost 111% more.
 
Last edited:
The AH of funding 8 SSBN is quite intriguing.....
 
Average costs per boat.
  • £1.675 billion - average cost of the 4 boats in No. 1.
  • £1,576 billion - average cost of the 5 boats in No. 2 - £0.099 billion (6%) less than No. 1.
  • £1.449 billion - average cost of the 8 boats in No. 3 - £0.228 billion (14%) less than No. 1.
  • £1.413 billion - average cost of the 8 boats in No. 4 - £0.262 billion (16%) less than No. 1.
It actually doesn't quite work out like that, since the proposed options all included some level of non-recurring expenditure. Adding Boat 5 would cost £1.18 billion; Boats 6-8 would be an average of £1.14 billion each. The overall effect is modelled well by assuming fixed costs of £1.55 billion, a first-of-class cost of £1.4 billion, and a 93% learning curve effect. In fact, it's so exact I wonder if that's actually the assumption that Mason-Duff used for the Trident C4 boats.

In your case of ten boats with Trident C4, it comes out to £13.51 billion. In a nice bit of coincidence, nine boats with C4 come out to cost £12.41 billion - almost exactly the same as eleven with Tomahawk!

When the Polaris boats were being ordered, nine to keep four on station was considered as an option, but deemed far too expensive. The data points that gives are quite interesting:
  • 4 boats gives 1 on station
  • 5 gives 2 on station
  • 8 gives 3 on station
  • 9 gives 4 on station
  • 11 gives 5 on station
The figures almost line up with 50% of available boats on station, plus one under refit, but not quite - there's an extra boat needed to support either one or three on station for some reason.

Incidentally, the report dismissed Trident D5 as too capable and too expensive. As things worked out, the increased capability of D5 meant that one boat on station could cover the Option 1 target set. The reduction in cost by eliminating one boat turned out to outweigh the increase in cost from the bigger missile.
 
Average costs per boat.
  • £1.675 billion - average cost of the 4 boats in No. 1.
  • £1,576 billion - average cost of the 5 boats in No. 2 - £0.099 billion (6%) less than No. 1.
  • £1.449 billion - average cost of the 8 boats in No. 3 - £0.228 billion (14%) less than No. 1.
  • £1.413 billion - average cost of the 8 boats in No. 4 - £0.262 billion (16%) less than No. 1.
It actually doesn't quite work out like that, since the proposed options all included some level of non-recurring expenditure. Adding Boat 5 would cost £1.18 billion; Boats 6-8 would be an average of £1.14 billion each. The overall effect is modelled well by assuming fixed costs of £1.55 billion, a first-of-class cost of £1.4 billion, and a 93% learning curve effect. In fact, it's so exact I wonder if that's actually the assumption that Mason-Duff used for the Trident C4 boats.

In your case of ten boats with Trident C4, it comes out to £13.51 billion. In a nice bit of coincidence, nine boats with C4 come out to cost £12.41 billion - almost exactly the same as eleven with Tomahawk!
For what it's worth I used to be a bookkeeper and I'm aware of the concept of fixed costs. Though what if instead of having the boats built by one or two yards (Vickers, Barrow and Cammell-Laird) the Government decided to have the boats built by a larger number of yards? Would that destroy the learning curve?

I've updated my workings spreadsheet on account for the above and put it into Microsoft Paint as follows.

Polaris Replacement Options.png

No. 5 (ten boats) in the old list is now No. 6.
Edit 04.05.23 the error that @Yellow Palace pointed out in Post 19 has been corrected.​

When the Polaris boats were being ordered, nine to keep four on station was considered as an option, but deemed far too expensive. The data points that gives are quite interesting:
  • 4 boats gives 1 on station
  • 5 gives 2 on station
  • 8 gives 3 on station
  • 9 gives 4 on station
  • 11 gives 5 on station
The figures almost line up with 50% of available boats on station, plus one under refit, but not quite - there's an extra boat needed to support either one or three on station for some reason.
The greater the number of boats the greater number that can be maintained on station correlates with my copy of Jane's 1968-69. The 41st First Generation SSBN was completed on 01.04.67 and the second paragraph of the preamble to the Strategic Warfare ships says.
Thus, the Polaris submarine force is capable of providing a deterrent force of about 20 submarines carrying 320 missiles on a continuous basis. During periods of crises the number of submarines at sea can be increased by delaying and shortening refit periods.
Or in other words about 50% of the force could be maintained on station.
Incidentally, the report dismissed Trident D5 as too capable and too expensive. As things worked out, the increased capability of D5 meant that one boat on station could cover the Option 1 target set. The reduction in cost by eliminating one boat turned out to outweigh the increase in cost from the bigger missile.
For what it's worth I expected as much.
 
Last edited:
What happens if the SSBN fleet is 8 or more armed with Trident C5?
Do we see a limit here eating into SSN numbers?
Or can the objective of upto 20 SSN and 8 SSBN be achieved?
What happens if both builders are producing nuclear submarines into the 90's when the Cold War ends?

Could a greater emphasis on nuclear submarines keep the UK from the disastrous gap in design and build?
 
It actually doesn't quite work out like that, since the proposed options all included some level of non-recurring expenditure. Adding Boat 5 would cost £1.18 billion; Boats 6-8 would be an average of £1.14 billion each. The overall effect is modelled well by assuming fixed costs of £1.55 billion, a first-of-class cost of £1.4 billion, and a 93% learning curve effect. In fact, it's so exact I wonder if that's actually the assumption that Mason-Duff used for the Trident C4 boats.

In your case of ten boats with Trident C4, it comes out to £13.51 billion. In a nice bit of coincidence, nine boats with C4 come out to cost £12.41 billion - almost exactly the same as eleven with Tomahawk!
If I've interpreted the 93% learning curve effect correctly the cost reductions are greater than those in the Report.

Polaris Replacement Options - 93% Cost Reduction.png

Edit 04.05.23 the error that @Yellow Palace pointed out in Post 19 has been corrected.​
 
Last edited:
If I've interpreted the 93% learning curve effect correctly the cost reductions are greater than those in the Report.
I've modelled it such that the cost for Boat 2 is 93% that of Boat 1, Boat 4 is 93% that of Boat 2, and so forth - a 7% reduction with each doubling. This seems to be the usual modelling for learning curve effects.

boat cost c4.png

Note that the £11.575 million figure is for eight boats with Polaris A4, which would be a wholly new UK-only missile. The relevant assumptions here are a 92% learning curve, £1.28 billion for Boat 1, a 92% learning curve, and £2.85 billion fixed costs:

boat cost a4.png

For what it's worth I used to be a bookkeeper and I'm aware of the concept of fixed costs. Though what if instead of having the boats built by one or two yards (Vickers, Barrow and Cammell-Laird) the Government decided to have the boats built by a larger number of yards? Would that destroy the learning curve?
It would presumably affect the learning curve, but as this is a gross approximation of a set of figures which is broken down into five sets of capital costs, plus running costs, careful interpretation would be needed. As a practical matter, I suspect that the larger forces would need to be split between two or more build yards.
 
Last edited:
What happens if the SSBN fleet is 8 or more armed with Trident C5?
Do we see a limit here eating into SSN numbers?
Or can the objective of upto 20 SSN and 8 SSBN be achieved?
What happens if both builders are producing nuclear submarines into the 90's when the Cold War ends?

Could a greater emphasis on nuclear submarines keep the UK from the disastrous gap in design and build?
Realistically, yes: the longer term limit has been getting enough crews, which (AIUI) is what drives the current UK submarine force size.

If the RN has access to a magic money tree (unlikely) and a limitless source of crew (even less likely), then maintaining two yards gets interesting. The OTL assumption was that a single yard could build a boat a year, with a service life of 25 years, giving a fleet of 25 boats. The maximum interval between boats to avoid skill fade is, or was, 22 months, which gives a minimum fleet of 13 to 14 boats to maintain a viable industry.

With two yards in the nuclear submarine game, you'd need to sustain that build rate in both yards, implying a minimum fleet size of 27 boats - which is very close to the figure where one yard will suffice. You can, of course, design your boats around a different service life. Perhaps the 20 years of operation in Duff-Mason. Perhaps the 35 years that the USN plans on. That has obvious implications for industrial capability.
 
But in the period they certainly got to 18 SSN if my memory serves.

The argument for more submarines is usually at the expense of surface Fleet and the Invincibles.

Currently the sustainable figure is aimed at 7 SSN and 4 SSBN. Though hopefully an expansion is coming.

Arguably a shorter operational life would cut the maintenance/upgrade costs and expand the operational limits on dive/surface cycles.

However two yards can potentially permit such shorter operational life and yet possess greater capacity. Which would feed into development of new concepts.

Arguably then Post Cold War, some SSBN slots in the build could be redesigned for Virginia style conventional weapon systems. Tomahawk, POLAR, ATACMS etc....
Or a specific Spec For boat or other specialist.

8 SSBN also shifts other factors.
Such as increased PWR-2 reactor production.
 
Note that the £11.575 million figure is for eight boats with Polaris A4, which would be a wholly new UK-only missile.
Hmmmm.....
Another interesting scenario.
A UK SLBM post Polaris.
 
For what it's worth I used to be a bookkeeper and I'm aware of the concept of fixed costs. Though what if instead of having the boats built by one or two yards (Vickers, Barrow and Cammell-Laird) the Government decided to have the boats built by a larger number of yards? Would that destroy the learning curve?
It would presumably affect the learning curve, but as this is a gross approximation of a set of figures which is broken down into five sets of capital costs, plus running costs, careful interpretation would be needed. As a practical matter, I suspect that the larger forces would need to be split between two or more build yards.
In the "Real World" 2 British shipyards built nuclear-powered submarines and after 1945 there were 4 British yards that built diesel-electric powered boats.
  • Vickers, Barrow-in-Furness built 2 out of 4 Resolution class and 5 out of the first 6 SSNs.
  • Cammell-Laird built the remainder, i.e., 2 out of 4 Resolution class (Renown & Revenge) and one of the first 6 SSNs (Conqueror).
  • All subsequent nuclear powered submarines have been built at Barrow.
  • The diesel-electric powered submarines of the Explorer, Porpoise and Oberon classes were built at 4 yards: Cammell-Laird, Birkenhead; Chatham Dockyard; Scotts, Greenock; and Vickers, Barrow-in-Furness.
  • Vickers built the first Upholder diesel-electric powered submarine and Cammell Laird built the other three.
I had previously thought that if 8-10 ALT-Vanguard class SSBN (or indeed 8-10 Real-Vanguard class SSBN) were built the Government might spread the work around 3 or 4 yards for interrelated political and economic reasons. That is to show that the money being spent was benefitting a large area of the country and to reduce unemployment in economically depressed areas.

Now I think that if its 5 boats, the extra boat will be built at Barrow-in-Furness and if its 8-to-10 boats half will be built in Barrow-in-Furness & the other half in Birkenhead. Cammell-Laird would be the second yard because they had already built 3 nuclear-powered submarines and built 3 of the Upholder class in the late 1980s to early 1990s.

However, if they were built in 3 yards the third would probably be Scotts, Greenock because the last submarine it built was the Australian Oberon Otama completed in April 1978. If it wasn't Scotts it would probably be another Clyde shipyard.

If a fourth yard was brought into the programme it wouldn't be Chatham because the last submarine it built was the Canadian Oberon Okanagan completed in June 1968 and it was closed in 1984 as part of the Knott Defence Review of 1981. Therefore, it would probably be Swan Hunter (which as far as I know was the only surviving shipyard on the Tyne) or possibly a second yard on the Clyde.
 
I've just remembered that the Vanguard class was built at a slower rate than planned so the Barrow-in-Furness yard might have had the capacity to build all the extra boats even if a total of 10 was built.
 
I'd have thought Upholders would be the first cut to sustain the SSBN production.
If Cammell-Laird was building some SSBN instead during the 90's.....
 
Wait, Perfidious Albion briefly considering french missiles ? must have been M2 or M20 https://fr.wikipedia.org/wiki/Missile_M20
M4, in fact: This system will be essential a MRV with some capability, albeit well below US MIRV standards, for engaging spaced targets. Missile range is estimated at 2200 nm. A UK warhead development would be needed but we cannot now assess its scale or test requirements.
And later: Fall-back options [to C4 with MIRV] if the US refued to sell us a MIRV system, are much inferior and far less cettain; it is not possible without further study and exploration to say which ould be best as between A4, C4 with MRV, and the French M4.

I do still need to give Duff-Mason a closer reading, but there's no immediately obvious point where M4 is dismissed in favour of the other two.
 
But in the period they certainly got to 18 SSN if my memory serves.
My memory (which may be wrong) is that before the Knott defence review of 1981 the target was 20 SSN and that after it the target was 17 SSN.

My other memory (which may be wrong) is that from the middle 1960s the plan was to built one SSN per year. That would have made a total of 25 ordered to 1986 (including Dreadnought, Valiant & Warspite).

However, the actual total to 1986 was 19 (including Dreadnought, Valiant & Warspite). The last of the 19 was Triumph (the seventh and last Trafalgar class boat) which was ordered in 1986 and completed in 1991.

If 25 SSN had been built at the rate planned Nos. 21 to 25 would have replaced Nos. 1 to 5 (Dreadnought, Valiant, Warspite, Churchill & Conqueror) which by the end of 1991 were all over 20 years old having been completed 1963-71.
8 SSBN also shifts other factors.
Such as increased PWR-2 reactor production.
Another as I understand it is that from the 1970s British nuclear powered submarines were built at an uneconomic rate for the supporting infrastructure, that is, they weren't building enough boats to keep the shipyards and their suppliers working economically.

Which if correct may mean that if 25 SSN & 5 SSBN were ordered to 1986 instead of the 19 SSN & 4 SSBN ordered to that date in the "Real World" there may be a significant reduction in the unit cost of the boats. Similarly, 8-10 second-generation SSBN were built subsequently instead of the 4 built in the "Real World". The information provided by @Yellow Palace bears this out.
 
If I've interpreted the 93% learning curve effect correctly the cost reductions are greater than those in the Report.
I've modelled it such that the cost for Boat 2 is 93% that of Boat 1, Boat 4 is 93% that of Boat 2, and so forth - a 7% reduction with each doubling. This seems to be the usual modelling for learning curve effects.

View attachment 698977

Note that the £11.575 million figure is for eight boats with Polaris A4, which would be a wholly new UK-only missile. The relevant assumptions here are a 92% learning curve, £1.28 million for Boat 1, a 92% learning curve, and £2.85 million fixed costs:

View attachment 698976
From the above.
I've modelled it such that the cost for Boat 2 is 93% that of Boat 1, Boat 4 is 93% that of Boat 2, and so forth - a 7% reduction with each doubling. This seems to be the usual modelling for learning curve effects.
I thought you meant each boat would cost 7% less than the immediately preceding boat. That is No. 2 would cost 7% less than No. 1, No. 3 would cost 7% less than No. 2, No. 4 would cost 7% less than No. 3 and so on.
 
I thought you meant each boat would cost 7% less than the immediately preceding boat. That is No. 2 would cost 7% less than No. 1, No. 3 would cost 7% less than No. 2, No. 4 would cost 7% less than No. 3 and so on.
You'd think so, but apparently that's not the way economists define a learning curve.
Which if correct may mean that if 25 SSN & 5 SSBN were ordered to 1986 instead of the 19 SSN & 4 SSBN ordered to that date in the "Real World" there may be a significant reduction in the unit cost of the boats. Similarly, 8-10 second-generation SSBN were built subsequently instead of the 4 built in the "Real World". The information provided by @Yellow Palace bears this out.
Oddly enough, I've looked at the ordering plans before. Because the SSBNs were ordered in rapid succession, they throw the '1/year' calculus off, and there are several years where no boats were ordered. Filling in the gaps would give an additional four SWIFTSURE class ordered in 1968, 1973, 1974 and 1975. Two additional TRAFALGAR would also be ordered in 1980/1981 - depending on when you count TORBAY - and 1983.
 
I thought you meant each boat would cost 7% less than the immediately preceding boat. That is No. 2 would cost 7% less than No. 1, No. 3 would cost 7% less than No. 2, No. 4 would cost 7% less than No. 3 and so on.
You'd think so, but apparently that's not the way economists define a learning curve.
Which if correct may mean that if 25 SSN & 5 SSBN were ordered to 1986 instead of the 19 SSN & 4 SSBN ordered to that date in the "Real World" there may be a significant reduction in the unit cost of the boats. Similarly, 8-10 second-generation SSBN were built subsequently instead of the 4 built in the "Real World". The information provided by @Yellow Palace bears this out.
Oddly enough, I've looked at the ordering plans before. Because the SSBNs were ordered in rapid succession, they throw the '1/year' calculus off, and there are several years where no boats were ordered. Filling in the gaps would give an additional four SWIFTSURE class ordered in 1968, 1973, 1974 and 1975. Two additional TRAFALGAR would also be ordered in 1980/1981 - depending on when you count TORBAY - and 1983.
Hang on.....

4 SSNs could have been built?
Would it not be reasonable to think 4 additional SSBN could have been built instead?
And that this, while additional and thus increasing total costs, doesn't directly conflict with slots in the SSN build rate?

So arguably the AH 8 SSBN plan could be achieved with the RL 19 SSN build achieved and the planned 25 later revised to 20 figures.

Even if it is at either increased spending or cuts elsewhere .
 
Which if correct may mean that if 25 SSN & 5 SSBN were ordered to 1986 instead of the 19 SSN & 4 SSBN ordered to that date in the "Real World" there may be a significant reduction in the unit cost of the boats. Similarly, if 8-10 second-generation SSBN were built subsequently instead of the 4 built in the "Real World". The information provided by @Yellow Palace bears this out.
Oddly enough, I've looked at the ordering plans before. Because the SSBNs were ordered in rapid succession, they throw the '1/year' calculus off, and there are several years where no boats were ordered. Filling in the gaps would give an additional four SWIFTSURE class ordered in 1968, 1973, 1974 and 1975. Two additional TRAFALGAR would also be ordered in 1980/1981 - depending on when you count TORBAY - and 1983.
This is my table.

SSN Table 1958-85.png
Jane's 1969-70 says that Valiant was originally scheduled to be completed in September 1965,
but work was held up by the "Polaris" programme.​
  • I think 3 extra Swiftsure class would be ordered (1972, 1974 & 1975) for a total of 9.
  • I think 3 extra Trafalgar class would be ordered (1980, 1982 & 1984) for a total of 10.
  • My SSBN table says the 4 Resolution class that were built were ordered on 08.05.63.
  • As far as I know the fifth Resolution was never formally ordered. My SSBN spreadsheet says "The intention to build a fifth Polaris submarine was confirmed by the Minister of Defence on 26th February 1964, but this intention was rescinded by a new Minister of Defence on 15th February 1965" and I think it came from the Wikipedia entry.
 
Wait, Perfidious Albion briefly considering french missiles ? must have been M2 or M20 https://fr.wikipedia.org/wiki/Missile_M20
M4, in fact: This system will be essential a MRV with some capability, albeit well below US MIRV standards, for engaging spaced targets. Missile range is estimated at 2200 nm. A UK warhead development would be needed but we cannot now assess its scale or test requirements.
And later: Fall-back options [to C4 with MIRV] if the US refued to sell us a MIRV system, are much inferior and far less cettain; it is not possible without further study and exploration to say which ould be best as between A4, C4 with MRV, and the French M4.

I do still need to give Duff-Mason a closer reading, but there's no immediately obvious point where M4 is dismissed in favour of the other two.

Problem: Mason review was 1974 and at the time french SLBMs were M1 (1971, Redoutable) and M2 / M20. M4 came by the early 80's afaik.
 
I thought you meant each boat would cost 7% less than the immediately preceding boat. That is No. 2 would cost 7% less than No. 1, No. 3 would cost 7% less than No. 2, No. 4 would cost 7% less than No. 3 and so on.
You'd think so, but apparently that's not the way economists define a learning curve.
Which if correct may mean that if 25 SSN & 5 SSBN were ordered to 1986 instead of the 19 SSN & 4 SSBN ordered to that date in the "Real World" there may be a significant reduction in the unit cost of the boats. Similarly, 8-10 second-generation SSBN were built subsequently instead of the 4 built in the "Real World". The information provided by @Yellow Palace bears this out.
Oddly enough, I've looked at the ordering plans before. Because the SSBNs were ordered in rapid succession, they throw the '1/year' calculus off, and there are several years where no boats were ordered. Filling in the gaps would give an additional four SWIFTSURE class ordered in 1968, 1973, 1974 and 1975. Two additional TRAFALGAR would also be ordered in 1980/1981 - depending on when you count TORBAY - and 1983.
Hang on.....

4 SSNs could have been built?
Would it not be reasonable to think 4 additional SSBN could have been built instead?
And that this, while additional and thus increasing total costs, doesn't directly conflict with slots in the SSN build rate?

So arguably the AH 8 SSBN plan could be achieved with the RL 19 SSN build achieved and the planned 25 later revised to 20 figures.

Even if it is at either increased spending or cuts elsewhere .
I've added the Vanguard class to the table in Post 33. The years in the first column for them are the years they were laid down rather than the year they were ordered as I don't have those dates.

SSN Table 1958-85 with SSBN 1986-93 added..png

The laying down dates of the first two Vanguards overlap with the laying down dates of the last two Trafalgars that were ordered in the "Real World" and the last 3 Trafalgars that would have been ordered if an ordering rate of one SSN a year had been maintained.

Do you know how many slipways were at Barrow-in-Furness and whether they were long enough to accommodate the "1985 SSN" and Talent & Triumph and Vanguard & Victorious?
 
My bad: M4 started by 1973

Le développement d'une nouvelle génération de missile, le M4, est décidé en 1973. Les développements sont lancés en avril/mai 1974 par le président par intérim Alain Poher, entre la disparition du Président Pompidou et la victoire de Valéry Giscard d'Estaing à l'élection présidentielle française de 1974
.
 
Problem: Mason review was 1974 and at the time french SLBMs were M1 (1971, Redoutable) and M2 / M20. M4 came by the early 80's afaik.
Two different Masons: Roy Mason was Minister of Defence in 1974, and published a defence review.

Sir Ronald Mason was the Chief Scientific Officer at the Ministry of Defence in 1979, and co-chaired a report into the future of the British nuclear deterrent.
Would it not be reasonable to think 4 additional SSBN could have been built instead?
Not within the OTL build schedules, unless you want to posit a new class of SSBN in the mid-1970s. Coincidentally, the timing actually is about right for (wholly imaginary!) new boats to carry Poseidon in place of the RESOLUTION/Polaris/Chevaline system.
I think 3 extra Swiftsure class would be ordered (1972, 1974 & 1975) for a total of 9.
You can, theoretically, fit an extra one in for 1968, but that might be pushing it a bit.
As far as I know the fifth Resolution was never formally ordered. My SSBN spreadsheet says "The intention to build a fifth Polaris submarine was confirmed by the Minister of Defence on 26th February 1964, but this intention was rescinded by a new Minister of Defence on 15th February 1965" and I think it came from the Wikipedia entry.
I believe this is backed up by The Silent Deep, which is invaluable for such discussions.
 
Hang on.....

4 SSNs could have been built?
Would it not be reasonable to think 4 additional SSBN could have been built instead?
And that this, while additional and thus increasing total costs, doesn't directly conflict with slots in the SSN build rate?

So arguably the AH 8 SSBN plan could be achieved with the RL 19 SSN build achieved and the planned 25 later revised to 20 figures.

Even if it is at either increased spending or cuts elsewhere .
I'm saying that the industrial infrastructure to build 25 SSN, 5 first-generation SSBN and 8 second-generation SSBN from the late 1950s to the late 1990s existed.

Whether, the money to pay for the 6 extra SSN, one extra first-generation SSBN & 4 extra second-generation SSBN existed without cutting something else is another matter.

I think doing it without cutting something else would only be possible in a timeline where the UK had a better performing economy, the UK defence industry was delivering more equipment on time & at cost and there was sufficient public support for spending a larger percentage of (the larger) GNP on defence.
 
Last edited:
Although I think that the Barrow-in-Furness shipyard had the capacity to built an extra Resolution class, 6 extra SSN and 4 extra Vanguard class from the late 1950s to the late 1990s some or all of the extra SSNs and the extra Resolution could have been built at Birkenhead instead. However, I'm sticking to having the 4 extra Vanguard class built at Barrow.
 
M Jones ,Official Hist,UK Strat.Nuc.Det,R'ledge,2017,P.44 and Silent Deep,P.247 chops Royal Sovereign, 29/1/65, announced soon after (I have 15/2/65). L.Freedman, Br & Nuc Wns,Mac,1980,P.34 "RN did not take issue...some feeling that, because of the difficulty of finding skilled civilian and military manpower, there was something to be said for making do with one less boat...Admirals seemed content with their minimum SLBM force and more worried about the rest of their fleet".
 

Please donate to support the forum.

Back
Top Bottom