Depression and Recession

In the final article of his series, the author reflects on the results of his research and concludes that the fundamental cause of the seeming shrinkage in the size of the Royal Navy is not so much the reductions in the defence budget, nor even directly the cost of individual ships, but rather a shift, whether or not intentional and justified, in the strategic priority accorded to the Navy. If he is correct, then the solution must also lie in the realm of real strategic thought rather than in the ‘tweaking’ of budgets.

My recent articles have focused on both the size of the Royal Navy and the cost of successive generations of ships in relation to gross domestic product (GDP). My broad conclusions were that the Navy’s size (certainly since the 1960s) is strongly correlated to defence spending as a proportion of GDP, and that ships (at any rate frigates and destroyers) are not much more expensive in GDP terms than they were 50 years ago, despite decades of apparently high defence inflation. One of the reasons for these strong relationships, I suggested, was relative stability in what ‘a British Navy’ looks like in terms of technology and disposition. Although each generation of ship is clearly more technologically advanced than its predecessor, there is a consistency in the structure, definition and purpose of a balanced fleet. In addition, defence inflation (or at least the cost of escorts) had marched more or less in time with growth in nominal GDP.

I wondered if this kind of analysis, crude as it is, might be applicable to earlier periods in the Royal Navy’s history. In particular, I was interested in the interwar years, when economic problems constrained the amount available for defence yet the Navy still sustained a significant global presence and still managed to place significant orders for new ships in most years. One of the reasons for this interest was reading an article cowritten by our Editor and Professor Gwyn Prins in 2007.[1] The work commented on the recent paucity of orders for new destroyers and frigates and made a comparison with the 1920s and 1930s. The contrast was stark. Despite serious economic difficulties, 24 cruisers and 65 destroyers were commissioned between 1926 and 1936, compared with just three escorts between 1998 and 2008, a period of sustained economic growth. I wondered how the nation could have afforded new ships despite a dire financial situation, and whether this would shed some light on how things are today.

Defence spending as a proportion of national income

The correlation of defence spending and number of major units was not going to work. During the period 1926 to 1935, defence spending averaged about 2.7% of GDP. This is actually the same as the average for 2001 to 2010 (Fig. 1). Yet the Navy of the 1930s operated (or wanted to operate) about 60 cruisers, 144 destroyers, plus battleships, submarines, aircraft carriers, tenders, supply ships, sloops and minelayers. There were dockyards all over the world, from Bermuda to Trincomalee, and Singapore was under construction. At the same time, technology, even in the newest ships, was fundamentally different from today. Certainly there was technological progress and there was also a sense that ships were getting bigger and more expensive because of the need to accommodate better wireless equipment, Asdic, anti-aircraft gunnery, improved living conditions and, eventually, radar. However, the blistering price tags attached to postwar missile technologies, action data systems and integrated communications were still a long way into the future. Fundamentally, ships were all about guns and torpedoes, and the only things which really mattered were size, numbers and armour.

Fig. 1. Defence spending as a proportion of GDP, 1926-35 and 2001-10.

The relative importance of the Navy within defence

If the Navy of the interwar years could afford to place orders for cruisers and destroyers, despite the overall defence budget being as constrained as it is today, there had to be another explanation. One of the reasons I believed the correlation between defence as a percentage of GDP and size of the Navy has held up for the past 40-50 years is consistency in defence policy and in the relative priorities afforded to the individual Services. If I could show that the Navy was somehow much more important between the wars, much better resourced (relatively speaking), then perhaps it would go some way to explain the ability to build and operate a much larger fleet, despite tough economic circumstances.

The first challenge was to try to produce some sort of estimate of how much the Navy actually costs to run today so that I could assess its relative importance within the defence budget. I reckon the modern Royal Navy cost about £8.5 bn in 2011, or about 0.6% of GDP. (For anyone having difficulty sleeping, I explain my methodology in a postscript to this article!)

With that done, it is back to the interwar years. First of all, the effect of inflation since the Second World War is thrown into sharp relief. In 1935, the Navy Estimate (the resources allocated to the Admiralty for the year ahead) was £64.9 m, barely 0.8% of the amount that I estimated the Navy costs today or just 10% of the estimated unit cost of one Type 45! (Supplementary Navy Estimates may have added to the amount of approved spending in-year, but still not enough to make several billion pounds’ worth of difference.) Over the ten years 1926-35, Navy Estimates ranged from a low of £50.2 million in 1932 to £64.9 million in 1935 as rearmament began to ramp up (Fig. 2).

Fig. 2: Navy Estimates 1926-35, £ million.

Where this starts to get interesting (or otherwise!) is when the Navy Estimates are set alongside the total for all three Services. Over this period, the Navy accounted for about 50% of defence spending, with the Army and RAF sharing the rest. The relative shares are in Fig. 3.

Fig. 3: Share of defence spending by Service.

This means that the RN was by far the most important service, even during the period of depression-led austerity, but with the RAF nevertheless steadily increasing in importance. Superimposing this onto the present day, if the RN were to have an equivalent priority in the defence budget, its share of the 2011 budget would have been close to £18.4 bn (more than twice its actual share). In terms of share of GDP over the period 1926-35, the Navy Estimates were at their lowest at 1.1% of GDP in 1930 and reached a high of 1.4% in 1935, and averaged 1.24%, again twice as high as today (Fig. 4).

Fig. 4: Navy estimates as a proportion of GDP.

Comparing the cost of warships

That the RN was a much more important part of Britain’s defence in the 1930s still feels like only part of the story. Even pro rata, rates of commissioning new ships were far greater than more recent experience, so perhaps ships in the 1930s really were simple and relatively cheap and the RN could make its money go a long way. It was hard to find information to prove this hypothesis, but I did manage to find a few numbers and, as in a previous article, I was able to calculate the cost in terms of % GDP. As before, it is worked out using GDP in the year the ship was commissioned, accepting the inevitable errors because money is actually spent throughout build.

The result was that the cost of a modern escort in terms of GDP is at least comparable to or cheaper than the cruisers of the 1930s. A big cruiser like the Cumberland (10,000t displacement) was just a little cheaper in GDP terms than HMS Daring. The smaller cruisers (Arethusa 5,200t and Leander 7,200t) are comparable to the later Type 22s. Only modern frigates stand apart, with Type 23s (and Type 26) actually looking pretty cheap (Fig. 5).

Perhaps this comparability is not surprising. After all, cruisers were the workhorses of the pre-Second World War fleet. They were valued for their ability to operate and deploy independently, useful as force elements within a grand fleet or on their own on a distant Empire station, perhaps with a squadron of destroyers to support them. This is not a million miles away from how frigates and destroyers are used today. Even before we started billing Type 45s as cruiser-sized destroyers, our modern frigates and destroyers were arguably cruisers in more modest hulls. Indeed, the smallest interwar cruisers weighing in at 4-5,000 tons were not much larger than typical post-Type 12 frigates. It seems, then, all of the immense economic progress since the 1930s has ‘just’ bought us more and more technology. The enduring fundamental requirement to have a capable independently deployable warship costs the same (or less) in economic output today as it did 80-90 years ago.

Fig. 5: Cost of selected ships as a proportion of contemporary GDP.

More fundamentally, it looks as though the Navy of the interwar period did not necessarily benefit from comparatively cheap ships. It took as much or more of the nation’s economic output to get a cruiser as it does to get an escort today. Yet still our recent history shows that we have been unable to afford anything like the number of escorts that were afforded in the 1930s, even after accounting for the relative decline in the Navy’s share of defence spending.

Comparing the cost of submarines

The final piece to the puzzle is the cost of submarines. In submarines, far more than surface ships, there has been a more complete (and expensive) transformation in terms of technology. While modern surface ships bear comparison to their 1930s ancestors in terms of cost, the escalation in the cost of a single submarine has been incredible. At least part of the modern Navy’s inability to afford as many surface ships as it might want is the requirement to spend far more of a constrained procurement budget on submarines.

Fig. 6: Cost of selected submarines as a proportion of contemporary GDP.

HMS Oberon was already 50% more expensive than the Trident-class (0.009% of GDP versus 0.006%), but the real shift came afterwards with the advent of nuclear propulsion. In GDP terms, Valiant cost the equivalent of seven Oberons or eleven Trident-class submarines; Astute cost the same as more than nine Oberons or 14 Trident-class. However, the jump in costs was not entirely due to nuclear propulsion. The last of Britain’s conventional submarines, the Upholder-class, were themselves a huge jump in cost over their predecessors. The total cost of the four Upholders was about £900 m, implying a unit cost of £225 m, equivalent to 0.04% of GDP at the time, or more than four Oberons.

Of course, the elephant in the room is Trident. I remember conversations in the late 1980s, almost heretical at the time, speculating about the potential conventional strength of the Royal Navy ‘if only’ Trident had been shelved. There was the very real feeling that the Royal Navy’s procurement budget had been decimated by the Trident programme. I don’t intend to debate the rights and wrongs of an independent deterrent, but for the purposes of this article, it is enough to say that £20 bn would buy an awful lot of Type 26s (between 57 and 80 for a constant unit cost of £250-350 m). The cost of postwar submarine programmes has surely been a factor in affecting what could be afforded in the surface fleet, as well as in the submarine fleet itself.

Cold War and after

The Cold War was obviously instrumental in setting up the seismic shifts in defence policy which reduced the relative importance of the Navy within defence, and which concentrated effort on excellence in anti-submarine warfare. At the same time, the desire to own an independent deterrent reflected lingering ambitions to stand tall on the world stage and, in the latter stages of the Cold War, it became a credible element of NATO deterrence in the face of overwhelming Warsaw Pact conventional strength. Compared with the interwar period, post-World War Two was an uncomfortable place for a globally deployed Navy with an expensive shopping list. The struggle to reconcile that global role with a European defence policy is well documented, but it ultimately caused the diminution of resources for the Navy and a reduction in commissioning rates compared with the interwar period.

Nevertheless, the effect was not immediate, and commissioning rates were still significant in the 1950s. Excluding the Type 15 and Type 16 conversions of wartime destroyers, the annual commissioning rate is shown below, together with a 10-year moving average:

Fig. 7: Annual commissioning of new escorts since 1950.

In 1957, 11 new escorts were commissioned, comprising two Type 12s, six Type 14s, one Type 61 and two Type 41s. Throughout the 1950s, an average of 3.3 new escorts were commissioned per year. Assuming an average life of 16-17 years, this implied a fleet of 55 escorts would be sustainable at this commissioning rate. Assume a 25-year life, and the sustainable fleet was more like 83 escorts. Additionally, 33 Type 15 and Type 16 conversions were commissioned, but this drew on the legacy of wartime construction and would not be repeatable. Ultimately, once these ships, and others dating from the war, had reached the end of their useful lives, the Navy would find it impossible to secure the resources to replace them one-for-one.

1963 was another exceptional year, with four Leanders, three Type 81s and three County-class destroyers being commissioned. Overall, in the 1960s, the commissioning rate was 4.4 per year, suggesting a much larger fleet could have been sustained (110 escorts for a 25-year life) if the commissioning rate itself were sustained.

However, the Defence Review of 1966 was a watershed moment in British defence policy, not only cancelling replacement aircraft carriers, but marking the end of an enduring global role. Britain’s defence would henceforward be focused on NATO, requiring fewer ships for out of area operations. Despite being a huge shock for the Navy, it appeared to mark the start of a period of reasonable stability. In the 1970s, the commissioning rate was 2.3 per year. In the 1980s, it was about two per year. With a 25-year lifespan, this appeared to be enough to sustain the ‘about 50’ escorts which the Navy settled into. If modern escorts are the true heirs of the interwar cruisers, this is not far off the average of 2.4 cruisers commissioned per year between 1926 and 1936.

With the Cold War ending, clamours for a peace dividend and a relentless focus on land campaigns made the situation more dire. In the 1990s, the average commissioning rate was 1.4, and between 2000 and 2009 it fell to 0.4. Given no new escorts are expected before at least 2021, the projected average annual commissioning rate for 2010 to 2019 is still only 0.5. Even if escorts are run for 35 years, this implies a sustainable fleet of just 17.

Conclusions

When we look at the past and lament the fortunes of the postwar Navy, it is easy to assume that the contemporary fleet is the poor relation, diminished by decades of cuts and utterly disconnected from the glory days of the past. Despite the unbelievable advances in technology and capability, the number of ships still feels important and it seems impossible to draw any other conclusion except one of depressing decline. Yet the modern Royal Navy still bears fair comparison to its past. No doubt the 1930s felt pretty grim to contemporary sailors and officers, even as rearmament got under way. The rate of commissioning of new ships was considered insufficient even then (even if it looks rather impressive today), so in terms of how people feel and their concerns about equipment, nothing much has changed.

What has changed significantly is that the modern Royal Navy is much less important, even within defence. The Cold War elevated the relative importance of the Army and RAF ‘permanently’. With this new paradigm established, the transformed (and relatively impoverished) Navy settled into a much more predictable relationship between defence spending as a proportion of national income and the size of the fleet. The Cold War and Britain’s maritime role within NATO also made the relative importance of the submarine in modern warfare expensively ‘permanent’. Finally, the desire to retain a credible independent nuclear deterrent locked the country into an expensive cycle of replacement programmes. The irony is that the Cold War still casts a long shadow over the Royal Navy, 25 years after it started to draw to a close.

In fact, the Cold War ‘continental strategy’ of a large Army and Air Force, still lingering on through Iraq and Afghanistan, has more than halved the resources committed to the Royal Navy compared with the last days of a far flung Empire and its complex web of trade routes. Had we continued to attach the same level of importance to the Navy, it could have been more than twice its current size, just on a pro rata basis. While that is still not numerically the same as 60 cruisers and 144 destroyers, it would be a fleet with about 40 globally deployable, technologically sophisticated escorts. None of those cruisers would have had anything like the availability of a modern warship and, at any point in time, some would have been in reserve, so perhaps it is just possible to equate a ‘fantasy fleet’ of 40 modern escorts with the 60 cruisers of the 1930s.

What has been lost is the contribution of the 144 destroyers, and the scores of sloops and other minor war vessels. Again, modern patrol vessels, mine countermeasures vessels and specialist hydrography ships fill some of the numerical gap, but perhaps this is where the change in technology is most keenly felt. Instead of screening a fleet with interwar destroyers, each with over 130 men on board, modern escorts have substituted missiles, sensors and combat systems to automate the engagement and prosecution of an airborne, surface or underwater threat. We no longer have 144 destroyers, but, in some functions at least, they have been superseded and, as in industry, there has been a huge increase in productivity as a result. One Sea Viper missile, and its associated systems, is surely a better match for contemporary air threats than a single interwar anti-aircraft cruiser or destroyer, and one Type 45 carries 48 missiles.

The adjustment after the Second World War was also the result of growing frigates and destroyers into some of the roles filled by cruisers, using technology to extend their reach and capability. The small destroyers and escorts of the interwar period (and wartime) were replaced in their droves by smaller numbers of ever more technologically advanced destroyers and frigates. Yet these new frigates and destroyers were more like cruisers in capability and reach, and, of course, in cost. Meanwhile, the cost of the deterrent and investment in submarines displaced not only numbers of surface ships, but numbers of submarines themselves. Even sustaining numbers of conventional submarines became increasingly difficult.

I started by wondering how the country could have afforded so many new ships during the depression of the 1930s and have ended up by concluding that it was, in part, the result of a very different set of priorities. Defence spending was no higher than defence spending today as a proportion of national income, so we cannot automatically blame welfare, health spending or anything else for consuming resources which should have been spent on defence. The state has simply expanded, with defence no more or less important than it was in the 1930s. Ships, or at least cruisers and escorts, were not much different in terms of cost then than they are today, so the advances in surface ship technology have not necessarily been at the expense of numbers in their own right. Arguably, submarines have cost the Navy some surface ship capacity, but those same advanced nuclear submarines keep a diminished Royal Navy at the ‘top table’ and sustain the nation’s nuclear deterrent. Perhaps, above all, the decision to become a ‘continental power’ has cost the Navy more than anything. Compared with the 1930s we have chosen to prioritise the Army and Air Force, and therefore afford, commission and operate fewer ships than ever.

Postscript: How much does the Navy cost today?

Even before the introduction of resource accounting, defence statistics gave only limited insight into defence costs attributable to the Navy and, since 2000, it has become even more difficult. The sinking of all procurement and support costs into tri-Service agencies like DE&S (and its predecessor organisations) makes the task more difficult still. While I am sure that someone, somewhere knows how much the RN costs, I do not know who that someone is, or whether they would share the information, and so I have resorted to my own rough calculations.

Until 2000, defence statistics helpfully broke out equipment and support costs for sea, land and air, and personnel costs for civilians and armed forces. Beyond 2000, it is almost impossible to find enough detailed information to determine the cost of each of the Services individually. The best anyone can do without inside information is to look at a historic trend and then make a subjective judgment on whether subsequent events would have derailed that trend or whether policy has actually been stable enough to allow an assertion that the trend still holds true today.

I began by trying to determine the cost of the Navy’s people, assuming that the untrained and trained strengths of each Service are indicative of each Service’s share of Armed Forces personnel costs. Until 2000, there was a global sum for Armed Forces personnel costs, which I attributed using this simple ratio. More recently, departmental annual reports have estimates of personnel costs, though I accept the move to resource accounting muddies the comparisons over time. At the same time, the average per capita cost is likely to be lower in the Army than in the Navy, because the Army has a higher ratio of Other Ranks to Officers, but to keep it simple I did not adjust for differences in the mix of ranks either between the Services or over time, nor did I try to adjust for the move to resource accounting. The results suggest that in 1986 RN trained and untrained personnel were about 21.1% of total Armed Forces personnel costs; in 2011, about 20.2%. In money terms, spending on RN personnel costs was about £739 m in 1986 and just shy of £2 bn in 2011.

Equipment and support costs used to be broken out into sea, land and air systems, so provided a ready indicator of RN procurement and support in terms of sea systems, though elements of land and air systems would have been applicable to the RN too. Without forensic examination of every programme, I had to accept that any estimate of the RN’s procurement and support costs would be too low, but I lacked any means to make a sensible adjustment. Over the period 1986 to 2000, sea systems spending fell from just under £2.5 bn to just under £2.3 bn. As a proportion of total equipment and support costs, sea systems ranged from a high of 33.9% in 1990 to a low of 23.4% in 2000.

Finally, there are civilian personnel costs and a big bucket of ‘other’ costs. Without any ‘intelligent’ means of attributing these costs, which would have included central overheads (for example, the MoD), I made a pro rata attribution based on the subtotals I already had for RN personnel and sea systems costs. Overall, then, I ended up with a crude estimate (probably on the low side) of the cost of running the Navy between 1986 and 2000, ranging from a peak of 29.5% of the total defence budget in 1988 (£5.2 bn) to a low of 22.4% in 2000 (£5.0 bn).

Moving to the present day, it appears nigh on impossible for the outside ‘lay observer’ to do the same exercise, so the most reasonable course of action is to take the historic budget share of the RN and project it forwards, making some ‘intelligent’ guesses about whether things are likely to have changed significantly either way. In this case, it seems fair to assume that the Navy has not done disproportionately well out of defence settlements since 2000. The emphasis on land campaigns, the cuts in the 2003 White Paper and the steady increase in the proportion of uniformed personnel who are in the Army all suggest that the Navy would have done well to sustain its share of the defence budget at about 22.5%.

Taking the near cash requirement as a reasonable equivalent to the defence budgets of pre-resource accounting days, this implies that the contemporary Royal Navy cost something like £8.5 bn in 2011. In GDP terms, that is about 0.6%. Which is where we came in!

References

[1] Blackham, J.J., and Prins, G.I., ‘The Royal Navy at the Brink’, RUSI Journal, April 2007, Vol. 152, No.2 pp 10-16.