The British Army and the Fall of France: A Recap

The winter of 1946 was like this

But the spring of 1955 was like this

I'd doff my hat to an old Simpsons reference, but I can't find it.

Try again:

In the spring of 1940, a British army in France failed to stop the Germans. The consequences were horrifying, the failure abject. 

On its face, this is a subject for military history; but the military-historical counterfactual never seems to turn on the fighting, where the crucial question (the BEF's decision to abandon the line and retreat on an evacuation point) is treated as beyond discussion. Instead, it focusses on David Low's point. The conclusion often drawn is that, had Britain spent more on machines in the 1930s (or perhaps, for much longer than that), things would have been different. It's all quite strained. If you had a car, you wouldn't have missed your flight. Martin Wiener, Correlli Barnett, Winston Churchill and your Dad all have explanations for why you don't have a car, but they're not very helpful for catching the next flight unless they come with a $30,000 cheque. 

I disagree. There is a much less strained counterfactual than the idea that Britain could have been, oh, say, six years "more advanced" than it was, had public school boys been forced to do maths instead of Latin. The fact is that the 10 infantry divisions of the BEF were scheduled to be joined by its first armoured division in the third week of May, and that an entire fourth infantry corps was to come over in June. The Battle of France was a near enough thing that these additional forces would probably have been enough to change its outcome. You wouldn't have missed the train if the power outage hadn't knocked out your alarm clock. Set your phone's alarm next time, and you're set. 

So forget fancy analogies and drive by sneers, and drill down, with laser-like focus, on one, simple fact. The BEF may have been as little as one month late in getting enough troops into the line to change the course of history. This is, mind you, ten months after a parade in London described by The Economist as featuring 28,000 marching representatives of the two million men and women who had so far volunteered for national service in the defence of the United Kingdom. It sure looks as though Britain had built up quite the war machine even before the fighting started. 

But, on the one hand, two million.  On the other, short four divisions. It seems to me that the gap between these two numbers is worth exploring.

Why did Neville Chamberlain not set his phone alarm? What were his reasons? I ask this because Neville Chamberlain, Chancellor of the Exchequer in 1923 and again from 1931 to 1937, and Prime Minister from 1937 to 1940, was more than any other individual the face of an era described by Harvard economist, Alvin Hansen, as one of "secular stagnation." The connection here may not be obvious, but Hansen himself claimed that the "secular stagnation" thesis was just another name for Keynes' "underemployment equilibrium." We broke out of that equilibrium during World War II. The best explanation for that just is WWII, and the question we should all be asking is whether we can have the outcomes without the consequences. The buildup to World War II gives us perhaps the best taste of what WWII would look like without the fighting. (There is also the interesting question of whether and when the impetus of the WWII faded away.)

So the questions are:

-How big was the actual 1937--39 build up. How real is the "two million"?

-What data do we have about the immediate consequences of this buildup?  

-How did this buildup fall short of what was needed?

-How do any failures identified reflect back on the persistence of "secular stagnation"?

Four questions, two economic, two military, all worth exploring. (I guess I've telegraphed my answer to the question of whether the one-time impetus of World War II "faded away." It's here, if you're into ghost towns.)

 Here's a look at the "rest strength" of the British Army. I've picked 1924--5 because, well, it seemed like a good, restful moment between all the strenuous episodes of strength changes between the wars. We've finished winding down from WWI, there's no crisis in Iraq, the Great Depression hasn't hit, and so we're not recovering from it, either.

 Authorised Strength of the Regular British Army in 1924—5 (Forces in India “+”)

Corps	Officers	Men
Cavalry	407+162	9100+3600
Artillery	1168+312	24,800+7000
Engineers	700+425	7051+690
Signals	240+156	5000+2200
Infantry	3000+1260	77,500+41,000
Royal Army Service Corps	430+150	5900+983
Tanks	213+81	4157+983
Medical	586+332	4252+774
Ordnance	250+10	3500+207
Military Police	0	740
Total		231,700

 Also, it's hard to be believe a Briton of the day could meet an Indian's gaze in the eye.)  Indian taxpayers were even on the hook for the pensions of their former occupiers!

But that's not all!

Authorised Strength of the Territorial Army in 1924—5

Corps	Officers	Men
Yeomanry (Cavalry)	362	5100
Artillery	1786	36,000
Engineers	426	12,000
Signals	330	6500
Air Defence Troops	374	8600
Infantry	3360	112,000
Tanks	112	1600
Royal Army Service Corps	112	1600
Ordnance	43	470
Engineer and Railway Staff Corps	60	0

I was going to total up these columns for you, but this is the “authorised strength.” No-one is even pretending that the Territorials will ever be close to authorised strength. Any even that brings it close to its authorised strength will cause it to be greatly exceeded. 

But that's not all! The whole point of having a "short service" army was to have an efficient reserve so that you could send an army over to the Continent tomorrow to intervene in the Franco-Prussian War, in case it happens again. Breaking with my attempt to achieve consistency, since 1924--5 was not typical in this respect, the Regular Army Reserve was, in 1939, 145,000; while the Supplementary Reserve of 1914 was 60,000. The Regular Reserve consists of men furloughed into civilian life at the end of their service, while the Supplementary Reserve consisted of men taken directly into the reserve with no regular service at all.  

It will not go unnoticed that quantifying the British Army is tricky as all heck. A bad actor could easily take two copies of the Estimates, paste "Ruritanian" over "British" in one, and then used motivated reasoning to prove that the Ruritanian Army outnumbered the British by almost 5 to 1. And by "could," I mean, so, yes, there's the historiography for you.

The Women's Land Army has a load of fertiliser for you.

Now let's get back to the Supplementary Reserve, which in 1914 existed to fill out the Regulars with eager young riflemen. Since the country was swimming in reservists in the postwar thanks to various oddities of wartime recruiting, in the postwar era, the Supplementary Reserve was about as useful as a women, men, fish, bicycles, etc. So it's on the list as (60,000); men not existing, money not spent, an authorisation waiting to be authorised. Until 1939, when the War Office finally plugged it in. Here is the S.R. as it was embodied in 1939. 

Provisional Estimates, Supplementary Reserve

Corps	Officers	Men
Cavalry	32
Armour	310	343
Artillery	362	5149
Engineers	372	6401
Royal Engineers, Transport	281	7129
Signals	319	4789
Foot Guards	104
Infantry	599	17,000
Military Police		800
Service Corps	465	11,200
Ordnance	72	8852
Paymasters		650

There’s a great deal here that is hard to parse, so I should probably explain that the “Royal Engineers, Transport,” while actually existing in the Regular Army, gets this huge bump because it has been decided to let men (and gentlemen-by-order-of-the-King) working in automotive factories to join up directly in the RET without going to the trouble of attending a summer camp and learning to square bash. They get paid, they get to wear snazzy new uniforms, put letters after their names –all those keen social perks, and no disadvantages at all. It’s like a sleazy car ad, only instead of punishingly high interest rates down the line, there’s a war. Of course, the whole point of spending all this money and effort is to persuade Berlin to back down and not start a war, so it really does seem like a good deal.

Also, since the “fitters and riggers” who maintain the army’s machines are, in service, split up between the service corps and Ordnance, the large numbers of both in the Supplementary Reserve reflects the fact that the War Office has gone nuts trying to recruit people to overhaul truck and tank engines. Quite a change from the Supplementary Reserve of 1914! 

Now for some more HR stuff before moving on to money.

Vote A Effective Strength, British Army, 1938 and 1939

Troops	1938			1939
	Officers	Men	Total	Officers	Men	Total
Cavalry	369	7708	8027	90	1754	1854
Armoured				622	11,919	12,544
Artillery	1329	24,857	26,185	1366	29,345	30,711
Engineers	649	9494	10,143	292	6,430	6,722
Signals	295	5991	6286	292	6430	6722
Infantry	3082	76,640	79,722	2723	78,649	81,372
Police		512	512		581	581
Royal Tank Corps	302	4475	4,771
Chaplains	135		135	147		147
Service Corps	454	7,058	7519	705	8385	9090
Medical	495	3973	4468	590	4766	5356
Ordnance	530	4123	4653	676	5167	5843
Pay Corps	184	780	964	196	962	1158
Veterinary	21	95	126	21	95	116
Educational	50	272	56	21	349	405
Dental	129	181	314	154	210	364
Total, Regimental Est.	8024	146,157	154,181	8108	154,599	162,707
Permanent Staff, Reserves
Supplementary Reserve	6	33	39	17	112	129
Territorial Army	547	1840	2387	672	2429	3101
Officers Training Corps	11	59	70	11	59	70
Colonial Militias, etc.	4	21	25	4	21	25
Total, Permanent Staffs	568	1953	2521	704	2621	3325
Staff
War Office Staff	313	157	470	371	202	573
Staff of Commands, etc.	798	215	1013	644	62	706
Total, Staff	798	215	1013	1015	264	1274

(Before you go all bug-eyed, the Vote A total strength for 1938 is 161,000, and for 1939, if I haven’t mentioned it already, 185,000. Note that I've left off an entire page of convicts in military prisons, students and faculty in various schools (almost 5500), and the Army's 75 groundskeepers.)

So there you go. The British Army had to hire 25 dentists in fiscal 1938 to keep up with growth, and a lot of infantry officers evidently wrangle transfers to the new Royal Armoured Corps. Where else can you go for this kind of quality information?

*

 And now what we've all been waiting for --Money!

Selected Army Finance Votes from a synoptic table in the 1938—9 Army Estimates.

Fiscal Year (£)	1936	1937	1938 (est.)	1939 (est.)
Vote 1, Pay of the Army	8,887,287	9,121,991	10,819,000	11,943,000
Vote 2, Pay of the Territorials, Reserves, etc	5,978,710	8.043,,919	9,775,000	14,022,000
Vote 3, Medical Services	1,106,003	1,075,272	1,105,000	1,458,000
Vote 4, Educational Establishments	943,628	1,033,277	1,327,000	1,542,000
Vote 7, Clothing	1,294,226	1,628,872	1,995,000	5,729,000
Vote 9, Warlike Stores	10,713,253	16,435,713	27,242,000+41,242,000	8,661,000+56,661,000
Net Cash Expenditure	54,195,089	62,735,520 (72,675,520)	85,357,000 (106,500,000)	81,923,000 (148,155,000)

There are a total of fifteen votes in the mid-30s Army Estimates, but I’ve made the authorial decision that Works, Lands and Buildings Vote does make the cut, even by blog-blather standards.

If you are wondering about the numbers in brackets, the Government is trying, for some reason, to keep expenditures financed by taxes separate from ones financed by the National Defence Loan introduced into the Commons on 20 April 1937, and since subject to the Great Forgetting, lest attention to it, and its maidservant-in-arms, the General Defence Contribution, upset modern orthodoxies. (1,2,3). 

Two centuries ago, the largest port on the West Coast between San Francisco and Alaska; today, even the old name is forgotten.

As the numbers show, 1937--39 was a period of huge run-up in British defence expenditures on the land forces. The only way to make it seem less remarkable is to compare it with the Air Force, which saw a run-up from £41 million to £221,000,000; with a much larger share going to warlike stores and to research and development (unfortunately not broken out in the Army Estimates), as the Air Force's prewar Vote A topped out at 100,000, with a proportionately even smaller reserve component. 

Also of some relevance to the experience in France, a large part of this spending on warlike stores is for antiaircraft guns. 

It was  nontrivial expense: here is the budget that was expected to be tabled on 3 April 1937.

This is against an ordinary revenue of 797,289,000 down a million pounds over the year, mainly due to declining income tax receipts, yet another hint of the recession on the horizon. This is, by the way, against a national income of £3.89 billions. So we have spending at very nearly modern levels and a (very small) deficit. However, I take Paul Shay(*) to be implying that the final bills for the supplementary estimates issued for increased defence spending in the fiscal 1936/7 year as adding to this. The Government was already borrowing to fund defence, and, in fact, had been doing so since 1935. It was just time to put it on a regular basis. 

There's something more than a little strange about getting your news coverage of these events from The Economist. Ordinarily, a weekly's coverage isn't that different from a daily, but, in this case, one opens the paper to the 24 April 1938 number, six days after the Chancellor-soon-to-be-Prime-Minister rose in the House of Commons to announce the National Defence Loan and a new tax on profits, the National Defence Contribution, and you encounter a dispassionate now-that-the-shouting-has-died-down discussion of the Contribution, and a parenthetical reminder that, after all, the Government has decided to borrow money to fund defence expenditures. 

It's all old news, man. Why are you dredging this old stuff up when we could be talking about Gracie Allen? Needless to say, the shouting had not died down. Chamberlain's Government would retreat from the original NDC in short order, albeit in a sideways retreat, since the old WWI Excess Profits Tax was revised to make up the difference. 

The question is, why? Answers abound: a new tax was needed to make up revenue; Labour was to be placated by the sight of the bosses paying out of profits, which would reduce militancy; a tax on profits would curb inflationary pressures resulting from high profits at the height of the 1937 boom; high profits in the aircraft industry were already a political issue; the Bank of England was behind it.

In a way, all of this reflects the aging state of the literature. Writing in 1975, Shay imagines the British economy of 1937 as humming along in a boom, with inflationary pressures from labour demands for higher wages only barely kept in check. The Economist, closer to the event and more aware of the spirit of the age, points out that the fall in the cost of living in the recent depression more than made up for the loss of income in the same period, so that the "salary and wage-earning classes" were actually better off, even if they didn't realise it. (No word on whether they all had colour tvs and cellphones, so that it didn't matter that they hadn't had a raise in twenty years.)

Today, we can look at this

and see the spectre of secular stagnation stalking the land. A hat tip to James McGraw, Jr., who acknowledges the possibility in the fall of 1946 as he complains that the long-term, "just" division of earnings between wages and capital gains that had held in America from the 1890s to 1941 might have beem generating more savings than could profitably be employed. Profits are too high. 

If that is the case, then money is being diverted into speculative investments, one of which, over the previous three years, had been aviation. It is probably no coincidence that the British aviation scene is littered with dubious startups and revivals of old margues. The Economist grimly warns that investment in industry is "reaching the limits of profitability." Which, given that the profit margin in aviation is hitting 20%, is not hard to believe! 

So it is a bit surprising that the National Defence Loan bonds were 7 year terms at 3% rate of return on the National Defence Loan bond. As a result, the issue was undersubscribed, and the Bank of England had to step in and buy the bonds, as previously and confidentially arranged. The outcome of that, of course, would have been an increase in the money supply --a quantitative easing of the tight money situation which had developed in Britain, to coin a phrase. 

It's this train of consequences that leads Shay to speculate that the National Defence Contribution was suggested by the governor of the Bank of England, worried that pumping money into the economy must lead to inflation, followed by a collapse in investor confidence in the soundness of the pound, a run-up of interest charges on the national debt, national insolvency, and the victory of Fascism and/or Socialism. Some even suggested, a little wildly, that inflation might increase the trade deficit; leading, in no time, to a situation in which Britain could no longer afford to import grain "and other raw materials," leading to unemployment and bread riots, and then on to Fascism and/or Socialism. Truly, all outcomes were bad

Or, rather almost all, because of the  "new mood of labour militancy" inspired, at the beginning of 1937, by rising profits, falling (albeit still high) unemployment, and declining real wages. Unlike The Economist, your average engineering operative was unimpressed that the family could afford more bread than it used to be able to do, and was focussed more on the pay packet. Strikes, on the rise as the budget estimates were being prepared, would obviously lead to wage increases, which would lead to inflation, which would lead to Fascism and/or Socialism.") Also incidentally-I'm-just-mentioning-it's-not-really-important-forget-I-said-anything, it would hurt people living on fixed incomes. 

The idea that Neville Chamberlain, a private man who reveals himself to us mainly through his letters to his two spinster sisters, gave only occasional thought to the fate of the poor rentier isn't just a  notion for a cynical historian to reject. As spring moved into summer, the new government had two fights on its hand. On the right, it turned out that campaigners for "sound budgets" were much less consistent than had been supposed when the battle turned from one of cutting transfers to the poor to cut the deficit; to raising taxes (especially on profits) to pay for rermanent. Debt for maintaining the shiftless idle was one thing. Debt for paying for guns was quite another! Simultaenouslyl, on the left, the Government had to go to the engineering sector unions and ask for a series of concessions on "the distressed areas" and on dilution.

I suspect this is a comment on a tax initiative that hit the poor. Low has similar compositions that explicitly make the point that armaments spending is bringing prosperity, etc; but I thought that this one captures the hovels of  the working class North a bit better than those ones. 

  I have talked about the unsuccessful negotiations with the labour unions in the summer of 1937 before, so I guess the bullet point summary is that they happened and they were unsuccessful. I notice that there I was on about the "distressed areas," the idea being that the solution to high unemployment in peripheral areas was to reduce wages there. That idea not getting very far, the 1937 negotiations shifted tone to the need to increase production. There were underutilised labour assets in the distressed areas, but they were underutilised because they weren't very skilled, and so--- Different problem, different plan, same solution, as it turned out. The engineering unions pointed out that the Government had it in hand ways of making it easier for labour to move around the country, and there it, for some reason, ended. (Spoiler: "ways" equals "spending money.") 

Dilution was a different plan, although you may take it from me at the head that I could just reuse my "spoiler" gag. The high profits in the aviation industry can be explained in three ways:

i) The companies recovered the research and development costs on the initial run of Hawker Hind variants they built for no now-discernable reason, were making windfall profits, and could now deliver aircraft at lower cost to the taxpayer if the contract were extended even further. This is what industry lobbyists tend to believe with special fervour  whenever parliament suggests that a production run be curtailed to save money. Cf. B-2, F-22, etc.

ii) The work force was getting better at making planes as they went along by "learning by doing," or or "innovating by doing" or what have you. Labour unions really like this idea. 

iii) Long production runs lead to increased productivity either by "division of labour," or by advancing technology. The former intepretation is preferred by mad Scotchmen of a certain age, while the latter was vigorously promoted by an AI emulation of Ray Kurzweil, run in a timeless state of pure virtuality by a Moores-Law-Gone-to-Infinity computer at the end of the universe. Given that a large part of the improved efficiency was from using production jigs, you have to have an extremely relaxed intepretation of "advanced technology" to agree with Kurzweil.exe.

I know that probably shouldn't I keep trying to make this argument through snide comments instead of  through sustained analysis, but it really seems to me that the argument about how "automation is taking our jobs" is impoverished by a sustained failure to engage with the way in which productivity improvements actually happen.

 So we're probably going to give a tip of the hat to the mad Scotchman.

Anyway, if (iii) is right, then aircraft production could be increased by "diluting" skilled labour with unskilled. You will also get more aircraft eventually by "dilution" if (ii) is right, but through learning-by-doing. That doesn't, however, mean that you have to pay your diluted labour more. 

The engineering unions could afford to remain agnostic on these issues because by the spring of 1937, unemployment and "short-timing," as they called it in those days, was on the rise. There was no "labour crisis," and if what the Government really wanted was more planes for less, it had already given up on the most obvious tool for achieving that when it retreated on the NDC, as it did. 

It was at this point, it seems, that the Government discovered that the Treasury was dead serious about the idea that the Government couldn't borrrow more money than the country was saving without something something inflation Fascism and/or Socialism. 

The Secretary of War, Duff Cooper, was called into the office and told that the army would have to "ration" rearmament, after all. To be fair, the Air Force was told the same thing, which is why it shifted over to emphasising fighters at this time; but since everyone agrees that this was a Good Thing, with even some Service advice to the effect that the country already had enough Hinds.

Actually the precursor Hart, but never mind. 

Cooper's presentation, which focussed on the fact that the War Office's plan to rearm the Territorial Army would have to be scrapped, endangering the Expeditionary Force and probably undermining industry's ability to deliver modern weapons entirely over the medium-term future, was overridden. Instead of resigning, Cooper agreed to be translated to the Admiralty, in what was not exactly his finest hour, and the Inskip Report, in the midst of scrapping bombers for fighters, took time out to recommend that the country moderate its ambitions to meet the limits of the country's precious borrowing capacity, and scrap the "continental commitment."

This was accordingly done. If you want to see the Army's opinion of that for public consumption, you can read the editorial in the January 1938 number of Army Quarterly. If you want some kind of sense of how the Army took it internally,  you can check out the long list of (forced) resignations from the Army Council at the same time. With his typpical tact and sensitivity, Chamberlain appointed a brash young businessman with strong Liberal leanings, of the Jewish persuasion, to be the new War Secretary. Ironically, once he had a grip on his ministry, Leslie Hore-Belisha was to be a strong and in some ways effective advocate for the Army, but, at the time, Colonel Blimp's reaction can be guessed. By Gad, at least incoming officers like Gort and Ironside were to cover themselves in glory! 

The one place you will not hear an account of all of this, unfortunately, is in the traditional historiography, which likes to present the decision for a "continental commitment" in March of 1939 as the sudden reversal of a policy of twenty years, as opposed to one of eighteen months. I am sure that it has nothing to do with the fact that the great Chichele Professor of Military History at Oxford University was advising the Secretary of War at the time, and, in fact, wrote several papers on how the Army should abandon preparations for war in Europe in favour of anti-aircraft and "imperial defence." I am sure it was because he was busy preparing a sequel to his history of the Royal Tank Regiment that explained how the former head of the Tank Corps and then Master General of the Ordnance, Lieutenant General Sir Hugh Elles, came to leave in 1938. (The career path of MGOs post-appointment here.) 

Shay, who has read the justifications for abandoning the continental commitment, notes that one argument presented was that the German guarantee of Belgian neutrality removed the need for it. He adds an old-fashioned (!) at this point, and I would go further. Someone is taking the piss. The Government is not on the level. It is interested in saving money somewhere, and it's not that the Army isn't being given more money. It's just not being given as much more money as the Air Force and Navy. So: why? Why must the Army suffer disproportionately so that public borrowing does not crowd out private, leading to a fall in investment in the export industries, leading to a widening trade deficit, leading to etc, etc.? Why the Army?

I have suggested before (and this why this is a "recap"), that the most obvious explanation is that the negotiations with the engineering unions and the funding decision are linked. I just think that they are linked at more than one level. In this case, the "human resources" stuff at the head of this blog post are linked to the "finances" stuff below. For the Army to be ready to make a "continental commitment" means making up a serious deficiency in manpower in the technical arms. 

There is only one way for the Army to do that in the short term. It will have to recruit from industry --Meaning pressure on wages.  

So let's review:

1937: Increased government spending on armaments, pressure on wages: boom.

1938: "Rationed" defence expenditure: bust.

1939: Defence spending panic: boom.

I am not going to make some crazy claim about causality here, because the larger causes of the 1937 recession are well known, and the obvious question about the 1939 boom is how bad will be the bust that follows. 

And how long it will draw out, and just how serious will be the consequences. 

What I am going to say is: Do you want to see an economy try to pull itself out of secular stagnation with "fiscal policy," rather than war? Here's a laboratory for you. It is a short-lived laboratory; but it is one that we have.

*A big round of applause to Princeton University Press for bringing this 40-year-old book out of its back catalogue and making photocopies available to the wider reading public for a mere $45 American.