Notes — Freedom’s Forge by Arthur Herman

My second deep dive pandemic read is Freedom’s Forge by Arthur Herman, covering the history of the United States’ industrial mobilization for World War 2. There is a good deal of resemblance between the mobilization to beat Covid (and coming soon: climate) and WW2 mobilization, so it’s a good history to have in your back pocket for the next decade.

As with my previous deep dive, on Barbara Tuchman’s A Distant Mirror, this is just a cleaned-up version of my livetweeting as I read the book. As with that book, this one too benefits from a lot of Wikipedia side-trails, and I’ve linked a bunch of them.

Next twitter live read for pandemic, Freedom’s Forge, on WW2 industry/economic mobilization recommended by James Giammona

Let’s see how this one goes.

Alright, this is gonna be a Great Man story. Two in fact: William Knudsen who left presidency of GM to head up office of production management, OPM under Roosevelt (funny how the OPM hacked by Chinese was “Personnel”), and Henry Kaiser, construction cartel boss turned shipbuilder.

“Their foes weren’t German or Japanese soldiers but Washington politicians and bureaucrats, shrill journalists, military martinets, the denizens of Big Labor as well as Big Government — and sometimes the forces of blind date.”

Ok, it’s clear what sort of book this will be 😀

Apparently 20x as many people were killed or injured in the war machine compared to the actual war 🤔

Now learning about the Nye Committee which investigated WW1 production machine in the run-up to WW2. So that’s where the phrase “merchants of death” comes from.

Damn the neutrality/non-engagement decades were really anti-military:

“From the 4th biggest military force in the world in 1918, the US Army shrank to number 18, just ahead of tiny Holland”

Roosevelt started rearming in 1936, years before officially dipping neutrality and joins the war. If you want peace, prepare for war?

Damn, early on Patton ordered nuts and bolts for tanks at his own expense from the Sears catalog, they were so short 😱

1940: Roosevelt asks for 700 million in army appropriations, up from… 24 million. This, coming just 7 years after dumping the gold standard must have been a huge shock. How much is that in 2020 dollars?

About 15.6B inflation adjusted. Drop in bucket compared to CARES act. But then, shit was low-tech and cheap in absolute terms then.

“…on policy matters FDR was a procrastinator. He preferred to put off decisions — or at least to keep news about them from going public.”

Hmm. The anti-Trump, who declares mission accomplished before it starts.

Bernard Baruch, organizer of WW1 War Industries Board (and target of Nye committee) declines to do an encore for WW2. Don’t blame him.
Apparently WIB didn’t manage to pull its act together in time for WW1 despite (because?) of strong economy. Baruch nominates Knudsen, who enters stage left.

Hitlers response to FDR plan of 50k planes/year: “What is America but beauty queens, millionaires, stupid records, and Hollywood”

Oh shit Hitler had 2016-America viewing palantir.

Knudsen starts off as an immigrant doing riveting work on the tough New York waterfronts and learning boxing. This is already cartoon grade. Marlon Brando plus Rocky.

Montage of early industry career and now Knudsen is systematizing Ford’s assembly lines by 1916. The real Taylorism is Knudsenism it seems.

Knudsen leaving Ford after high-level head-butting with Henry Ford, and going to Sloan at GM reads a bit like Wolverine leaving Magneto for Prof. X.

GM is more interesting as an early business case study than Ford, even though Ford steals the technology honors. The invention of the federated conglomerate was a bigger deal than the assembly line.

“Speed produces nothing in manufacturing,” Knudsen likes to say — which was one reason he eschewed the complicated time-and-motion studies of production gurus like William Taylor. “Accuracy is the only straight line to great production.”

I LIKE THIS GUY

He’s basically discovered “slow is smooth, smooth is fast” thinking in 1922. True flow thinking including lean vs fat tradeoff. Unlike charlatan F. W. Taylor. There’s an element of Elon Musk first-principles thinking to Knudsenism.

Knudsen set a “one to one” goal of matching Model T sales and got from 18:1 to 2:1 in a few years. He also invented model year system of annual updates. Damn.

Aside: this was clearly the OG example of the Mac vs PC pattern. Dunno why I didn’t see it before. Probably because this is the first time I’ve read about the Chevrolet story.

Kaiser was a dropout hustler who led with sales looks like. Makes his start in photography in upstate NY, then heads west to grow a hardware business, then lands in highway construction. So a hacker-hustler pair saved the US in WW2.

Kaiser builds thriving road construction business in BC but WW1 makes the name Kaiser acliability do he he adds back to US and grows big there.

Kaiser building thousands of miles of roads in the west. Early adopter of innovations. Steve Ballmer type seems like. Hard driving and hard working but subtler than Ballmer… not a bull type. Seems like he elevated roadwork from mostly manual labor to mostly machine.

Now we have Kaiser orchestrating the Six Companies to build Hoover Dam. He spent most of the time in DC managing the political bosses.

Huh, Kaiser apparently pioneered the internal competition model with Grand Coulee dam, which he later used in WW2 work. I guess that’s where the X/Y flyoff model originated perhaps?

Okay. Chapter on 1939-40 world fairs as transition, and we’re finally into WW2. Knudsen meets FDR.

Interior secretary Harold Ickes skeptical of Knudsen; “I have heard that Knudsen even makes his own notes in handwriting” (presumably as opposed to a secretary transcribing shorthand into typewritten notes) Moron. If you aren’t taking your own notes you’re not thinking.

Knudsen pulls together a sort of fellowship of the ring from across industry. They have to figure out what the military needs and tell them, because the military if 1940 doesn’t know.

Finally into the war production story proper. We’re talking tanks, airplanes, engines. Building for US, British, and French needs all at once, 1939-41. Just a vast amount of action getting underway.

Lessons and anecdotes at every level from rivets vs welding through contract laws to amortization regulations, to foreign policy. The blog post, Reality has a surprising amount of detail by John Salvatier, is a good side read here. This is that times a million.

Nice anecdote: a banker who has to transfer top secret Merlin engine plans meets the battleship bringing them from the UK with an empty suitcase. The captain laughs: the plans occupy a whole railroad-car sized crate of paper (would be a container today… or a largish USB stick)

I’m honestly kinda excited for the post-Covid rebuilding.

Spark plug and steering gear auto subsidiary companies commandeered into making machine guns. We’re still on Knudsen. Kaiser is trying to get in on the action but Knudsen doesn’t trust him. Yet.

The auto industry had 1050 factories and $3B in facilities and largest pool of engineering talent. Like Silicon Valley today. Except luckily SV capability is a better match for Covid response.

“It was also an industry of associations”

850 companies, 1/20 of workforce,… saw this while at grad school at U. Michigan. Auto industry is ridiculously full of industry orgs associations densely knitting it together. SV is much looser.

“What Knudsen saw in the defense buildup was more than just rearmament… he saw a way to revitalize American business and industry”

“The arsenal of democracy” has a nice ring to it. Need something similar to capture strengths for pandemic. China may have started it and democracies may have suffered way more casualties but I suspect long term liberal democracies will win the war against Covid better.

Vignette about unions opposing Knudsen and proposing alt Reuther plan by UAW chief Walter Reuther. Book is clearly anti-union but Reuther’s plan does look like stupid wrong-problem (fighters overran bombers) vaporware objectively speaking, from engineering and other lenses.

Why FDR resisted urge to appoint war czar and take over all economy: “A victory small enough to be organized is too small to be decisive” — Eliot Janeway (who? an economist who influenced FDR it seems).

Alright Kaiser back in the story to build merchant freighters for British because America’s biggest shipyards were at capacity for navy orders.

Knudsen gets full authoritah over Office of Production Management (OPM) but FDR sez he has to share authoritah with a labor leader. Guy named Hillman.

Now we have Kaiser going brrr. Gets shipyard location swamp drained in 3 weeks instead of 6 months that people expect. Richmond shipyards. So that’s the story behind that huh.

Apparently Kaiser revolutionized shipbuilding by treating it like a construction industry rather than a steel industry.

Ship building go brr.

Knudsen’s people are now making a huge Soviet style central production planning and demand forecasting ledger.

Guy named Stacy May did a massive project to compile the ledger but is now forgotten. Not even a Wikipedia page smh

The production priority lists of “strategic”, “critical” and “essential” had to be repeatedly redone to balance regular and military economy needs.
Everybody cool in 1940s USA had quoted nicknames.

“Babe” Meigs, “Powder” Johnson.

“That was the magic number: 18 mo. That’s how long Knudsen estimated it would take for American business and industry to make the arsenal of democracy a reality. One year to build new plants and retool the old ones, 6 mo for conversion. Everywhere he looked that number held true”

Fraught period of Knudsen jockeying for position with labor leaders and losing, and FDR vacillating on commitment to war effort under pressure from isolationists. But crucial period of ramping up machine tools is underway so big risk mitigated.

Interesting that 2nd gen companies like Cincinnati Milling Machine stole a march over first-gen more artisan NE tool makers in this period, including sneaking large naval gun drill out of Nazi Germany.

Fascinating glimpse of balancing civilian and military production. No point diverting raw materials to military if you then can’t build houses for civilians to work on it, or roads to transport material to factories. Economy is intertwingled. Military/civilian is fake divide.

The breakdown of various attempts to partition the economic web by fiat in human meaningful terms (civilian/military, essential/non-essential, strategic/non-strategic) is like the explainable AI problem. Like declaring pi=3 like Arizona once tried to do. Might blog this.

Picking this one back up after a while. Now into 1941. Defense spending up 12x through the year. US approaching Nazi Germany in output. 1942 overtake. Capitalist exponential network effects slower to start, but faster once past a point.

Now at Pearl Harbor… “[the army plan]… still saw Japan as a problem to be put off until at least July 1943. The timetable had suddenly, catastrophically speeds up.”

Timetables speeding up is a general theme across everything I’m reading right now. “Slowly, then suddenly”

The auto industry was like Silicon Valley of 1940-45 looks like. Knudsen weathered the 1940s techlash by simply ignoring it, refusing to defend himself, and just doing what he thought necessary. You can do that if you’re kinda indispensable.

Though this book is obviously strongly pro-capitalism/anti new-deal crowd, the state of play still filters through. Looks like Big Labor at its height was as corrupt as Big Finance at its height today. Absolute power corrupts absolutely etc.

New dealers get Knudsen fired from OPM. OPM and SPAB replaces by new agency WPB. War Production Board. Knudsen devastated but lands 3-star generalship on military side of effort. Only civilian to get that. Kicked… sideways?

Rule of three: In the first year after a production order output was bound to 3x. In second year, 6x. In third year, at supply limit: materials and labor.

Chapter on Six Companies work in construction in pacific islands, and civilian workers role in Battle of Wake Island. Kaiser role there was supplying cement in bulk carriers instead of bags, for efficiency, assuming the risk.

Now into Liberty ships story.

Summer of 1941: 4000 workers in Richmond yards. After Pearl Harbor, by end of 1942: 80,000. 20x.

WW2 manufacturing tech was highly scalable with relatively low-skill/low-training labor. Kinda like driving Uber today I guess.

Fascinating story of race between 2 Kaiser shipyards, Richmond 2 (Clay Bedford) and Portland (Edgar Kaiser) to drive production faster using prefab deck house sections and assembly line techniques. Time cut from 220 days to 10 over a couple of years. Some photos:

Bedford takes back record with Robert E Peary built in < 5 days. Something of an engineered PR stunt of course, but these record setting races did drive genuine advances in construction techniques. China is in this mode today. Like those Wuhan hospitals.

Alright let’s pick this up again. Now into the story of Knudsen successor Don Nelson, ex Sears guy who learned how 135k products in the Sears catalog were made and ran the WIB.

Nelson faced down trust-busting threats from Roosevelt AG and FDR eventually suspended trust-busting for the war. Also fought Truman on hiring of dollar-a-year crowd. Basically anti-corporatism suspended.

By end of 1942 US was producing more arms than all 3 axis powers combined

Interesting. Massive levels of direct open cooperation among dozens of aircraft makers. Kinda like Silicon Valley with open source. Sorta open source mechanical and industrial engineering under wartime pressure.
Now reading about Andrew Jackson Higgins who designed 92% of the small boats used in ww2. Landing craft, PT boats etc. Hitler called him the new Noah.

Aluminum production went from 327m lbs to 2.2b lbs between 1939 to 1943. Today the US makes about 1 million metric tons which curiously is about the same (2.2b lb) and imports about 6 million. 🤔

Henry Kaiser builds a big steel plant in California. Looks like this was when California started transforming into an industrial powerhouse. This plant inspired Ayn Rand’s Rearden Steel in Atlas Shrugged and was also where Terminator 2 climax was filmed.

Back to Detroit. After a survey of impressive production statistics, the story of Ford’s Willow Run plant for making B24s, and Knudsen’s evil twin Charles Sorensen who pushed through a dating vision to make planes like cars. 25k parts instead of 1.5k.

Excellent story. Huge problems with production design and labor. And dual purpose of both parts supply and full assembly resolved to focus on latter. Innovation of field modification allowed production to get past constant stream of design mods and hit pace of 300/mo.

B24 Liberators were less popular with crews than B17s but more capable. VLR version won’t the Atlantic war by closing the gap in coverage for U-boat defense. Funny I never looked up this story despite being in Ann Arbor for years and a fan of WW2 aircraft lore.

Generally impressive the extent to which this generation of engineering leaders had mastered enormously complex command economy production, just a couple of generations after boutique early mass production. Artisan production craft turned to manufacturing science in 50 years.

A big part was being asshole dictators. Sorensen was apparently called the Mussolini of Ford. Leaders of this era punched each other up. Still the culture in large-scale manufacturing but less so.

1943, Big Labor broken as public and even FDR lose patience with strikes disrupting war effort. This is the weakest part of the book. I’m guessing there was more than this cartoon villainy portrait to labor side of war story. Congress passes War Labor Disputes Act over FDR veto.

More production go brrr. Kinda gets lost in the record breaking that much of this output was used to bomb Germany to rubble and kill on an industrial scale. The engineering story is great but the slaughtering race is depressing to contemplate.

Parade of stories big and small, from GM’s plant in Iran to assemble trucks for Russia to small garage startup making machine tools and tank parts. The sheer number of things being made by unlikely companies in impressive. Frigidaire made machine guns for eg. Wtf.

This is mainly a mechanical, metallurgical and chemical engineering story but tons of electrical too. GE made ridiculous numbers of types of lighting, motors, etc. Also bazookas. The generality of industrial capacity in the 1940s is amazing.

Key principle: WIB could restrict consumer demand, regulate wages and prices, and stop the production of non-essentials but could not order companies to make specific things. Producers chose how to insert themselves into war effort. Based on skills, benefits.

Not a free market but not a command economy either. More like a 50-50 mix. Shape demand and restrict supply but leave matching free. Regulate the macro and boundary conditions but not the micro.

Starting in 1941 Business Week apparently ran an advice column responding to reader queries about war materiel production contracts and opportunities.

This book published in 1942 was a guide to getting into war production. Your Business Goes to War.

Half a million new businesses, including Pacific Hut which saved steel by replacing steel Quonset huts with coated plywood huts. The war transformed the economy into modern form.

On the labor side, 20 million migrated to the new industrial base. A step function in industrialization I guess. Up from what was already the highest level in the world. No wonder 1930s and 1950s seem worlds apart. The 50s seem familiar in a way the 30s do not.

Wages rose by 70%. About 7 million left farming for military or industry, driving farm automation and fertilizer use. The war was pretty good for the American economy. The US bought itself a modernization via wars wrecking Europe and Asia.

Blacks fought for a big social level up: Executive Order 8802, patchy desegregation in industry (though nothin the military), race riots in Detroit. Book skips rather lightly over this bit. Sounds like a whole other book could be written about this alone.

8802 introduced minimal anti-discrimination protections fir race, color, religion, creed, but not sex. But women made the biggest gains despite headwinds. Kinda tedious how incumbents predictably resist any new entrants to anything.

“By July 1944, 36% of all workers in prime defense contractors were female”

• Steel: 22.3%
• GM: 30.7%
• Kaiser yards start Richmond: 70%

“Rosie the riveter” archetype was really 3 women: Vera Lowe of Lockheed, used in Lockheed PR after a photo appeared in Life magazine Geraldine Huff for “we can do it” ad council poster by J. Howard Miller May Doyle, Norman Rockwell’s model for Saturday Evening Post

Shipyard diary of a woman welder, 1944 bestseller. On google books

Back to Henry Kaiser. Now he’s building baby aircraft carriers and developing a very high public profile. Then ships cracking mysteriously topple his rising reputation. Brittle steel, not his fault.

Kaiser partners with Howard Hughes to build an airplane version of the liberty ship, the gigantic Spruce Goose. Spoiler: project went nowhere.

Now into B-29 story, probably the most complex project of the war, costing more than the Manhattan project. Bridge between early and modern planes, except for jet engines it had most familiar features of today.

Lots of production innovations: multi lining being the big one. Unlike one long assembly line like at the B-24 Willow Run plant. Interesting that this is still how it’s done if you visit the Boeing plant in Everett. 40k parts instead of 25k for B24. Built at 4 plants.

Serious crash kills test pilot and sets back B-29. Knudsen called in to rescue the program. Wings prone to catching fire.

Such a lovely plane. Gotta remember though — only plane to actually drop nuclear bombs in war.

I will admit I got sucked into watching a bunch of bomber videos on YouTube today. Well, back to the book.

Knudsen working through endless B-29 production troubles. Manhattan project starting up. Apparently the B-24 and B-17 couldn’t handle the nuke. So if the B-29 hadn’t made it in time, it would have been British Lancasters that did the job.

“This nation seems to be able to do more by accident than any other country can do on purpose” — Bechtel-Cone B29 modification employee. Shades of British empire described as created in a fit of absent-mindedness. Imma call it imperial serendipity. Or serendipitous imperialism.

The Marianas campaign in the Pacific war was fought just to create a staging base for B-29 missions. I didn’t know that. Must resist WW2 bunny trail. This read is for industrial production story.

B-29 missions going very poorly until they start using incendiary bombs based on magnesium “goop” developed as a byproduct of Kaiser’s magnesium plant. Hence firebombing of Tokyo. Worst attack of the war after the nuclear bombs. Damn. War is ugly.

Curtis LeMay, architect of the low-altitude firebombing strategy that brought Japan to its knees. Makes me really question whether Hiroshima was even necessary. Even this was rightly considered over the line. Ender Wiggin level total war.

83k civilians killed in the firebombing of Tokyo. Dresden was 25k. That’s about the US Covid deaths so far. WW2 was one bloody war.

Into the epilogue now. Another wall-of-statistics view. The US produced more than all other combatants combined AND did so with smaller fraction of economy AND kept consumer economy growing AND wages rising.

• $183B in arms
• 141 aircraft carriers
• 807 other combat ships
• 203 subs
• 52m tons of merchant shipping
• 88,410 tanks
• 257,000 artillery
• 2.4m trucks
• 2.6m machine guns
• 41b rounds of ammo
• 324,750 aircraft, 170/day since 1942
• Raw materials to UK and USSR
• $50B lend-lease

47% of economy as opposed to 60% for Britain and more for Germany and USSR YOY consumer goods also grew through war. Americans ate more meat, used more gas, electricity etc than before Hitler invaded France Guns AND butter. All mostly free market. Germany used 17m serfs

Total economy of production doubled, wages rose by 70%, Americans 2x more productive than Germans and 4x Japan (true even before war started). Underlined just how qualitatively next-level the US economy was with true mass production tech. Old world was still half artisanal.

A huge theme of this book has been how artisan production methods got properly mechanized, with British designs being ported being the best illustration. The brits were great at design but sucked at radical industrial scale. This was true since mid 1850s. They never did learn.

Though this book is almost embarrassingly pro-capitalist and paints a cartoon straw man view of unions, the picture rings true to me (admittedly a proud neoliberal shill). The unions failure was ultimately one of imagination. They never quite navigated the huge 100-year leap leap

Though the unions evolved from guild/trade union based to general mass organizing, like the British, they never quite mentally came to terms with the vast gap between 1830s artisan tech and 1930s mass tech. No wonder in 2020, dregs of unions are a humanities/liberal arts cause.

Book says post-war everybody was scared of depression/slump and big demob unemployment. To believe aggregate demand boom would persist into consumer economy was to be contrarian in 1946. Alfred Sloan was one. He turned out to be right.

Small blip in 1946 with 20% inflation and 3.9% unemployment then of course it’s all history as we know now. 30y of uninterrupted growth as the production engine of a bombed out world. Private capital investment tripled from 10.6B in 1945 to 30.6B in 1946. Stocks up 92% by 1947.

3 jobs waiting for every returning soldier. Women went back to homes and feminine mystique era. 2 decades of 4% growth. The China of mid-century Ofc Cold War superpowernomics and military industrial complex and stuff. Not all rosy. But war machine go brrr in peacetime basically

1948, Knudsen dies of brain hemorrhage. Health destroyed by the war years. Book argues that in postwar years the labor/new deal left tried to push a revisionist narrative of this being primarily a labor and government spending success, indication of new deal economics.

So ideological argument of book is:

• New deal failed
• Business won the war AND the peace despite New Dealer interference
• This was by Knudsen etc protecting it
• Post-war New Dealers rewrote narrative and took credit AND control
• They eventually killed golden goose in 30y

Needless to say I’m basically sympathetic to this narrative and hostile to the labor-sympathetic narrative, but I’m not inclined to buy the fully hagiographic version of either story, valorizing either capital or labor. The world is messy. Heroes and villains on both sides.

William Whyte Organization Man is a perfect natural sequel to this, both in terms of narrative continuity and ideological harmony. This is your basic business conservative/social liberal posture that eventually turned into neoliberalism in the 80s.

Aside: for Econ 101 in 1993, I had a Keynesian prof who used Samuelson text but the other section was taught by a Friedmanite prof. Pure chance that I landed in one section rather than the other (we couldn’t choose). Took me a decade after to discover supply side Econ in my 20s.

Ah, now for Kaiser’s story. He called the postwar boom correctly, and expanded ambitiously. Bought Willow Run and Willys the Jeep company. All failures. Lost his entrepreneurial cred. Still recovered financially with global success in construction. Plus aluminum go brr.

Kaiser dies in 1967 at 85. Outlived Knudsen by 19y. Much of his industrial legacy in heavy industry got outsourced or unionized away. That’s why today we know the name primarily as an employee healthcare firm. His legacy was decent non-union employee treatment I guess.

Okay done. This book was fun and honest on economic and engineering side, but a bit lightweight and hagiographic/demonizing on the human side. Despite my strong sympathies I’d give it only a qualified “good enough for topic, not great; needs ideological caveats” recommendation.

Thanks @jamesgiammona for the recommendation. Scratched my itch to apply WW2 industrial mobilization story lessons to Covid reboot and climate action.

My takeaway in that front is… not optimistic. The US lacks the kind of economic guts that allowed this story to happen. Outside of parts of Silicon Valkey tech economy this spirit is basically missing. And SV does not dominate the US as strongly as Detroit etc did in 1940s.

Also an equally worrisome trio of red, blue, and green new-deal crowds is at work today, pushing the exact same sorts of bad thinking as in the 1930s, requiring the same kind of protection/interference to allow a recovery.

• Green New Deal: bad climate thinking inside huge big tent sjw bullshit and MMT package
• Red New Deal: Profiteering gangsterism under maga pretending 1950s rewind is possible.
• Blue New Deal: keep protecting financialization and unproductive predatory Wall Street denture elites

A Freedom’s New Forge pretty much has to come out of Silicon Valley since there are no other candidate golden geese. But trends there are not promising either except for isolated pockets.

Okay gonna let this simmer.

Series Navigation

<< Notes: A Distant Mirror by Barbara TuchmanNotes: Astounding by Alec Nevala-Lee >>