Freedom's Forge: How American Business Produced Victory in World War II

by Herman, Arthur

  Starting in 1933, congressional legislation had placed sharp restrictions on how much war suppliers could make on their government contracts. During the First World War, cost-plus contracts were common, meaning that the government would pay all expenses relating to making an airplane or artillery gun, in addition to a fixed fee or percentage of cost—8 percent was fairly standard. The postwar reaction against “war profiteers” led Congress and the Treasury Department to impose sharp curbs on the profit companies could make on orders larger than $25,000. They also required an advance audit to guarantee that the company’s profit would be no more than 8 percent even before the contract was signed.19

  In addition, every government contract for a new airplane or tank or vehicle required bidding companies to pay for the production of their prototype and the new machine tools to manufacture it. No money was ever advanced, even to the winner of the bid. As one executive from Boeing put it, “There was no sound of coin in Uncle Sam’s jeans.” An aircraft maker looked at an average of a half-million-dollar investment just to enter a bid—with no guarantee of winning.20 Even if he did and problems or delays developed, the government was not above pulling the contract, leaving the company high and dry.

  Then there were the tax laws. Every American company’s investment in new plant construction, tools, and other physical resources necessary to produce a plane or tank or aircraft, even with a contract, took sixteen years in order to be fully deducted as a business expense. Knudsen saw at once that these amortization rules (Knudsen had to get John Biggers to explain to the president and Cabinet what amortization meant) made the short-term investment in plant, property, and equipment necessary for the defense buildup almost impossible.21

  “Mr. President,” Knudsen said, “do you want statistics, or do you want guns?” If the latter, he explained, then amortization should be drastically shortened to five or six years. Hitler’s German companies, he pointed out, worked on a seven-year rule. Companies would get their investment back quicker, and be more willing to take risks on manufacturing things that had no commercial value to them but were crucial to the defense effort.

  “The government can’t do it all,” Knudsen told Roosevelt. “The more people we can get into this program”—in other words by offering incentives instead of threats—“the more brains we can get into it, the better chance it will have to succeed”22

  Knudsen also insisted that a “letter of intent,” meaning an official War or Navy Department letter stating the government’s intention to do business with a particular firm before a formal contract was drawn up and signed, should be enough to get a company advance funds from their bank—and to protect the company’s out-of-pocket expenses in case the contract never went through. It was a practice he borrowed from the British, and critics would hound him for it, decrying the fact that he had abandoned the costly, time-consuming process of competitive bidding. But Knudsen sensed that the time for slow, deliberate action was over. The government had to be willing to work with those companies willing to work with it.23

  Roosevelt was deeply dubious. He pointed out that never in the history of the United States had such a provision been made for government contractors. Knudsen replied it was time to shatter precedent. Otherwise, he said, they might never get a job of that size done in time. Roosevelt’s 50,000 planes a year would remain only a pipe dream, while Hitler’s Luftwaffe ruled the skies over Europe.

  With Stimson firmly backing Knudsen, the White House at last conceded. In the next weeks and months, people would catch a glimpse of Knudsen cornering some industrialist after an NDAC meeting and saying, “George, in that plant of yours—Plant Number Four, I mean—I’ll make sure you get the tools, see?” There would be a grateful handshake, perhaps an appreciative pat on the back.

  Then at parting, “I’ll see that you get a letter. I won’t forget.”24

  At the same time, everyone else’s attention was on what was happening in Britain.

  On any given day in July 1940, Winston Churchill half expected to see German paratroopers landing on the outskirts of London. On July 12 there was a serious discussion in the War Cabinet about whether the government should encourage the populace to attack German invaders with scythes and stones.25 Meanwhile, fighting off German air attacks on British shipping in the Channel became a top priority. In Washington the British demands for war materiel, especially planes, came with alarming frequency. Knudsen was forced to confront a new truth. American industry was going to have to satisfy the needs of war on both sides of the Atlantic at once.

  Matters came to a head on July 23 at a meeting in the Treasury Building. Knudsen began peppering the British Purchasing Commission’s Purvis with questions about how many planes the British would need, and when. Purvis gave his answer: 1,000 planes a month. Britain was putting all its own production into making fighters, Hurricanes and Spitfires, in order to halt what was promising to be a massive German onslaught from the air. It needed every other kind of plane America could provide, from bombers to trainers.26

  Knudsen swallowed hard. Right now America’s aircraft manufacturers were barely producing 550 a month, with 250 a month slated for overseas. He promised that the United States could hit that number for Britain by the end of 1941, by which time the Army Air Corps could count on 2,000 a month for its own force. President Roosevelt’s call for 4,000 or more a month still looked a long way away, but this would get things started.

  The next morning, however, Purvis stopped by Treasury Secretary Morgenthau’s office. The truth was, he said, 1,000 a month might not be enough. By the end of 1941—assuming Britain survived—they would be needing closer to 4,000.

  Henry Morgenthau was a dry, precise man who refused to be surprised. In spite of himself, however, he had to smile.

  “Pass the ball to Knudsen,” he said. “He’s the kind of production man who will rise to a challenge like that and meet it.”27

  So Purvis did. Once Knudsen had recovered from his shock, he promised he would see what he could do.

  The result was a sharp change in the direction, but not the pace, of aircraft production in the United States. Almost incredibly, the U.S. now agreed to slow its own buildup in order to keep Britain in the air.28 For all of 1941, one-third of all warplanes produced in the United States—and one-half of all tanks—would be slated to be sent to Great Britain. It was an extraordinary sacrifice of priorities, for the sake of preventing Britain’s defeat. Still, figuring out how to achieve production numbers like that meant more late nights at Knudsen’s Federal Reserve office huddled with his aviation expert, Dr. Mead.

  Knudsen and Stimson were already pushing for radical changes in the work schedules for America’s aircraft plants: adding a night shift, for example.29 Still another question haunted the back of Knudsen’s mind. By the end of 1940, British wartime orders in the United States would be almost half of what the federal government itself was spending on armaments.30 How was Britain going to pay for all this? Already he sensed that a day of reckoning was coming, even supposing Britain survived. And as Stimson’s reports were telling him at the end of July, those chances did not look good. The problem of how to supply a customer, in this case Britain, who might suddenly lose the ability to pay arose for the first time on that sweltering morning of July 24. It would not go away until the passage of Lend-Lease eight months later.

  At the same time, Knudsen was getting grief from the other direction. The Army was starting to panic. There was real fear that the British demands for planes and tanks and guns would not leave enough for the Army’s own revised mobilization plan. On the twenty-fifth Knudsen and Stimson met to discuss the complaints against NDAC from various Army administrative heads who felt the agency was moving too slowly—while Harold Vance, Bill Harrison, and the others felt the men in uniform were shutting them out from major purchasing decisions.31

  For example, the rising star of Stettinius’s team handling materials and supplies was the former president of Sears and Roebuck, Donald T. Nelson. T
all, owlish, and bespectacled, Nelson was a little shocked at the Army’s intransigence regarding uniform buttons. It told him that it had to have horn or ivory for its uniform buttons, as in the First World War. Nelson pointed out an American company called Rochester Button was ready to make thousands of perfectly fine celluloid buttons, and that horn and ivory had to be imported from South America or Czechoslovakia—the latter now under Nazi rule. Horn or ivory, the Army said, and for the time being that was where things stuck.32

  Then there was the problem of what to do about tanks.

  Purvis and the British were starved for them. Churchill told Purvis to ask Knudsen and the Americans for help. So on August 6, 1940, Knudsen called a summit meeting with Purvis, British and American officers, and representatives from the truck, railroad, and heavy equipment industry, including American Car and Foundry, the only private company currently making tanks in America besides the Army’s own Rock Island Arsenal. Knudsen made them put their heads together and come up with a bulked-up version of the Army’s current light tank. It would carry a 75mm gun—equal to the heaviest German tanks at the time—and a 30mm gun mounted on a turret. Purvis announced that this new model, the M3 Medium, would be acceptable to His Majesty’s Government. But it would want no fewer than one thousand a month.33

  The Army officials were stunned. American Car and Foundry was barely making thirteen a month of the old model. How on earth could they ever meet numbers like that—not only for Britain but for the United States? They were discussing possibly asking Baldwin Locomotive, one of the country’s leading manufacturers of heavy equipment, when Knudsen cleared his throat.

  “How about a car company?” he asked.

  As everyone else in the room looked at one another, Knudsen went back to his office. Donald Nelson watched from around the corner as Knudsen picked up his phone. “Give me Detroit,” he asked politely, and within minutes he was connected with K. T. Keller at Chrysler.

  “We have a problem, K.T.,” he said. “We have to make more tanks than any corporation has ever made in the past. Can you do it?”34

  Kaufman T. Keller was Detroit old school. He knew Knudsen from the automobile’s pioneering days and knew the respect his old boss, the ill and retired Walter Chrysler, had for the industrious Dane. Keller replied, “Sure. When do we meet?”

  Tomorrow, Knudsen answered. With that he flew off to Detroit and the pair spent the day bent over a desk sketching out plans for a facility that would be able to produce up to five hundred tanks a month.

  It was an impossibly formidable task. Chrysler was moving into uncharted territory for an American car company—and neither Keller nor his engineers had ever seen a tank. But two priorities stood out at once. Knudsen and Keller agreed there was only one person who could design the right facility: Albert Kahn. The second was that it would not just have to be a new factory, but a new way of making tanks, as well.35 It was Chrysler’s engineers, not the Army’s, who would have to figure out how.

  At Knudsen’s insistence, on August 9 the other members of NDAC met to approve his and Keller’s plan, and the War Department immediately authorized $20 million for construction.36 It was also agreed that the tank engines would come not from Chrysler but Continental Motors, a nearly defunct automotive company based in Muskegon. Knudsen knew their president, Jack Reese, like he knew everybody. He figured if anyone could come through, it was Jack Reese.

  Continental was a dying business when the War Department contract for two hundred engines a month arrived. Reese was a cigar-chomping fireplug of a man who, like Knudsen, worked best in shirtsleeves and a hat. Reese had been made president in 1939, and was Continental’s last hope of turning the company around.37 In addition to firing executives who collected salaries but didn’t do any work, one of his plans had been to sell the company’s big Detroit plant and transfer everything—machines, workers, tools, the works—to the more efficient operation in Muskegon.

  A year had passed, and Reese found no buyers. That proved lucky for the Army, because now Reese had an empty facility in which to start making tank engines. Even better, Reese and his engineer offered to redesign the 440-horsepower aircraft engine they had already starting tooling up for in order to get back in business, to fit into the M3 tank. The Army specified an air-cooled engine, and this was an air-cooled engine, even though it had been designed for the skies, not fighting on the ground. Still, the Army jumped at the chance to get into production months before anyone thought they could. The Continental end of the deal was signed.38

  Across town at Chrysler, Keller’s chief engineer was on the job. Ed Hunt was, like Reese, a physically formidable man, built low to the ground with a natural glower and no-nonsense attitude. Army officers called their visits to Chrysler to see him “going to Fort Hunt.” The first time he and Keller showed up at Rock Island Arsenal, outside Davenport, Illinois, the Army showed them several hundred pounds of tank blueprints. Hunt emphatically shook his head.

  “I’ve never seen a tank on the hoof,” he said. “If I’m going to build ’em, I’m gonna have to see one. How about it?”39

  That afternoon, Hunt, Keller, and their Army escorts took one of their prototypes out for a run. When they were done, Hunt stepped out and asked, “Is that what you want?”

  “Yes,” the general said, “but lots of them.”

  K. T. Keller glanced at Hunt. “All right, Ed. How about it?”

  The Chrysler engineer thought for a moment. “Well,” he finally said, “I guess we’d better not get any bright ideas of our own. We’d better let the Army design ’em. We’ll just make ’em.”

  The final design of the M3 came out in early 1941. By March, Chrysler’s new facility, dubbed the Detroit Arsenal Tank Plant, was finished—just seven months after Keller broke ground for it driving his own tractor. Hunt and his associates put in fourteen- and eighteen-hour days, making their first tank entirely by hand. Albert Kahn’s dream factory took shape overhead and around them, with empty bays where the ordered machine tools would eventually go.

  The Chrysler men then discovered the Army-designed M3 had problems. Certain elementary engineering mistakes had crept in, which the Army’s Ordnance Board had missed. There were tank treads that worked fine on regular roads but slipped and slid in muddy ditches. There were air-cooled engines, which Continental was contracted to build and Chrysler to install, but which were time-consuming and expensive to machine, and which required cool air to cool. That was going to be in short supply in Egypt’s Western Desert, where the M3 would first be deployed.

  Then there was the problem of the chassis springs. Hunt and his men scratched their heads over the drawings. They had never seen springs like this: certainly nothing like anything used by the automotive industry, not even for the heaviest trucks. They looked into every manufacturing nook and cranny. Then one of Hunt’s assistants came up with the answer, in a railroad manual dating back before World War I. They were so-called volute springs originally designed for freight cars. The railroad industry had abandoned them long ago. Now the Army was using them for its vaunted M3 tank.40

  Keller and Hunt had to fight to persuade the Army to let the car company, which routinely spent huge sums designing and testing chassis springs, give them a new spring design. In the end, Chrysler would basically redesign the entire vehicle—just as later, Chrysler, Ford, and GM would reengineer the M3’s successor, the M4 Sherman. The M3 Grant would finally be ready to fight Rommel in the desert in 1942 with Continental’s less-than-ideal engines. But months, and not a few lives, were lost because the Army insisted on pushing ahead on its own design without once asking if the professional experts might do it better.

  It would take some time before American companies learned to challenge the War Department on how to design and build the weapons it wanted (aircraft makers of course had known this for years). Meanwhile, Knudsen himself had to step up to deal with the even bigger problem stalking the M3’s makers, that of rivets.

  Ed Hunt confessed he had never seen
anything like it. Trying to get holes bored through the two-inch steel plate took forever. He called in one of American industry’s legendary toolmakers, Hank Krueger, to help out. Slowly the problem, and a dilemma, took shape. There were two kinds of steel plate for tanks, cast homogeneous armor, or CHA, and rolled homogeneous armor, or RHA.41 CHA was easier to work with but hell to cut. RHA was clearly the way to go, but riveting so many separate plates raised manufacturing costs to the roof.

  Knudsen proposed a solution. Why not weld the plates instead? Army engineers scoffed. There was no way any welding job, no matter how carefully done, could hold these heavy steel plates together and withstand the kind of rough handling a battle tank had to endure—including shell fire. They had tried it once, they told him. Honestly, rivets were the only way to go.42

  Knudsen, however, knew people who knew a good deal more about welding than the Army did. There was Bill Smith, for example, whose company, A. O. Smith Corporation of Milwaukee, did welding for car frames and for massive steel piping used by oil companies. Knudsen had already contacted him about using high-efficiency welding techniques instead of rivets for the Maritime Commission’s shipbuilding program (which were later adopted across the industry).

  Sure, Smith responded, you can weld steel plate of that kind of thickness. Then there were welding people Knudsen knew at General Motors, and so he set off once again for Detroit.43