The Dawn of Innovation (34 page)

Since almost all the steamboats were built locally, the New York–New Jersey harbor was lined with massive foundries and engine and machine shops, as well as shipyards of the largest dimension. By 1850, they had produced thirty-eight major steamships, including two for foreign governments. William H. Brown was the biggest yard, accounting for about a quarter of the hulls. Other important yards included William H. Webb, Westervelt & Mackay, Simonson's, and Jacob Bell. There were also multiple dry docks. Webb's, built in 1854, was the biggest at 325 by 99 feet with a 38½ foot depth; it had twelve pumps and was operated by two large steam engines, one on each side, each with a locomotive boiler. About half of the steamboat engines were made by the Allaire Works on the East
River, a pioneer of the marine compound engine,
^bf
with the Novelty Works likely in second place. (Vanderbilt gradually absorbed both the Allaire and Simonson businesses to improve his profit margins.
⁴⁵
)

Transatlantic shipping was largely the province of American sailing ships until two British steamships, the
Sirius
and the
Great Western
, arrived in New York Harbor within twelve hours of each other in the spring of 1838. The
Great Western
was much the more formidable ship—it had twice the tonnage of its rival and was a third greater in length and much faster (it had left England three days later). It had been designed by Isambard K. Brunel with an engine from Maudslay Sons and Field—respectively the sons of Mark I. Brunel and Henry Maudslay, who created the great pulley-block factory at Portsmouth. As was the convention until well into the second half of the century, both ships were under steam power and sail the entire voyage to conserve fuel. A number of other British competitors jumped into the fray, and the Cunard Lines won the premier position by virtue of a coveted mail subsidy—a generously calculated payment for handling British-American mail.
⁴⁶

E. K. Collins of New York, scion of a generation-old transatlantic packet-boat and freight operator, responded by launching four steamships in 1849–1850, all of them built in New York harbor, buttressed by an even more lucrative mail subsidy from the American government, justified on “national interest” grounds. The first two, the
Atlantic
and the
Arctic
, had engines from the Novelty Works, while the engines of the
Pacific
and the
Baltic
were from Allaire. The engine cylinders were all ninety-five or ninety-six inches in diameter, while the pistons had nine- or ten-foot strokes. There were other steamship lines, but the Collins and Cunard ships were on the same routes and even used the same terminal in New York, so they were most visibly engaged in a head-to-head competition. On the early matchups, the Collins ships—besides being much bigger—proved much faster. Cunard commissioned two big new ships, the
Asia
and the
Africa
, which were a marked improvement. But they did not decisively trump the Americans until the launch of the very large and very fast
Persia
in 1855.
⁴⁷

Cornelius Vanderbilt, in the meantime, had been gaining experience as an ocean steamship operator by running a lucrative Pacific line to take advantage of the California Gold Rush. The route went from New York to a port on the coast of Nicaragua. Passengers then embarked on a combined river-lake-transit road trip across the peninsula to meet steamships to San Francisco. His first ship commissioned for the East Coast leg was the
Prometheus
, which in 1850 made the 5,600-mile run, including stops at Havana on the way down and at New Orleans on the way back, in the extraordinary time of nineteen days, while consuming about a third less coal than any comparably sized ship would have required.
⁴⁸

The secret of the
Prometheus
's performance was the engine. All of the oceanic steamships were side-wheelers. British design dogma, which had been adopted by Collins, required that oceanic ships have their engines located low in the hull for stability, so operating the side wheels required complex gearing with poorer operating efficiency. Vanderbilt insisted on a traditional walking beam engine, with the beams on the deck and the drive shafts connected directly to the side wheels, and proved the naval architects wrong. The walking-beam engine was much lighter and more efficient, and the ship, like most side-wheelers, was very stable.
⁴⁹

Vanderbilt launched the
North Star
in 1853, and in a slow market, he adopted it as his personal yacht. At 2,500 tons, with 260 feet at the waterline, 34-foot side wheels, and two walking-beam engines, it was among the largest and most capacious, most lavishly outfitted, and fastest commercial-scale ships in the world. He took his family and retainers on a world tour, in part just to introduce himself to the international scene but also to advertise the capabilities of American shipping and to underscore his challenge to the Collins mail franchise, which Vanderbilt bragged, doubtless correctly, that he could operate at half the price.
⁵⁰

Vanderbilt returned from his tour, refitted the
North Star
as a passenger ship, and began work on the
Vanderbilt
, even bigger and faster than the
North Star.
Plausibly, the United States looked poised to seize shipping
supremacy from the British. It didn't happen. By the eve of the Civil War, the Collins line had collapsed, and Vanderbilt had sold off most of his steamship interests, turning over his larger ships to the government for war transport.

What happened? The immediate causes were economic. Congress decided to make drastic cuts in the Collins subsidy, just as the economic crisis of 1857 devastated freight and passenger revenues. More subtly, the British, acting on pure free-trade doctrine, had opened their registries to foreign shipping in 1849. As British shipping interests had predicted, the Americans, with their unmatchable clipper ships, quickly dominated international freight shipping, most especially the lucrative China and India trade.

But theory for once was triumphant. All sensible American investors plowed their money into the wooden sailing-ship industry. When the Crimean War ended and the British began a full-scale transition to iron ships
^bg
and propeller drives, Americans were simply not in the game. And at the end of the day, compared to the opportunities in the American West and in railroads, ocean steamships did not look like an especially attractive business, since it was so heavily dependent on subsidies.
⁵¹

After the Civil War, the Pennsylvania Railroad attempted to re-create a steam-based merchant marine industry by building four creditable ships, but it finally merged them into a Belgian combine in 1884. In 1902, at the height of an American leveraged buyout boom, J. P. Morgan led a roll-up of two American and two British shipping lines. The new entity, International Mercantile Marine, was arguably insolvent on its first day of business, and as an investment, was an unmitigated disaster.
⁵²

Matthias Baldwin was a Philadelphia jeweler, a whitesmith—a metalworker in silver, pewter, and tin—and clearly a gifted mechanic. When he
and his partner moved to a bigger shop in 1824, he built a stationery steam engine to power their machinery. Although it was his first engine, it was remarkably efficient and had enough useful innovations to attract a number of orders from other Philadelphia workshops. When the partnership broke up in 1829, Baldwin opened his own workshop as a jobbing machinist and specialist in steam engines.
⁵³

This was at the dawn of the railroad age, and in 1830 Baldwin was commissioned by the Franklin Institute to build a realistic working model of a steam-driven railroad. That led to contract from the new Philadelphia-Germantown Railroad to build a full-scale engine. A second contract with another line followed a few months later, and others after that; in the single year of 1835, he built fourteen locomotives.

From the time Baldwin entered the business, he pushed the product boundaries. The second engine he produced incorporated Jervis's four-wheel swiveling truck at the front of the engine, which improved its performance on curves. Baldwin followed that up with a bigger three-axle engine, including the truck, and then added a power train to the truck. In both 1836 and 1837, Baldwin sold forty locomotives, employed some three hundred men, and became one of the wealthier Philadelphia manufacturers. An 1838 Treasury survey showed that there were 350 steam locomotives in the United States, 122 of them from Baldwin. Since 78 engines had been imported, Baldwin had 45 percent of domestic production, and had begun to win contracts on the European continent.
⁵⁴
Among some twenty-two other American locomotive producers, Baldwin was clearly the dominant player.

When the 1837 Bank Panic shut down most railroad construction, Baldwin was exposed to the perils of big-ticket capital goods production. Since he supplied much of his customers' financing, he found himself desperately squeezed for cash, scraping through by stonewalling suppliers and groveling to creditors. He preserved the company by working out an informal receivership with his creditors that allowed him to retain control until markets recovered, although he lost his position as industry leader. For Baldwin, a deeply religious man with high business principles, it was
a time of terrible strain.
⁵⁵
The half-century-long railroad boom that followed the Civil War finally put the company on a solid footing, and by the 1880s, it was the world's largest locomotive maker.

Despite the financial headaches, between 1850 and 1870 Baldwin made extraordinary productivity gains, as shown in
Table 6.2
.

 

That accomplishment is the more impressive if one considers the much greater complexity of the 1870 over the 1850-vintage locomotive, plus the fact that dominant railroads like the Pennsylvania and the Baltimore & Ohio were increasingly specifying highly customized designs.

How did they do it? The original Baldwin manufacturing strategy was, at heart, that of the classic machinist's job shop. But superimposed on that was a prototype for the modern systems integrator, like Boeing. Boeing has great manufacturing skills and is arguably the world leader in mega-wing design and production, but the vast majority of the millions of components in one of its planes are sourced from outsiders, including many of the most critical, like the engines and the avionics. The ability to manage that complexity is the deep skill that makes Boeing such a successful company.

Baldwin's challenge was similar. By the 1840s, standard locomotives weighed 40,000 tons and had some 4,000 parts, far more than Baldwin could make itself. Many of the components, moreover, like boiler plates, axles, wheel castings, and large flues, were beyond Baldwin's shop capacities. From the earliest days, therefore, Baldwin formed strong relationships
with key vendors. Each new contract produced a flood of parts orders that could usually be specified by the variations from a previous order. On a standard engine delivery schedule of sixty days, the first four to six weeks were devoted to making, or purchasing and finishing, the components, followed by two weeks of building the engine. Several engines could be worked on simultaneously, but each one, along with virtually all its components, was essentially built from scratch, and few of the engines were ever completely identical. Throughout the decade, depending on the market, Baldwin turned out a locomotive every ten days to two weeks.

Baldwin did that with almost no management to speak of. As the company entered the 1850s, Baldwin was the sole owner and chief executive. He had elevated a shop foreman, Matthew Baird, as the de facto number two, although Baird spent most of his time with customers. The rest of management consisted of a general superintendent, George Hufty, who oversaw day-to-day production, and one more executive, a shop floor veteran with a flair for numbers who took care of the accounts. Baldwin, besides managing day-to-day financing issues, was immersed in tooling strategies and product design, a major source of the company's reputation. And that was it—four managers in a four-hundred-person company, two of them in sales and accounting rather than line positions, supported by a handful of clerks.

Operating management lay with the shop foremen. There were seven shops in 1850: foundry, boiler, smith, painting, and three machine shops. Each foreman was effectively the master of his house. Baldwin and Hufty would establish a production schedule, set the timelines for each of the shops, and order the required components and raw materials. From that point, the shop foremen were on their own. In effect, they were running sizeable businesses with about fifty workers each, and like Baldwin, they had mastered every detail of their shop work, hired and fired their workers, purchased their own shop supplies, and organized their internal work flow as they saw fit—with the major difference that they had only a single customer. The relation with major suppliers was somewhat similar. Over the years they must have developed high levels of trust, and Baldwin could not have met its schedules without top-priority treatment by all its vendors.