When he challenged historical orthodoxy nearly 40 years ago, the Nobel Prize-winning economist Robert W. Fogel changed the way we look at railroads' impact on the U.S. economy; now, two Minneapolis Fed economists re-examine Fogel's work and find that railroads' contribution may be much greater than Fogel concluded.
Published December 1, 2001 | December 2001 issue
The world of today differs from that of Napoleon more than his world differed from that of Julius Caesar; and this change has chiefly been made by railways.
Thomas Curtis Clarke, 1889
Writers who have held either that railroads were crucial to American economic growth or enormously accelerated this growth implicitly asserted that the economy of the nineteenth century lacked an effective alternative to the railroad and was incapable of producing one. This assertion is without empirical foundation.
Robert W. Fogel, 1964
On May 6, 1856, a steamboat struck a pier of the first railroad bridge to span the Mississippi River, serving the Chicago, Rock Island and Pacific Railroad. The ensuing conflagration not only destroyed the boat but also burned a portion of the bridge, precipitating a landmark legal case that went beyond the parochial interests of the aggrieved boat and bridge companies, and pitted such water transportation partisans as the St. Louis Chamber of Commercewhich hoped to protect its city's role as steamer nexusvs. the burgeoning railroad industry.
Representing the Rock Island Bridge Co. when the case finally went to trial in 1857 was a young attorney who was later termed one of the best railroad lawyers in the West, Abraham Lincoln. Just five years later, as president, Lincoln would sign the Pacific Railway Act, setting in motion the construction of the first transcontinental railroad in the world. But in 1857 such plans were still a dream.
A former steamboat pilot, Lincoln investigated the scene of the accident, measuring currents and interviewing witnesses, to make his case that the crash was the result of pilot error. But he would also argue another, much broader, point when he faced his Chicago jurorsnational economic development. As the Chicago Daily Democratic Press reported when transcribing Lincoln's courtroom speech (taken here from the Collected Works of Abraham Lincoln, Volume 2):
St. Louis as a commercial place, may desire that this bridge should not stand, as it is adverse to her commerce, diverting a portion of it from the river; and it might be that she supposed that the additional cost of railroad transportation upon the productions of Iowa, would force them to go to St. Louis if this bridge was removed. ...
But there is a travel from East to West, whose demands are not less important than that of the river. It is growing larger and larger, building up new countries with a rapidity never before seen in the history of the world. He alluded to the astonishing growth of Illinois ... and the other young and rising communities of the Northwest.
This current of travel has its rights, as well as that north and south. If the river had not the advantage in priority and legislation, we could enter into free competition with it and we would surpass it.
This competition was already keenly felt, Lincoln noted later in his comments:
What mood were the steamboat men in when this bridge burned? Why there was a shouting, a ringing of bells and whistling on all the boats as it fell.
In a victory for the railroad, the jury deadlocked; an Iowa court later found against the bridge builders, but the Supreme Court overruled and declared that railroads could bridge rivers. "Had Lincoln never done another thing for the railroads, he had earned their gratitude on this one," writes Stephen Ambrose in his history of the transcontinental railroad, Nothing Like It in the World.
Lincoln was not alone among his contemporaries in his belief in the necessity of the railroads for the economic growth of the budding nation, nor would he be alone among future historians and other observers of the late 19th and early 20th century U.S. economy. The list of encomiums for the railroads and their supposed multitudinous benefits rarely lacked for hyperbole.
And this praise often encompassed more than pecuniary benefits. Asa Whitney, the early visionary and professional proselytizer for a transcontinental railroad, intended not only to "annihilate distance," but with it, "ignorance, want and barbarism," according to David Haward Bain in Empire Express: Building the First Transcontinental Railroad. Whitney wrote: "[W]e shall have the whole world tributary to uswhen the whole commerce of the vast world will be tumbled into our lapwhen this vast and now useless waste and wilderness (and it ever must be so, without this road) shall become, not only the thoroughfare of the vast world, but its garden, feeding, clothing, comforting and enlightening millions, who are now starving, homeless, naked, ignorant and oppressed; and who can oppose such a work?"
Enter, the iconoclast
Who, indeed? When Robert W. Fogel set out to reconstruct the railroads' influence on U.S. economic growth, he knew he was taking on one of U.S. history's most sacred cows. "To the men of the Gilded Age the primacy of the railroad in the promotion of American economic growth became an indisputable fact," Fogel wrote in his 1964 book, Railroads and American Economic Growth: Essays in Econometric History. Four years prior, Fogel had begun his reassessment of the railroads with the publication of his master's thesis, The Union Pacific Railroad: A Case in Premature Enterprise.
While others of Fogel's era were beginning to question the indisputability of this so-called fact, nobody had yet applied the mathematical rigor of econometrics to the question. This contribution, along with analysis of other issues in U.S. history, including slavery, led to Fogel's 1993 Nobel Prize in Economic Sciences.
Fogel's 1964 work addressed four broad questions, and while we won't discuss each of these questions in this article, it is useful to consider the scope of Fogel's work to appreciate the effect he had on current research:
Did the interregional distribution of agricultural products depend on the existence of the long-haul railroad?
Were there social gains due to the availability of railroads in the intraregional, or short-haul, distribution of agricultural products?
Is there an empirical justification for the "take-off" thesis of the stages-of-growth theory in American economic development? This theory suggests that of the five such stages of economic growth, the takeoff period represents the crucial turning point in modern societies; in this case, railroads were considered key to such a period.
How can quantifiable magnitudes be determined with only fragmentary data? Here, Fogel challenges the assertion that the growth of the American iron industry prior to the Civil War depended on railroads' consumption of iron.
In addition to applying economic models and sophisticated statistical measures to the above questions and raising some sound skepticism about previous claims, Fogel went even further and attempted to measure the precise amount that railroads contributed to U.S. economic growth in one representative year1890. In the end, Fogel came up with the decidedly underwhelming figure of 5 percent (at most) of gross national product; that is, the United States was already well-served by the water transport industry and, likewise, railroads only contributed up to 5 percent to U.S. national income in 1890. Needless to say, Lincoln would not have called on Fogel as an expert witness.
How competitive is competitive theory?
To make his case, Fogel had to make an assumption about how prices and productivity in the water transport industry would have developed without the addition of railroads. The theory he chosecompetitive theoryis the basis for the critique of Thomas J. Holmes, University of Minnesota professor of economics and visiting scholar at the Minneapolis Fed, and James A. Schmitz Jr., Minneapolis Fed senior economist. Holmes and Schmitz recently published an article in the Minneapolis Fed's Quarterly Review ("Competition at Work: Railroads vs. Monopoly in the U.S. Shipping Industry," which challenges Fogel's basic assumption and demonstrates the competition at work in long-distance transportation.
But before addressing Holmes and Schmitz's challenge, we must first briefly describe Fogel's application of competitive theory and how he arrived at his 5 percent number. Competitive theory suggests, in this case, that prices and productivity in water transportation over the long run would have evolved independently of the railroads; that is, water transportation was its own competitive environment uninfluenced by the railroads. Hence, and this bears restating, the water transport prices that were charged in 1890 were precisely those prices that would have been charged if no railroad was ever built.
This leads to a rather simple calculation: We can take a bushel of wheat, for instance, check the price of transport via water or rail in 1890, and then multiply that price differential by the number of units to receive a total difference. This was Fogel's technique, and he applied his measurements to agricultural commodities, in general, and to wheat, corn, beef and pork, in particular. As suggested above, Fogel examines such costs on two levels, long- and short-haul routes. In the case of long hauls, water would have been used in the absence of railroads, and with short hauls, water and wagon would have been the alternative modes of transport.
When he summed all the extra costs of moving agricultural products by water and wagon alone in 1890, Fogel arrived at a figure of 1.8 percent of GNP; he then speculated the extra costs for all goods, agricultural and nonagricultural, to be no more than 5 percent.
Upon further review
For now, we'll put that number aside and look more closely at competition within the water transport industry, which forms the basis of Holmes and Schmitz's work. Essentially, the economists' counterargument with Fogel rests on two points: that there is a strong tendency toward monopoly in the water transport industry that results in higher prices and inefficient technologies, and that railroads greatly weakened these monopolistic tendencies over time.
On the first issue, there are three features of the water transport industry that make it prone to monopolistic behavior:
Holdup points: A group can establish a toll along a trade route if it can capture a certain point along the line through which all users must pass. Dock worker unions in the 19th century were able to do this at places where cargo was moved from one form of transport to another, for example, from riverboats to oceangoing vessels. A large portion of a trade route may be competitive, but a group at a holdup point can extract profits as if they had a monopoly on the whole route.
Scale economies: In some markets the efficient scale of production is large relative to the size of the market. For example, to transport cargo efficiently from Coastal Point A to Coastal Point B may take the services of a large shipping company, but the actual A to B market may be so small that once one large company has captured the market there is no incentive for another shipping company to compete. Consider some present-day airline marketssmall city to small citythat are served by one airline; since that airline faces no competition from another carrier it can exploit the dependent travelers.
Government regulation: As Lincoln alluded to in his courtroom speech, there may be no U.S. industry with more protective legislation and regulation than the water transport industry. The authors discuss three types of laws that have protected this industry: those prohibiting foreign ships from transporting between two U.S. ports, those requiring U.S. shipping companies to purchase U.S.-made ships and those requiring U.S. shipping companies to hire a certain portion of U.S. citizens.
In Holmes and Schmitz's analysis, dock worker unions are shown to have significant power to, among other things, exploit holdup points and charge prices above competitive levels. Hourly wage rates, for example, at such holdup points as New Orleans, the Great Lakes and New York, were two to three times higher than those paid to other freight handlers. Further: "[W]e present evidence that dock worker unions clearly chose to use inefficient technologies." The intentional use of inefficient technologies does not necessarily square with the conventional view of monopolistic behavior, but the facts (and theory) in this case support the claim.
Reprinted from Holmes, Thomas J., and Schmitz, James A., Jr. 2001. Competition at Work: Railroads vs. Monopoly in the U.S. Shipping Industry. Federal Reserve Bank of Minneapolis Quarterly Review 25 (Spring): 20.
On the second pointthat development of the railroads significantly reduced the monopolistic tendencies of the water transport industryHolmes and Schmitz offer the example of cotton shipments from Louisiana to New York. Prior to railroads, the only option to a coastal route, which was subjected to high holdup costs at New Orleans, was up the Mississippi River and eventually to the Erie Canal, a path subject to holdup points all its own. With the advent of railroads, cotton could be shipped to other ports besides New Orleans or, for the first time, could be shipped overland to New York without the cost and time delays associated with wagon hauling.
Holmes and Schmitz present evidence for the reduction of dock wages due to railroad competitionfrom both the 19th and 20th centuriesas well as for their claim that railroad competition led to more efficient practices in the water transport industry. Regarding the latter, Holmes and Schmitz write, "Some dock worker unions in New Orleans changed their work rules over the period 1885-1905 so as to increase the pace of work 20 percent. Some unions also dropped their work rules which required unskilled work to be done by skilled workers."
In the end, the two economists make their case that researchers have not properly recognized the contributions made by railroads to reduce the effects of monopoly in the water transport industry. In other words, if researchersbased on Fogel's workhave quantitatively shortchanged this contribution, then they have also underestimated railroads' contribution to economic growth. Five percent of GNP doesn't fit the bill. As Lincoln might have observed: Five percent of anything, let alone GNP, wouldn't be enough to get a bunch of steamboat men whooping and whistling and ringing their bells.
So what's the right number?
Good question. And one that Holmes and Schmitz don't answer in their current paper. However, they provide a quantitative exercise which"when fully implementedwill likely yield a number "much larger" than 5 percent.
In this exercise, the two economists first sketch Fogel's calculation by including the extra cost of transportation with monopoly. Next, they apply recent advances in general equilibrium theory, which suggest that those costs of monopoly were likely large in the water transport industry. (General equilibrium theory is a highly abstract theory about how all markets fit together and reach equilibrium.) Holmes and Schmitz plan to explore this line of research in future work.
"Rather than throw out a new number here," Holmes and Schmitz write, "we will end by pointing out that the transportation sector is a special sector, closely linked to the rest of the economy. Changes in this sector reverberate widely. At the least, improved transportation increases specialization and competition in other sectors. These benefits thus magnify the benefits of resource savings from simple reductions in the cost of transportation."