For anyone on the green side of fifty who didn't start historical browsing in the playpen it may be quite hard to see the present appeal of statistical theory and method in perspective. To one lucky enough to have been a student abroad in the 1920's, it is merely one of the consequences of a fundamental shift, which was firming in that decade, in conceptions of the economic historian's job. Essentially the shift consisted in making the economy and the social institutions in which it is embedded analytically distinct. Voices from the Polanyite school still claim that this step was as wrong as Adam's eating of the apple. Milder critics complain only that some of us let economic analysis run away with the ball to the neglect of social analysis and of the interplay between the two. For workers on the recent past this is defensible, because the heavy fall-out of purely economic data clamors to be dealt with in its own terms. The preindustrialist, who has to dig harder for data, and seldom turns up such pure economic ore, is more inclined to think in terms of interplay.

The shift in the locus of European trade from the markets of the Mediterranean to the North Atlantic overthrew a centuries old pattern of commerce and established the basis for the predominant role of North Atlantic Europe in the era of industrialization. While the expression “commercial revolution” no longer has quite the currency that it once enjoyed, students of the early modern economy have not been negligent about trying to understand the causes of the commercial shift. The impact of entrepreneurship and Weltanschauung, capital accumulation, technical innovation in shipping and industry, and the economic and political organization of nation-states have all received attention from students of the age.

Recent accounts of medieval agricultural organization begin not with a description of “the” manor but with a warning that no single archetype may be considered appropriate. From place to place, from generation to generation, a property might be exploited as a single unit, divided between tenancies and demesne, or wholly partitioned into tenancies; and the land would be worked by a correspondingly varied mixture of “employees” and independent labor. This complex historical record is still far from being explained by a suitable model. In the secondary sources, the varying size of the demesne is most frequently attributed to changes in relative prices: the demesne is buoyant, it is argued, when output prices are high and wages low; when prices fall and wages rise, rents become more attractive and the demesne is parcelled out. The difficulty with this explanation, however, is that rents are not independent of output prices and wages: in simple economic theory, rent so adjusts to prices and wages as to always equal the surplus that could be earned by direct exploitation, and the proportion of tenant land on an estate is a matter of indifference. This does not of course mean that it was so in fact; it does mean, however, that one must carefully specify which of the many considerations neglected by that simple theory are in fact critical to the problem at hand.

We are right to see power, prestige, and confidence as conditioned by the Civil War. But it is a very easy step to regard the War, therefore, as a jolly piece of luck only slightly disguised, part of our divinely instituted success story, and to think, in some shadowy corner of the mind, of the dead at Gettysburg as a small price to pay for the development of a really satisfactory and cheap compact car with decent pick-up and road-holding capability. It is to our credit that we survived the War and tempered our national fiber in the process, but human decency and the future security of our country demand that we look at the costs. What are some of the costs? Robert Penn Warren, The Legacy of the Civil War (New York: Random House, 1961), pp. 49–50.

A discussion fifty years ago of comparative economic history would have taken a broader view and would probably have been concerned very largely with exploring along the trails blazed by Max Weber and Marc Bloch. They were interested in many other aspects of economic history besides economic growth and I hope that similar broader interests will shortly show signs of reanimation. In spite of the present popularity of quantitative studies of changes in production, I hope some discussions at this meeting will examine comparative studies of forms of economic organization and the human qualities those structures reflected or generated. But my remarks here accord with the present preoccupation with that kind of economic history in which the all-important questions relate to the causes of economic growth. And I limit myself to one aspect only, the influence of governments.

The analysis of fluctuations longer than the nine-year business cycle has been somewhat confused by a failure to distinguish sharply and to relate three distinct phenomena: the forces set in motion by a leading sector in growth, stemming from the introduction and progressive diffusion of a new technology, and its deceleration; the forces set in motion by changes in the profitability of producing foodstuffs and raw materials, whether from the side of prices or technology, including their effects on investment in new territories and mines, on capital movements, interest rates, terms of trade, and domestic and international income distribution; and the forces set in motion (notably, in housing and urban infrastructure) by large waves of international or domestic migration or other forces changing the rate of family formation, housing demand, and the size of the working force. As the world economy unfolded after 1783, these three phenomena operated concurrently and related to each other in complex ways not easy to disentangle. Nevertheless, the substantial literature on long waves can be usefully clarified by distinguishing these phenomena more sharply than is sometimes done and then relating them to each other at particular times and places. The exercise of doing so has, I believe, some general implications for economic theory, for they are all aspects of the process of adjustment towards a moving equilibrium, never attained—a process for which we lack an adequate dynamic theory.

Why Did “Uncle Remus” exhort the post-bellum South to reduce its cotton-growing in favor of corn? His complaint was prompted in the immediate sense by the low cotton prices of the early 1890's, but such comments reflected a continuing discontent over the region's abandonment of self-sufficiency in foods after the Civil War. The ratio of cotton output to com was probably at an antebellum peak in 1860, but this ratio had been easily exceeded by 1880, as Table 1 indicates. In the leading cotton states, per capita corn production and the per capita stock of hogs were only about half of what they had been twenty years earlier. Coinciding as it did with a major era of stagnation in world cotton demand, this shift into cotton is of great importance for the subsequent economic development of the South. Despite its size and significance, the shift lacks a satisfactory explanation in the historical literature.

The successful demonstration of reaping machines by Obed Hussey and Cyrus McCormick in 1833 and 1834 inaugurated long series of events that eventually revolutionized the harvesting of small grains and grasses, drastically altering the lives and productivity of grain farmers. Given the ultimate success and widespread impact of the reaping machine, historians have long pondered why almost twenty years elapsed between the date when Obed Hussey sold his first machine in 1833 and the first wave of popular acceptance in the mid-1850's. Why did it take twenty years for a significant number of farmers to begin to exchange their cradles and scythes for reapers and mowers? What important economic and technological factors governed the initial diffusion of this invention? These are important historical questions, the answers to which can significantly influence our broader perceptions of the problem of technological diffusion.

A struggle for industrialization is one of the major themes of the recent history of not only Brazil, but virtually the whole of Latin America. The historical evolution of that struggle finds a common denominator in the structural similarities of the national economies of the region. Assessment of Brazil's experience in the post-1929 period should therefore yield insights into the problems that Latin America as a whole faced during the international upheaval spawned by the financial collapse of 1929. Such an appraisal may also provide empirical assistance for the elaboration of future comparative studies. The specific analytical focus of this article is the attitude of Getúlio Vargas' government toward industrialization and planning, a theme whose re-evaluation serves, furthermore, to do justice to the enigmatic and reluctant revolutionary who governed Brazil during critical years and depression and war.

To the public of the 1860's, federal loans and land grants to the pioneer Pacific railroads represented aid necessary to secure an economically and politically desirable technological feat; to the public of the 1870's they represented plunder. Scandal, hinted before the driving of the legendary Golden Spike in 1869, sporadically dominated the national political debates of the 1870's and the 1880's. The issue subsided by the 1900's, leaving only a residual popular impression of ample government aid to railroads. Among professional historians, the debate was revived in the mid-1940's by Robert S. Henry's revisionist article justifying the land grants. The ensuing discussion largely restored the original consensus that the land grants represented excessive subsidization.

The dominance of steam over sail was not finally established on long distance routes until the 1880's. When it came, steam's triumph was due primarily to the adoption of the high pressure compound engine, a marvel of its day and one of the major engineering achievements of the nineteenth century. Yet within two decades experiments were being undertaken with two forms of marine propulsion—the steam turbine and the marine diesel—which were to render the compound engine uncompetitive over a wide range of uses. An experimental steam turbine was built by Parsons in 1884 and an improved design powered the Turbinia in 1894. The heavyoil or diesel engine was first adapted for marine propulsion in 1902. In 1910 a Dutch tanker, the Vulcanus, was fitted with a 450 BHP engine and in 1911 Burmeister & Wain began constructing the first ocean going cargo liner, the Selandia. By the early twenties steam turbine and diesel were replacing conventional steam vessels. Above all, the inter-war years witnessed the success of the marine diesel (motorship) which was adopted, particularly by European shipowners, for the propulsion of a wide variety of ship types and sizes.

Douglass North and Robert Thomas recently proposed a model of the rise and fall of the manorial system. There is much to be admired in this work, which explores an unusually broad historical vision with great analytical acumen; but my purpose here is to examine some of its less compelling features. In section II, I consider the nature of feudalism, and develop empirical arguments against the wholly voluntaristic and non-exploitative interpretation proposed by North and Thomas. Section III examines their analysis of the “classic” manor, which I believe errs both in limiting the feasible contracts to forms of direct barter and in attributing the lowest transaction costs to labor dues even within that restricted set. Section IV reviews the proposed explanation of the later evolution of the manorial system, disputing both the continued use of the transaction costs model and the extensions of it that consider custom as an institutional barrier to efficiency, and population growth and inflation as the exogenous motors of change. Section V provides a brief concluding summary and evaluation.

“The town has that sign of prosperity of new buildings, which never deceives,” wrote Arthur Young of Nantes in 1788. In fact, on the eve of the Revolution, Nantes—the infamous ville des négriers (city of the slavers)—was one of the richest cities in France and one of the most important ports in the kingdom. For almost a century Nantes had been the capital of the French slave trade, and by 1789 local merchants were investing almost 10 million L (livres) annually in the trade. Yet, in spite of the general activity, Nantes traders were experiencing difficulties with their business. Although a few slavers managed to create vast fortunes almost overnight, problems in France, the colonies, and Africa meant that most traders had to struggle to obtain even modest profits, from slaving. During the last decade of the old regime investments in the slave trade were usually no more profitable than investments in offices or land, although the trade alone offered the possibility, however, remote, of quick, vast profits.

The textbook theory of tariffs, and their converse, the movement to freer trade, has more elements than we need for the nineteenth century, but also lacks some. In the usual comparative statics, a tariff may be said to have ten effects: on price, trade, production (the protective effect), consumption, revenue, terms of trade, internal income distribution, monopoly, employment and the balance of payments.

Did preparations for the Second World War account for the precipitous drop in the growth rate of Soviet industrial production from 10–12 percent per annum in the period 1928–1937 to only 2–3 percent per annum in the period 1937–1940? According to some who study the Soviet economy the answer is “yes.” This view has been succinctly expressed by Stanley Cohn: “After 1937, the rising spectre of Hitler forced the Soviet leadership to shift resources into armaments on a massive scale. As a result, the growth rate fell drastically to 3.6 percent per year between 1937 and 1940.” Such a sequence of events, however, has never been empirically demonstrated. The purpose of this paper is to investigate formally the validity of this explanation, via aggregate production functions, particularly of the CES (constant elasticity of substitution) variety, as well as to explore an alternative hypothesis, espoused, among others, by Naum Jasny, Alec Nove and Warren Nutter. This hypothesis stresses a domestic factor as the major contributor to the disruption in industrial production: namely, the impact of Stalin's terror in the form of chaos-producing political purges.

This paper should begin with a brief defense of its title. “Variety” and “unsuccessful” are doubtful if not dirty words to most economists and many economic historians. The “success stories” of rapid development in Western Europe, Russia and Japan have been the most frequent subject of this Journal's articles on non-American topics. And the discovery of uniformity in the past, rather than variety, is admittedly essential to the development economist's search for predictability in the future that has informed so much of the economic history written since the Second World War.

As a nation becomes economically developed, the pace of development usually tends to vary between sectors and between regions of the nation. Industrialization usually involves more rapid growth and technological change in some industries than others. Often regional specialization accompanies industrialization, and, in fact, can become an important determinant in the industrialization process. With differential growth among industries, and increasing regional specialization, the performance of the factor markets in mobilizing and reallocating capital and labor among industries and regions becomes extremely important.

The contract construction industry today represents one of the most important sectors of the American economy. Total new construction in recent decades has accounted for nearly one-half of the Gross Private Domestic Investment, while non-farm residential building alone accounts for nearly one-quarter of GPDI.

An issue of continuing interest in Canadian economic history is the lag between the formulation of policies directed toward settling the prairies and the appearance of any significant agricultural population. Proposals to develop the region preceded the union of the British North American colonies in 1867. By 1872 the first Homestead Act had been passed and a commitment made to construct a transcontinental railway linking the western provinces to Central Canada. Yet except for a brief speculative boom in the early 1880's, occasioned by the CPR reaching Winnipeg, the rate of settlement remained well below expectations. Homestead entries averaged under 3,000 from 1874 to 1896, and in many years there were nearly as many cancellations as new entries. In the same period adjacent American lands were filling up, in large part with emigrant Canadians. Settlement of the Dakotas, beginning in 1870 but depressed from 1873 to 1878, boomed from 1879 to 1886. Over the thirty years from 1870 to 1900 an estimated 120,000 Canadians chose the American prairies over the Canadian.