Feeding the Brethren: Grain Provisioning of Norwich Cathedral Priory, c. 1280–1370

This dissertation is a study of food supply in a late medieval English urban community. In late medieval England, between the thirteenth and sixteenth centuries, the vast majority of population, around 85 percent, lived and worked on land. At the same time, England underwent a significant process of urbanization. Urban centers emerged and expanded, participating in and contributing a great deal to economic and mercantile activities of the country. By the end of the thirteenth century, Norwich grew into the second largest city in England (after London), housing perhaps as many as 20,000 people.Footnote ¹ At the heart of the city stood Cathedral Priory, one of the wealthiest landlords and one of the most important food producers and distributors in the entire region. The grain provisioning of Norwich Cathedral Priory forms the core of this dissertation.

The importance of this topic is manifold. First, it shows how people were fed, what were different strategies of producers and distributors, and what were the channels, by which finalized food products reached their consumers. Second, it reveals the fragility of everyday life, which meant that the degree of success or failure of food supply varied from place to place and from season to season. Finally, it deals with the question of diet standards, composition and cultural preferences and choices of different social strata in late medieval England. These important aspects can be linked to larger sociocultural questions and problems, which are involved in ongoing scholarly debates.

Despite intensive forms of agrarian regime and fairly sophisticated management of landed estates, food provisioning was a quotidian problem, both in the countryside and in towns. The success of food supply was often subjected to the vagaries of endogenous and exogenous factors, and this was especially apparent in the fourteenth century. This was a long century of crises, which had a profound impact on all strata of the society. The most important ones were the Great Famine of 1314–1322, the Great Cattle Plague of 1319–1321 and the Black Death of 1348–1352.Footnote ² To make things worse, England was also involved in a violent war with Scotland (1296–1328) on the one hand, and with France (1337–1453) on the other. These crises spread chaos over the country and made food provisioning more difficult and challenging than ever before. One of the main objectives of the study, then, is to see how Norwich Priory authorities coped with these crises in order to ensure a steady food supply to their house.

The research is based entirely on unpublished archival material, deposited at various national and local archives and other repositories of the United Kingdom. The sources in question include over eight hundred manorial and monastic rolls, compiled in that period.

Chapter 1 is introductory in its character. It presents sources, problems methodologies, and objectives.

Chapter 2 looks at the extent of Norwich's rural hinterland, geography, and regional specializations of crops. It argues that the geography of grains in the Norwich region was determined by a combination of several factors. One factor was the distance between the point of production (demesne) and the point of destination (Norwich). A close analysis of the accounts reveal that the Von Thünen theory of geographic specialization and distribution of “grain zones” seems to have worked in the case of late-medieval Norfolk.Footnote ³ The second factor was the urban demand. The fact that the Norwich community demanded large quantities of barley ale explains why barley growing was the specialization of suburban manors. Finally, the type of soil was the third factor, accounting for the fact that sandy and acid soils around Norwich were unsuitable for wheat growing.

Chapter 3 deals with grain production on the landed estates of the priory in Norfolk. Here, I examine the agrarian production and productivity over the period. My analysis reveals that Norfolk obtained high wheat yields, compared to other regions in England, which agrees with Campbell's earlier findings.Footnote ⁴ This success depended, to a large degree, on careful management of manorial officials, who did their best to respond to various economic and ecological crises and shifts. To a large degree, this was a history of wheat's expansion at the expense of rye, reflecting demographic and price trends, as well as changes in living standards. After the Black Death of 1348–1351, there was a visible contraction of the legume sector, which gave way to fallow and grazing land. This reflects the post-pestilence (relative) shift from arable husbandry to pastoralism. The ratio of oats-to-barley remained largely unchanged after the pestilence.

Chapter 4 looks at annual acquisition, components, and disposal of grain supply on the countryside. About 80 percent of the total grain receipt came from annual harvest, while the remainder derived from tithes and purchases. In the pre-1350 era, about 25 percent of the annual receipt was invested in seeding; about 20 percent were sold at local markets; further 30 percent were dispatched to Norwich, for monastic consumption; additional 20 percent were distributed among local peasants; and the remaining 5 percent, or so, were fed to animals. After the Black Death, however, different patterns emerge. The proportion of sales accounted now for 35 per cent of the total grain receipt; the share of seed remained at its pre-pestilence levels; the share of grain dispatches declined to about 15 percent; additional 15 percent were allocated to the peasants; and the share of animal feed increased to further 10 percent. The distribution reflects wider demographic, economic, and agrarian shifts in the post-Black Death period.

Chapter 5 is concerned with transportation of grain from the countryside to the town. In the Middle Ages, large-scale transportation was a challenging task, because of inefficient road systems, relatively high transportation costs, slow moving speed, and potential road hazards. The chapter shows how the Cathedral officials attempted to withstand these challenges and maximize the profits and efficiency of grain transportation. First, the chapter establishes the size of the draught sector within the demesne, and calculate relative costs of its maintenance.Footnote ⁵ It then discusses the customary carting dues of the local peasants, which undoubtedly saved the Norwich authorities a great deal of money. The actual carting strategies are discussed in a later section of the chapter.

Chapter 6 studies the physical space within Norwich priory, in which the grain was stored and processed. These included a granary, water mill, stables, bakery and brewery, all controlled and operated by different monastic officials. The chapter argues that the topological planning of the cluster of five adjacent buildings, all located close to the canal leading to the backwater of the Wensum River, allowed the most effective cooperation between a dozen of monastic officials. These convenient locations, plus the reliance on horses and water as the power source for the mill, undoubtedly optimized the level of production, on the one hand, and decreased the risk and transportation (or, rather, moving) costs, on the other.

Chapter 7 deals with the relation between the priory population, its annual grain supply and demand. It establishes the size of the monastic community, and the quantity of grain supply that reached it on an annual basis. The latter came from two channels: the manors and local markets. The chapter argues that the share of grains that were dispatched and purchased changed from year to year, subject to price fluctuations. For instance, when prices were relatively low, the priory authorities made larger purchases, while the “free” grain from the manor was stored in the granary, in case of famine or food shortage in the future. On average, about 70 percent of wheat came in the form of annual receipt from the manors, while the remaining derived from either the market, or the leftovers from a previous year. As much as 80 percent of malt was carted from the demesnes; over 14 percent remained in the granary from a previous year and almost 6 percent came from the market. Only a marginal percentage of oats (less than 3 percent) came in form of manorial receipts. Over 70 percent was purchased and about 25 percent was left untouched from a previous year. Conversion of the grain into its approximate calorific and financial equivalent reveals that the supply must have normally exceeded the demand, an impressive achievement for the period of food shortage.

Chapter 8 deals with processing, consumption, and redistribution of grain products at Norwich priory. First, it examines different kinds of bread and ale produced by the priory bakers and brewers. Then, it estimates how many loaves of bread and gallons of ale was the annual grain supply capable of producing. Here, I attempt to convert the volume of grain into its approximate calorific equivalent. Finally, I estimate the contribution of cereal products to the daily monastic diet, on the one hand, and the overall calorific intake of an average Norwich monk, on the other. These estimates, crude as they can be, reveal that the total energy intake of an average monk exceeded his diet requirements by far. The estimates reveal that cereal products, namely bread and ale, contributed, on average, between 55 and 60 percent total calorific intake of a Norwich monk. The remainder came in form of fish (about 30 percent), meat, and dairy products (about 15 percent). Joining the grain accounts with the cellarer's rolls, we arrive at some astonishing figures, indicating that on a non-fast day an average monk was offered a daily plate worth of around 6,500 kilocalories. These figures, however, should be approached with care: it is obvious that the brethren were not capable of consuming such heroic proportions of food supply. As complimentary evidence from Westminster Abbey suggests, about 40 to 45 percent of the daily food allowance was passed over to laborers, servants, and paupers. Subtracting forty per cent from the allowance of c. 6,500 kilocalories, we arrive at some 3,900 kilocalories, which accords with Barbara Harvey's estimation for the Westminster monks.Footnote ⁶ This energy intake, though not as quite deadly as 6,500 kilocalories, is still large for passive male consumers, since it exceeds the suggested 3,000 kilocalories recommended for active men. This indicates that the Norwich brethren may have suffered from overweight, which agrees with recent archaeopathalogical studies suggesting that late medieval English monks tended towards obesity.Footnote ⁷ This confirms the stereotypical dichotomy between the “fat monk” and the “slim peasant.”

In Chapter 9, I discuss the charitable activities of Norwich priory, in terms of bread and ale distribution among the three groups of the poor and needy: the town hermits, castle prisoners and street paupers. Alms distribution, whether in coins or in food, was an important behavioral element of the medieval culture. As far as our sources reveal, the priory authorities practiced generous charity activities among the poor and the annual amount of grain allocated for that purpose was enough to feed perhaps as much as 150 paupers, on the daily basis. This phenomenon is examined in a larger context of late medieval charity. In particular, an analysis of the earliest surviving wills from Norwich (for the two periods: 1314–1339 and 1370–1383) reveals that bequests of bread to the paupers was a commonplace among both religious and lay testators.

The epilogue places the process of grain provisioning of the Norwich monks into wider environmental and socioeconomic contexts of late medieval England. It provides an overview and summarizes the main contributions of the thesis to the present state of research

To conclude, this dissertation is an attempt to follow each and every step within the process of food provisioning of a late medieval urban community in England. The process of conducting food from the field to the plate was a complex one, requiring careful planning. Food supply was especially challenging in the fourteenth century, a century of numerous ecological and socioeconomic shocks and crises. As surviving evidence shows, however, the authorities of Norwich Cathedral Priory and their manorial reeves managed to secure a steady supply of grain produce, notwithstanding these challenges. One aspect that this dissertation failed to deal with is the question: To what extent does the Norwich case reflect the general picture of those days? To answer that question, a comparative analysis of other religious communities, such as Canterbury Cathedral priory, Westminster Abbey, Durham Cathedral and Ely Cathedral is required. This is one direction I intend to take my dissertation to in the future.

PHILIP SLAVIN, Yale University

REFERENCES

• Campbell Bruce M. S. “Arable Productivity in Medieval England: Some Evidence from Norfolk.” The Journal of Economic History 43, no. 2 (1983): 379–404.

• ______.“Four Famines and a Pestilence: Harvest, Price, and Wage Variations in England, Thirteenth to Nineteenth Centuries.” In Agrarhistoria. På många sätt. 28 studier om människan och jorden. Edited by Britt Liljewall et al. Stockholm: KSLAB, 2009.

• Campbell, Bruce M. S., James. A. Galloway, Derek Keene, and Margaret Murphy. A Medieval Capital and Its Grain Supply: Agrarian Production and Distribution in the London Region, c. 1300. Belfast: Historical Geography Research Series 30, 1993.

• Harvey, Barbara. Living and Dying in England, 1100–1540: The Monastic Experience. Oxford: The Clarendon Press, 1993.

• Langdon, John. “The Economics of Horses and Oxen in Medieval England.” Agricultural History Review 30 (1982): 31–40.

• Newfield, Timothy P. “A Cattle Panzootic in Early-Fourteenth-Century Europe.” Agricultural History Review 57, no. 2 (2009): 155-90.

• Patrick, Philippa. “An Archaeology of Overindulgence.” Archaeological Review from Cambridge 20, no. 2 (2005): 98–117.

• Rutledge, Elizabeth. “Economic Life [in Medieval Norwich].” In Medieval Norwich, edited by Carole Rawcliffe and Richard Wilson. Hambledon and London: London, 2004.

Market Integration and Economic Development: A Comparative Study of India and Europe, 1700–1900

The idea that the reach of the market is associated with the level of economic development, and that the expansion of markets leads to economic growth, has made Adam Smith one of the best known economists of all time.Footnote ¹ It has also become one of the most popular explanations for economic development since Smith first asserted that connection more than 200 years ago. This concept of Smithian, or trade-led growth more generally, has not only thrived within academia, but also in popular economics, and in policymaking around the globe.Footnote ² Yet the empirical evidence on when and how markets became integrated, and on whether, when, and under what circumstances expanding markets promoted economic growth, is actually amazingly thin and often remains ambiguous. This is where this contribution comes in. It is fore and foremost an attempt to improve the empirical evidence that often prevents any sound assessment on these issues. Thus a substantial amount of new evidence has been collected for both Europe and India with the aim of assessing the efficiency of markets as well as the processes of market integration and globalization through time.

By doing so, this study contributes to another major debate in economics and economic history, namely the one on the economic development of the big Asian economies. Spurred in particular by the rapid growth of Asian economies in today's global economy, this debate—which came to be labeled the “Great Divergence” debate—centers on the comparative trajectories of economic development in Europe and Asia and on explanations thereof. In this respect, studying markets and economic integration helps to assess the accuracy of the contradictory claims made in the Great Divergence debate, and makes it possible to roughly gauge whether larger and more efficient markets may have played a central role for Europe's early economic takeoff.

The analysis is broadly divided up into two main parts—macro and micro story. Part I charts the big, or macro, picture of market integration in India and Europe. To begin with, chapter 2 provides an overview of the political and economic background of the time period, and it also maps out the transport infrastructure and its evolution in both India and Europe. This chapter therefore describes what are normally viewed as the most important determinants for efficient market structures and the processes of integration or disintegration of markets. Chapter 3 moves to the quantitative analysis and starts out with a discussion about how both the extent and the process of market integration can be measured. Thereafter, the database that has been compiled for this task is presented. By measuring co-movements and the speed of adjustment, it then tries to pin down and compare the levels of integration in India and Europe. Chapter 4 then looks at concomitants of market integration, investigating price volatilities and price convergence. At the end of this chapter, I then try to link this new evidence on market integration with the available evidence on economic development. By doing so, I aim at exploring whether the assertion that Smithian growth played an important role for the rise of Europe seems plausible or not in the light of the new evidence on economic integration.

For reasons of simplicity and space limitation, the results of all quantitative investigations are in this abstract summarized without using numbers—and they can for India be summarized as follows.

Prior to the mid-nineteenth century, the grain trade in India was essentially local, while more distant markets remained fragmented. This is not to say that no grain was traded over longer distances, but the extent was very limited, as the prices from cities all over India still exhibited various characteristics of isolated markets. First, annual price fluctuations were extremely high. Second, differences in price levels between markets were very pronounced and persisted until well into the nineteenth century. Third, apart from neighboring villages or cities, price series from different markets did not show co-movements at all. Fourth, where prices do show co-movement, adjustment processes to spatial price differentials were slow. It was in the second half of the nineteenth century that these backward structures got transformed rapidly, so that by the end of the century, price volatility had been massively reduced, prices across regions had converged, and prices for even very distant markets showed clear co-movements and adjustments to price disequilibria. Nevertheless, although it may be possible to speak of a national grain market at the turn of the twentieth century, the formation of an integrated market was still incomplete and showed distinct regional differences.

These findings correspond very well with the general history and the state and development of the transport system and they confirm more mainstream accounts of Indian market integration.Footnote ³ In view of the highly fragmented political landscape, the missing security and—centrally—the very poor state of the transport system before 1850, to find anything but a very limited degree of economic integration would be very astonishing indeed. In a country without a reasonably dense system of made roads that still relied mostly on energy-wasting transportation by pack animal or human carrier, the size of the market for grain must have been confined to short distances. Not only this early pattern but also the rapid changes observed in the second half of the nineteenth century are consistent with the qualitative accounts on integration as well as the broader historiography. Since for the period after 1860, abundant statistical information is available, and the developments during the later period have actually been rather uncontested. It is widely acknowledged that—driven by the rapid improvement of transport facilities, coupled with the process of political integration—commercial activity was fundamentally transformed in this period, leading to an unprecedented extent of trade in general, and of grain trade in particular. Previously highly fragmented markets became unified to such a degree that by the end of the century, India not only had a national market for grain, but was integrated in the world markets for both rice and wheat.

For Europe, a summary of the big picture on market integration looks very different indeed. Already in the early eighteenth century, grain was not only traded locally, but even regional and some interregional markets showed already clear signs of common influence, such as substantial co-movement of prices or a relatively small level of prices disparities. Long-distance trade, on the other hand, was still very limited, as the connection between prices remained very weak. From the mid-eighteenth century onwards, the gradual increase in the size of the market was gaining pace in Europe, so that by the turn of the nineteenth century, already markets that were several hundred kilometers away were closely connected to each other. An encompassing integration of the long-distance trade in grain, however, needed decades more of steady market expansion. By the mid-nineteenth century, there seems to have been a truly European grain market, even though the erection of new trade barriers in the face of an emerging global grain market restricted the trade flows in some parts of the continent. This gradual nature of the integration process goes clearly against the popular “Big Bang” theory arguing that the process of economic change in Europe was a very fast and drastic process that only started at the beginning of the nineteenth century.Footnote ⁴

As for the comparative debate, which is motivated by the Great Divergence discussion, a picture of distinct differences between India and Europe arises. First, the level of market integration was higher in Europe throughout the entire examination period. Moreover, this difference varied greatly over time. Europe started at a considerably higher level of market integration than India; the correlation results suggest that Europe's level at the beginning of the eighteenth century already surpassed the one India attained in the late eighteenth and early nineteenth centuries. By the latter time period, due to an early and steady expansion of European markets during the eighteenth century, the extent of trade and the efficiency of markets were radically different in these two regions. This early expansion of European markets compared to India continued through the first half of the nineteenth century and made the differences even more pronounced. India's economic integration only gathered pace in the second half of the nineteenth century, but then it proceeded at a very fast pace, so that in comparative terms, this was a period of catch-up for India. Nevertheless, the improvements were not quite far-reaching enough to close the gap; market integration remained higher in Europe still.

This investigation on market integration thus clearly rejects claims by “World Theorists,” Indian Nationalists, or the “California School” of “Asia” having a similar economic potency as Europe until the late eighteenth or early nineteenth century.Footnote ⁵ At least as far as India is concerned, all indicators point to lack of economic development until the mid-nineteenth century as measured by the efficiency and size of markets. It seems safe to conclude that by the late eighteenth century, India was already very different from developed parts of the world such as large parts of Europe or as some advanced parts in China in terms of market performance. In view of the political history and the state of the transport infrastructure of the subcontinent, such a divergence between Western Europe and India needs to be shifted at least back to the seventeenth century, if not more. Thus, the results of this study also go against any “Big Bang” theory in terms of comparative development. The bifurcation leading to the rise of the West did not only happen towards the end of the eighteenth or at the beginning of the nineteenth century, and it did surely not depend on a relatively sudden shift of relative European and Asian economic trajectories. Explanations of this divergence therefore have to shift away from coal and colonial exploitation and back to more traditional factors such as commercial organization, institutions, or technological progress.

All in all, the findings on the big picture of market integration—both in terms of the trajectories within India and Europe as well as with respect to the comparative pattern—are not only consistent with each other, but furthermore find broad support when looking at the general history of development as well as the mainstream accounts on market integration and the transport infrastructure. Thus, the present results clearly confirm market integration as a plausible source for economic development.

Part I of the thesis compares the “average” performance of two very big geographical entities, even though these two regions labeled “Europe” and “India” were by no means homogeneous entities. Thus to get a better understanding of how markets actually worked and to better map out differences between Europe and India, as well as differences within these territories, part II provides complementary, more “micro” comparisons. These “extensions and differentiations” to the big picture are twofold: First, chapter 5 expounds on differences within Europe, after which chapter 6 comes back to the differences between India and Europe, and presents a geographically much more narrow comparison of a subregion in India with a subregion in Europe.

The differentiation on the intra-European picture (chapter 5) concludes that physical geography shaped the patterns and extent of economic integration probably more than any other factor. As a consequence, it may be misleading to speak of a European “level” or “path” of market integration. Instead it would be more accurate to distinguish “two Europes,” lowland Europe and landlocked Europe. Their respective path of economic integration is one of some common features but plenty of distinctive differences. In lowland Europe, markets clearly extended to much bigger geographical areas before the arrival of steam transportation and the creation of extensive road networks, which can be explained by physical geography that had endowed lowland Europe with easier and cheaper transportation. Again, is seems that a sensible argument for Smithian growth can be made. Since lowland Europe had bigger markets in early modern Europe, one would predict that northwestern Europe should, ultimately for geographical reasons leading to more trade, enjoy higher economic growth than other regions.

Chapter 6 returns to the description of differences between India and Europe, and narrows the geographical scope by focusing on a comparison of particular subregions to get a more specific understanding of differences between Europe and India. This comparison controls for physical geography, which—based on the insight of chapter 5—is arguably the most important differentiation to make in the early modern trade environment. I thus compare in this chapter the region around Pune in the Western Ghats for India and Switzerland for Europe, as they share a similar natural endowment: they have no access to harbors and are not located in plains with easy and widespread access to navigable rivers. The outcome of this exercise is that despite their shared characteristics in terms of physical geography, the market environment in the two regions was completely different and the comparative picture is basically congruent with the key “macro” results of part I—which points out that these results seem to stand on rather safe ground.

The overall conclusion from this dissertation must be that it will not stop the concept of Smithian, or trade-led growth more generally from thriving. Larger and better functioning markets did indeed seem to have come hand in hand with economic progress.

ROMAN STUDER, London School of Economics

REFERENCES

• Findlay, R., and K. O'Rourke. “Commodity Market Integration, 1500–2000.” In Globalization in Historical Perspective, edited by Micheal Bordo, Allen Taylor, and Jeffrey Williamson, 113–64. Chicago: University of Chicago Press, 2003.

• Frank, Andre Gunder. ReORIENT: Global Economy in the Asian Age. Berkeley: University of California Press, 1998.

• Pomeranz, Kennneth. The Great Divergence: China, Europe, and the Making of the Modern World Economy. Princeton, NJ: Princeton University Press, 2000.

• Roy, Tirthankar. The Economic History of India, 1857–1947. Oxford: Oxford University Press, 2000.

• Sachs, Jeffrey. The End of Poverty. Paperback edition. London: Penguin Books, 2005.

• Smith, Adam. An Inquiry into the Nature and Causes of the Wealth of Nations. First published in 1776, edited by E. Cannan (New York: Random House, 1937).

Real Wages and Wage Inequality in China, 1858–1936

In recent years, there has been great interest among economists in China's economic development in the late nineteenth and early twentieth centuries. Prior to the 1840s China was a closed, largely agrarian economy. Under international pressure, China gradually opened its economy to foreign trade and, later, to foreign direct investment. By the early twentieth century, the coastal regions were industrializing and commercializing, and Chinese-made manufactures had begun to replace imports and even to be exported. Historians and economists have made substantial progress in sketching the broad outlines of this transformation. What has been less understood, however, is how these accelerating economic changes affected people's real income and standard of living.

This dissertation studies trends in real wages and wage inequality in China from 1858 to 1936. In the first two chapters of my dissertation, I present the first detailed real wage series by skill group that span this critical period in Chinese history. I construct nominal wage series from the records of employees in the China Maritime Customs service (hereafter CMC) for nearly fifty Chinese cities. I also construct group-specific cost of living indices from price data and household budget information contained in CMC trade statistics and surveys. The resulting nominal wage series and cost of living indices make it possible to estimate long-run trends in real wages and skill premia for three basic categories of Chinese workers: unskilled, skilled, and highly skilled.

Using these newly constructed real wage series, I find that skill premia rose rapidly during the first three decades of industrialization. After the 1920s the wage gap between skilled and unskilled labor began to decline, while the gap between highly skilled and unskilled labor continued to rise.

In this dissertation, I further explore the causes of the changing skill premia. I find evidence suggesting that technological advances initially increased the demand for skilled and highly skilled labor, while supply of skilled and highly skilled labor was extremely scarce. Therefore skill premia increased rapidly. The decline of the skill premium after the 1920s, I show, is very likely to be driven by two factors. First, educational progress increased the supply of skilled workers, thereby reducing the skilled wage. The second possible factor driving down the skill premium after the 1920s is the effect of the trade boom in China during the early twentieth century. I study the effects of trade boom on real wages and wage equality in the third chapter of my dissertation, which is a joint work with Kris James Mitchener.

My wage and price data are collected from CMC archives and publications. These archives are among the richest resources for the study of trade, economics, politics, society, international relations, and many other aspects of Chinese history from the 1850s to 1949. The CMC was probably the only bureaucratic organization in China that continued to operate uninterruptedly throughout the period from 1859 to 1949. The CMC's geographical reach grew from just fourteen stations during the 1860s to nearly fifty during the 1920s, covering both the coastal and inland cities. The most studied of CMC's publications are its trade statistics. These trade statistics record the quantities and the values of all commodities passing through each treaty port and therefore provide a precious source for the study of trade patterns and commodity prices in these cities. Besides the trade statistics and reports, there are 55,000 volumes of the CMC's original archives kept in the No. 2 Historical Archives of China, around half of which pertain to labor. They include surveys of local wages and standards of living, CMC wage scales, and most importantly, the Service Lists—that is, the individual personnel records of CMC employees from 1875 to 1936.

The complete wage data set contains 44,600 observations. Each observation records, for one particular year from 1858 to 1936, the wage received by a Chinese employee of the CMC, as well as biographical information about the employee, such as name, job title, hometown, current port, the year beginning the service, the year promoted to current position, and the year transferred to the current port. The CMC treaty “ports” were not just coastal or river cities: 12,700 data points came from inland cities, constituting 28.48 percent of the whole sample. There were more than 100 specific positions in the CMC system. My wage data set focuses on the 35 most common occupations, which I categorize into three groups: unskilled, skilled, and highly skilled labor according to the education requirements of these occupations.

Because the CMC was a large bureaucratic organization that was not able to adjust wage scales instantaneously, it is possible that temporarily the CMC wages might have been different from the market level. However, the CMC tried hard to set its wage scales to match the market equilibrium level, and the CMC's survey data also suggest that the wages paid by the CMC, local Chinese firms, and foreign firms did not statistically distinguish from each other.

In order to estimate the nominal wages of three skill groups, I use hedonic regressions to adjust for changes in the attributes of the sample over time, as well as to control for characteristics of the individual employees. The dependent variable is the log wage earned by a particular CMC employee, and the independent variables are characteristics of the employee under analysis. The estimation shows that the nominal wages of all three skill groups began to grow rapidly in the mid 1890s and that those of highly skilled workers rose fastest. In order to test the reliability of the estimated nominal wage indices, I conduct four types of robustness tests. These robust tests do not change the qualitative features of my results obtained.

Changes in the nominal wage series largely resulted from movements in the general price level. The widely cited Nankai Index, which is an unweighted general price index, shows a drastic inflation since the 1890s. Other than this index, there is no other country level, weighted cost of living indices available for the period I study. Furthermore, because the price changes usually affected all groups of workers in different ways, ideally I should deflate the nominal wages of three skill groups with group-specific cost of living indices in order to capture the group-specific inflation effects. This is completed in the second chapter of my dissertation.

In chapter 2, I develop a set of group-specific cost of living indices to measure real wages and wage inequality. These are the first set of nation-level cost of living indices that span a long period from 1864 to 1936. I collect the price data from the CMC trade statistics and use the CMC's family budge surveys to estimate the weights of these commodities consumed by these three groups. Twenty-eight consumer goods are selected into the consumption basket. These commodities are highly representative of the total household consumption because they are of high weights in the household budgets.

By applying these weights to the price series, I construct the indices of different groups as well as the general cost of living index. Two features are conspicuous from the constructed group-specific cost of living indices. First, the general trend of the cost of living indices of all three skill groups is the same. Second, the cost of living of the unskilled group rose significantly faster than that of the skilled and the highly skilled groups. This shows that low-income people suffered more from the inflation than high-income people. This was likely driven by a faster rise in food prices than the prices of manufacture goods. As low-income people spent a much larger share of their expenses on food, higher food prices hurt the poor much more.

I deflate the nominal wage series of the three skill groups by the respective cost of living indices to obtain the real wage series. Real wages for unskilled labor rose by about 20 percent in the deflation period between 1858 and 1890, fell back in the next decade, and then leveled off until experiencing significant growth in the 1930s. Skilled real wages rose much faster from 1858 to 1890, declined after 1890, but then resumed growth during the first two decades of the twentieth century. Although skilled real wages began to fall again around 1920, the drop was followed by another rise in the 1930s. Highly skilled real wages rose from 1858 to 1890 and fell back from 1890 to 1900, just like wages for the other two groups. But the highly skilled real wage increased continuously from around 1900 to 1930, followed by faster growth during the 1930s.

I divide skilled and highly skilled wages by unskilled wage and obtain two skill premia. The two skill premia rose by 15 percent from 1858 to 1880, leveled off after that, and began to grow rapidly again from 1900 to 1920. After 1920 the wage gap between the highly skilled and the unskilled continued to rise, but the gap between the skilled and the unskilled began to decline.

The real wage series for the three groups of Chinese workers suggest that the initial rise in the skill premia was driven by demand and supply of skilled and highly skilled workers. Demand for all three types of workers increased rapidly as a result of two waves of economic change: trade liberalization beginning in 1842 and rapid industrialization beginning in 1895. The large stock of unskilled labor was more than sufficient to meet any increase in demand, but, given China's traditional Confucian system of education, it was more difficult to find workers equipped with the knowledge and skills demanded by the modern sectors of the economy, including engineering and mechanics, reading and writing in native and foreign languages, bookkeeping and accounting, and business management. Therefore, the supply of skilled and highly skilled workers was limited, and the skill premia accordingly rose.

Beginning around 1920, however, the two skill premia began to diverge. While the wage gap between highly skilled and unskilled workers continued to rise, that between skilled and unskilled workers declined. A good candidate for this decline is changes in the educational system that occurred after 1905, changes that fundamentally transformed the traditional Confucian system of education. Trying to emulate Japan and Germany, the central government abolished the Imperial Civil Service Examinations in 1905 and began to establish a modern schooling system. The new system was composed of primary, secondary, and tertiary schools. Secondary schools, including general high schools and vocational schools, were the major suppliers of skilled workers. The subjects taught in high schools were tightly connected to the industrial demand, for example, reading and writing of Chinese and foreign languages, physics, chemistry, law, economics, and geography. As a result of the policy change, modern high schools began to grow rapidly in the early twentieth century. By 1936 the number of enrolled high school students had grown to 627,246, a level six times that of 1912. This number was small relative to the total population. It was also only about 8 percent of total employment in the modern sector. However, the modern sector employed many more unskilled than skilled workers, and the demand for skilled workers should have been only a small fraction of total employment. Furthermore, the estimated stock of high school graduates has not included those who attended vocational schools and who attended but did not high school. Taking all these factors into account, it is highly plausible that the rapid developments in high school substantially increased supply of skilled workers and drove down the skill premium.

Another possible factor that drove down the skill premium between skilled and unskilled labor was the trade boom. During the first three decades of the twentieth century, China experienced a tremendous growth in trade with the rest of the world. The nominal value of exports sextupled and imports rose roughly by the same amount. To examine the factor content of Chinese trade, we assemble new detailed estimates of exports and imports at the industry level from CMC trade publications. Since CMC trade statistics were primarily published at the port level, the units of measurement and currency sometimes varied across ports and over time. We therefore standardized the measurement and currency units and then aggregated the product-level data to the national level. After creating a database of quantities and values of all the traded commodities, we classified exports and imports into two broad groups, unskilled and skilled, based on economic activity and skill intensity. Using either wages or education as our measure of skill, we show that exports became more unskilled-intensive over the entire sample period. In contrast, the share of unskilled imports declines substantially. The data set further shows that the exogenous shock of World War I dramatically raised the price of Chinese exports, increased the demand for Chinese goods overseas, and increased the demand for unskilled workers producing these goods. These trends continued even after the war ended. We show that the timing of the rise in export prices is consistent with the observed decline in the skill premium in China. Although we cannot completely rule out alternative explanations, the evidence on factor content of trade and unit values suggests that the rapid expansion in exports from Chinese trade significantly altered the skill premium in China. It appears that once export prices began to rise dramatically during the second decade of the trade boom, wages of unskilled workers relative to skilled workers changed sufficiently to alter the slope of the skill premium. By the 1920s the skill premium had reversed course and declined by roughly 8 percent.

SE YAN, Peking University

Evening Schools and Child Labor in the United States, 1870–1910

Economists and historians have made considerable progress in understanding the expansion of public common schooling in the United States during the first half of the nineteenth century and the impact of legislation regarding child labor and compulsory school attendance in the late nineteenth and early twentieth centuries. However, the quantitative literature has neither documented nor analyzed a coincident feature of the educational movement—widespread efforts to enable children to combine work and schooling. Non-traditional forms of education—including evening schools and employer-operated schools, as well as part-time, half-time, and continuation schools—emerged in response to the new industrial environment to facilitate the education of working children.

Evening schools, in particular, were common throughout the United States, and enrollment in these schools continued to increase throughout the later part of the nineteenth century. By 1900 more than 845 evening schools operated in 145 cities and towns, with a total enrollment of at least 203,000 students.Footnote ¹ This represented a small fraction of total school enrollment at the national level, but the proportion was much higher in heavily industrial cities. In 1880, for example, in Holyoke, Lowell, and Salem, Massachusetts, evening school enrollment represented 15, 14, and 11 percent (respectively) of total public school enrollment. Evening schools accounted for 8 percent of enrollment in both New York City and Philadelphia, 12 percent of total enrollment in Newport, and 25 percent of enrollment in Paterson.Footnote ²

By examining this relatively neglected aspect of American education, this dissertation aims to fill an important gap in the literature regarding the economic history of education in the United States. Evening schools effectively lowered the opportunity cost of education for children from poor families since they could attend school without forgoing wages. The schools also encountered less political opposition from (and often found advocates among) poor parents, some teachers, and manufacturers who were unwilling to support compulsory school attendance legislation or enforcement. As such, the schools provide an excellent example of the flexibility and adaptability of American educational institutions around the turn of the century. This is of particular interest since the relative openness of educational opportunities in the United States helped facilitate economic and social mobility during a time of rapid economic change and helped position the United States for economic leadership in the “Human Capital Century.”Footnote ³

To examine the diffusion and impact of evening schools, this dissertation brings together information from the census, various Annual Reports of the U.S. Commissioner of Education, a number of states' school superintendent's reports, and factory inspection reports, survey data on working youths in New Jersey, and a variety of secondary sources. After a brief introductory chapter, I begin by using these sources to piece together the first detailed history of the expansion and evolution of public evening schools, which emerged to facilitate education for working youths and to aid in the assimilation of immigrants' children (chapter 2). The schools held great potential for helping such children acquire a basic level of education, obtain preparation for work or citizenship, or in some cases continue to higher education. The historical evidence gathered and synthesized in this chapter suggests that many youths seized these opportunities and benefited from the system of public evening instruction. The same aspect of evening instruction that held so much promise, namely youths' ability to combine work and schooling, also posed challenges to the success of the schools, however. Indeed, critics of evening schools often emphasized problems such as fatigue among evening school students and teachers, irregular attendance, and compressed instructional time.

Despite these challenges, evening school enrollment continued to increase throughout the late nineteenth century. The expansion was far from uniform across space, however. Some states and cities embraced this nontraditional form of education much more readily than others. Since the potential benefits of evening schools (as well as the problems mentioned above) were widely discussed among educators in the nineteenth century, one is left questioning why some large cities chose to open evening schools while others did not. The third chapter of this dissertation examines this question. The chapter begins by discussing the extent to which state legislation regarding evening school provision and child labor laws explain differences in evening school provision across states. I then turn to the more significant question of why cities within states differed in their provision of evening schools. Standard public choice models would suggest that cities with more educated and prosperous residents, greater resources available to the schools, and more homogeneous or stable populations should have been more willing to educate children in the community. Since these predictions need not hold, however, in the presence of externalities to education, I also consider whether—as educators claimed—the extent of child labor, the size of the immigrant population, and even the extent of overcrowding in the public (day) schools may have influenced cities' decisions regarding evening school provision.

To econometrically test these (often competing) hypotheses, I collected city-level data on the existence of evening schools, receipts of the public school system, and the spaces available for study from the Annual Reports of the U.S. Commissioner of Education for 1871, 1880, 1900, and 1910 and matched these administrative school data with city-level characteristics calculated from Integrated Public Use Microdata Series (IPUMS).Footnote ⁴ The quantitative evidence indicates that—consistent with a model of school provision where externalities to education are substantial—cities with high proportions of working youths, sizable immigrant populations, and overcrowded day schools were most likely to establish evening schools. Although the results square well with the historical narratives of evening school provision, some care must be taken in attaching a strong causal interpretation to the estimates. It is possible, for example, that the observed positive correlation between child labor and evening school provision reflects an endogenous response of child labor to the availability of evening schools, or the presence of unobservable city-level factors that influenced both child labor and evening school provision. Therefore, after constructing a “simulated instrument” for child labor that exploits differences in industry structure that should have affected the prevalence of work among children but not the provision of evening schools, I estimate two-stage least squares and maximum likelihood models of evening school provision that allow for endogeneity in child labor. The results suggest that exogenous differences in the proportion of children working do not appear to account for city-level differences in evening school provision. Rather, the political economy of evening school provision appears to have hinged more on school boards' responsiveness to plausibly exogenous differences in immigration levels and school overcrowding.

Raw evening school enrollment figures suggest that, especially in industrial cities, many youths took advantage of these educational opportunities when they were made available. Of course, it is unlikely that every evening school enrollee was a net addition to the city's public school rolls. There is evidence, for example, that evening schools also attracted students who had enrolled in the day schools at some point during the school year, but found themselves unable to attend regularly due to work obligations and family responsibilities. Moreover, since evening school access lowered the opportunity cost of working as well as the opportunity cost of attending school, some day school students may have entered the labor force when evening schools were made available. In such cases, evening school attendance was a substitute for day school attendance and the total number of students enrolled in the public schools that year would not change.Footnote ⁵

Access to evening schools should have improved (or at least not worsened) literacy outcomes for youths whose next best alternative to evening school attendance was full-time work or idleness. Evening schools may (or may not) have been a better fit for students that would have otherwise enrolled in day schools. Consequently, the overall effects of the schools on educational outcomes such as literacy, numeracy, and educational attainment are ambiguous and depend on the productivity of time spent in evening school and the extent to which evening school enrollment “crowded out” day school enrollment. The fourth chapter of this dissertation uses large samples of census data, matched with city-level information regarding the availability of evening schools, to examine the overall effect that evening school access had on literacy and school enrollment. Here, I also examine the possibility that evening schools had the unintended consequences of drawing day school students into the evening schools and/or into the labor market.

Although contemporary critics argued that such schools would be ineffective due to the compressed instructional time and exhausted students, the empirical results in this chapter suggest that the schools did effectively promote school attendance and improve literacy outcomes. In accordance with the stated goals of the schools, the results indicate that the benefits of evening schools were concentrated among working children and the children of immigrants. The magnitude of the estimates suggest that if evening schools had been made available in all cities in the sample, the gap in literacy rates between native-born children (not reporting an occupation) and working youths or the children of immigrants would have been nearly eliminated. The empirical evidence is also consistent, however, with evening school enrollment “crowding out” day school enrollment in some places. It also appears that the ease of combining work and school may have induced some youths, especially native-born girls, to enter the labor force. Taken together, the findings regarding literacy outcomes and school enrollment lend credibility to the claims of evening school proponents who argued that the schools improved educational access and were a better fit for working children and immigrants who could not maintain regular attendance at day schools.

The annual education reports, from which much of the historical information and data in the preceding chapters has been taken, provide broad characterizations of the students enrolled in evening schools during this period. For example, we know that the children of immigrants and other children from low-income families were more likely to attend. These descriptive accounts, however, provide little basis for assessing how individual-level characteristics (e.g., gender, nationality, occupation, or household composition) influenced the likelihood that a working child would take advantage of opportunities for evening instruction. Fortunately, a unique survey of working children in New Jersey offers further insight. In 1903 enumerators for the State of New Jersey surveyed 943 working children, collecting information about wages, working conditions (e.g., hours worked and lunch breaks), and personal and household characteristics (including nationality, occupation of father, etc.).Footnote ⁶ This survey also included detailed questions on the amount and type of schooling, ability to read, write, and do math, and—specifically—included retrospective questions on night school attendance. Because the federal Census of Population recorded school attendance without distinguishing between day and evening schools (except in years when enumerators were specifically instructed not to count evening school attendance), this survey of working children is especially valuable. Moreover, since both child labor and evening schools were prevalent in New Jersey, this state provides an excellent context to examine the evening school attendance of working children. After providing some contextual information regarding child labor and compulsory schooling legislation in New Jersey, the fifth chapter makes use of the survey data to characterize the factors influencing working youths' decisions about attending evening school. I then use the data to examine the impact of evening school attendance on working youths' literacy, numeracy, and educational attainment.

Although we may not be able to generalize the specific results from this survey to the overall population of working youths, the New Jersey data do provide useful insights that cannot be discerned from other sources. First, although evening schools lowered the opportunity cost of schooling for youths from relatively poor backgrounds and private returns to education appear to have been high in this period, evening schools did not induce the majority of working children to pursue part-time schooling. Rather, the immediate importance of children's contribution to household production (especially when the mother was absent) and to family income weighed heavily in children's work and schooling decisions. Attracting a larger share of working children into schooling apparently would have required going further, perhaps even compensating children or their parents for time spent in school. Second, the evidence suggests that patterns of law enforcement mattered—establishments subject to inspection were far more likely to have child workers pursuing schooling than those that were insulated from inspection. Finally, night school attendance was associated with improved educational outcomes, namely literacy, numeracy, and ability to write.

This dissertation's concluding chapter highlights the contemporary relevance of these findings. In recent decades, when faced with the challenges of promoting school attendance of poor children, developing countries and urban school districts in the United States have also implemented nontraditional education programs that organize school schedules around children's work. Like their nineteenth-century counterparts, contemporary critics of these education alternatives voice concerns about the quality of instruction and potential for unintended consequences of the schools. The results presented in this dissertation, however, suggest that evening schools may be an effective way to promote education, even in settings where compulsory schooling (or child labor) laws are found to be ineffective policy instruments or where overcrowded schools cannot accommodate growing population pressure.

LINDA K. CARTER, Baylor University

REFERENCES

• Carter, Susan B., Roger L. Ransom, and Richard Sutch. “The Historical Labor Statistics Project at the University of California.” Historical Methods 24, no. 2 (1991): 52–65.

• Fishlow, Albert. “The American Common School Revival: Fact or Fancy? In Industrialization in Two Systems: Essays in Honor of Alexander Gerschenkron, edited by Henry Rosovsky. New York: Wiley, 1966.

• Goldin, Claudia. “America's Graduation from High School: The Evolution and Spread of Secondary Schooling in the Twentieth Century.” The Journal of Economic History 58, no. 2 (1998): 345–74.

• ______. “The ‘Virtues’ of the Past: Education in the First Hundred Years of the New Republic.” NBER Working Paper No. 9958, Cambridge, MA, 2003.

• Kaestle, Carl F., and Maris A. Vinovskis. Education and Social Change in Nineteenth-Century Massachusetts. Cambridge: Cambridge University Press, 1980.

• Ruggles, Steven, et al. Integrated Public Use Microdata Series: Version 3.0 [Machine-readable database]. Minneapolis: Minnesota Population Center, 2004. Available at http://usa.ipums.org/usa/.

• United States Bureau of Education. Annual Report of the U.S. Commissioner of Education. Washington: GPO, 1880.

• ______. Annual Report of the U.S. Commissioner of Education. Washington: GPO, 1900.

Your Job is Your Credit: Creating a Market for Loans to Salaried Employees in New York City, 1885–1920

In the first decade of the twentieth century, a market in the personal debt of corporate and government employees was thriving in New York City and other major urban centers in the northeastern and midwestern United States. A set of shadowy entrepreneurs, colloquially known as “loan sharks,” offered short-term, high-rate advances that they called “salary loans.” Despite operating in violation of the law, primarily the prohibition against usury, the operations of these intermediaries had by 1912 reached an imposing scale. At least eighty-one such offices operated in Manhattan and Brooklyn alone, with millions of dollars in loans outstanding. Of these eighty-one offices, thirty-four belonged to interstate chains, the largest of which stretched over sixty-three cities in the United States and Canada.

By the mid-1920s these lenders had all but disappeared. However, a new type of intermediary, the industrial loan company, had emerged and was serving substantially the same customers. More commonly known as Morris Plan banks, after their chief progenitor, Arthur J. Morris, these firms accommodated borrowers through a complex set of transactions designed mainly to disguise their true interest charges, which were more than double the statutory maximum for similar transactions. As with many of the salary lenders, these businesses typically belonged to interstate chains. Unlike their less scrupulous forebears, however, they operated with the full blessing of public officials. At the close of 1926, fourteen such firms operated in New York City, with $46 million in loans outstanding. The largest among them, the Industrial Finance Corporation, oversaw offices in more than one hundred cities from Maine to California.

Focusing on New York City, my dissertation examines the emergence of these intermediaries, their business strategies, and the transitions between them.Footnote ¹ Through the use of court transcripts, incorporation certificates, records of municipal and state agencies, correspondence in the archives of a charitable agency that dealt with oppressed debtors, and, in the case of the Morris Plan companies, business records preserved in the Library of Congress, I have reconstructed the operations of these firms and individuals. I then applied theories from the economics of information, cooperative and non-cooperative games, and capital structure to understand their business models and explain their successes and failures. In completing this project, I hope to contribute to the literature on financial innovation, market creation, and institutional change. My story puts forward a neoinstitutionalist interpretation of the evolution of a market for loans to individuals who, rather than owning their own businesses, relied upon others for their incomes.

At the heart of every credit transaction is the problem of asymmetric information. Any pool of prospective borrowers will contain individuals of varying creditworthiness. Some will enjoy more stable incomes or practice more frugal spending habits, while others will be less fortunate or conscientious. But all have the incentive to represent themselves as good risks. Furthermore, once they receive their loans, debtors have an incentive to engage in behavior that increases the probability that they will be unable to repay, for they enjoy all of the benefits of spending the funds but share the cost of default with the lender. They can spend lavishly, exert less effort, or simply disappear. If confronted, they can falsely cite circumstances beyond their control that, if true, would render fruitless any attempt to collect. Finally, debtors may simply refuse to pay if they believe that their creditors will consider it not worth their while to pursue a claim. To minimize the effects of these incentives, lenders must expend resources to screen out bad risks, monitor debtors' behavior, audit assertions of misfortune, and enforce claims against the recalcitrant.Footnote ²

In the earliest days of the market for loans to salaried employees, lenders used collateral to elicit information from borrowers. More specifically, they required that applicants sign chattel mortgages on their furniture as a condition for their loans. In so doing, they sought to change borrowers' incentives in two ways. During the screening process, they hoped that risky borrowers would self-select out of the applicant pool because of the increased cost to them of default. They further believed that while the loan was active, the threat of dispossession would motivate debtors to monitor their own behavior in ways that increased the probability of repayment.Footnote ³ Pledges of furniture thus shifted some of the lenders' information costs upon borrowers. Under the right conditions, this procedure could be reasonably effective, but it had two distinct flaws that limited the supply of loanable funds. First, applicants' willingness to pledge personal property said little about their ability to pay. They were typically desperate and lacked financial sophistication, so they were likely to overestimate the funds they would have available in the future. It therefore required additional expertise to evaluate their creditworthiness, which contemporaries averred took years to accumulate.Footnote ⁴ Second, delegating appraisals was risky, for there were many ways that hired appraisers could falsify or obscure information to defraud their superiors. Attempts to monitor their decisions were time-consuming and often unreliable because appraisers' judgments relied upon “soft” information that was difficult to convey verbally or verify empirically. Because of these factors, chattel mortgage lenders tended to limit the size of their enterprises to the few close relatives they could trust. Firms remained small and growth sluggish.

In contrast, the bureaucratization of corporate and government hierarchies made information about applicants cheaper to gather and more relevant to the lending decision. More specifically, the reorganization of labor accomplished three tasks of interest to lenders. First, they connected prospective borrowers' incomes to well-capitalized entities whose financial condition was easy to determine and verify. Second, they standardized jobs so that particular titles were associated with particular ranges of pay. Third, they concentrated large numbers of workers within the same building, where they interacted regularly due to the nature of their white-collar occupations. Bureaucratization thus made the attributes of their employees more stable and transparent. In addition, it changed the type of information available to lenders. Applicants either worked where they said they did or not, and their incomes were inferable from their job titles. Such data were simpler, more objective, and more easily verifiable than that available to chattel mortgage lenders. They were also reducible to numbers or had a relatively predictable range or responses. Hence lenders could codify the process of evaluating applicants to the extent that they could safely delegate the task to trained subordinates.Footnote ⁵

For those who understood their implications, these characteristics prompted a transformation in the structure of their lending operations. The most successful salary lenders established large, bureaucratic hierarchies to administer chains of offices distributed across multiple cities. They employed teams of trained auditors to oversee these offices from centralized headquarters and teach their firms' methods to local office managers. At the local level, managers recorded their transactions in detail and aggregated those numbers into weekly or monthly reports. These strategies conferred three advantages. First, they created economies of scale. Once a firm had established its rules, auditing infrastructure, and training procedures, the marginal cost of establishing new offices was relatively low. Second, increased size and scope diversified their portfolios of loans over more transactions and across multiple cities and industries. Relying upon the law of large numbers, owners could calculate the prospects of these portfolios according to statistical regularities, and they could control their attributes from afar, simply by changing the rules. Third, remote supervision distanced the principals legally and geographically from the transactions they administered, thus protecting them from law enforcement. A few lenders went further and hired only female managers for their local offices, for they found women faced substantially lower penalties when convicted of usury.

But however successful they were at overhauling the screening and monitoring process, salary lenders faced challenges in collecting loans from delinquent debtors. While their formal opportunities to secure legal recourse or coerce debtors' employers appeared strong, in practice such avenues were uncertain, cumbersome, and risky. In recognition of this reality, salary lenders turned to persistent campaigns of extralegal intimidation and obfuscation to collect their debts. They harassed, and they humiliated. They exaggerated their advantages to debtors and employers. When rebuffed, they resorted to informal, personal sanctions aimed at disrupting the working relationships and private lives of debtors and their employers. Collections became costly wars of attrition in which salary lenders' main weapon was their quarry's fear and uncertainty about the probable costs, both pecuniary and personal, of resistance.Footnote ⁶ While relatively successful in the short run, these tactics rested on shaky foundations. For them to continue to succeed, several conditions had to persist. Borrowers had to remain ignorant of their rights, and employers could not realize the extent of the business among their workforces. Legal authorities had to be kept oblivious or complacent. Salary lenders maintained this fragile equilibrium by relying upon the private sanctions of shame and guilt, shaping their transactions to conform to favorable legal categories, and selectively paying off public officials. In these efforts, they benefited from the relative anonymity of urban life and the fragmentation of political institutions.

However, these conditions evaporated in the early 1910s, when the newly established Russell Sage Foundation instigated a mass movement focused on reforming the market. As an endowed foundation, the RSF was an institutional innovation designed to overcome the informational segmentation and institutional discoordination of its time. Its charter gave it free rein to investigate the causes of social problems, analyze potential solutions, and encourage efforts to implement solutions. Independent funding freed it from obligations to any persons, parties, or institutions and imbued it with an aura of neutrality, which gave it credibility with the press. Connections between members of its board of directors and corporate and government elites bestowed a trustworthy reputation on the organization and facilitated cooperation between the organizations they managed. Thus bypassing traditional channels of communication, the RSF mobilized a coalition of corporate, government, and reform elites against the salary lenders and, more important, maintained the pressure over the long run. As a result, most salary lender found it impossible to continue operations.Footnote ⁷

The RSF campaign and its outcome underscored a proposition that has received less attention from scholars of financial intermediation than the screening, monitoring, auditing, and enforcement activities of lenders. They suggest that to be successful over the long run, lenders must also establish credible commitments that constrain their own opportunistic behavior. Salary lenders earned their loan shark moniker by pursuing fraudulent debts and engaging in collection tactics far out of proportion to the underlying obligations. Such malfeasance ensured that they would rarely gain favorable treatment from civil courts and invited costly reprisals from corporate and government officials should they become public. (Under such circumstances, even a compromised politician had to make a show of prosecution, if only to maintain appearances.) For their parts, salary lenders could not alter this treatment simply by announcing a newfound appreciation for the benefits of responsible behavior. A more sustainable business model would require the establishment of costly and binding consequences for opportunism.Footnote ⁸

Arthur J. Morris addressed the problem of credible commitment through an organizational structure that resembled the designed contract enforcement institutions described by Avner Greif and others.Footnote ⁹ He and his associates established the Industrial Finance Corporation, a $1.5 million firm, in New York City and then recruited “representative men”—individuals with high profiles in their respective localities—from across the country to invest in affiliated firms in their cities, with the IFC taking a 25 to 49 percent share, enough to give them a stake in the performance of the affiliate but not enough to gain control. These men then made public representations about the responsibility of the parent firm to gain charters and other legislative concessions that would enable it to operate in compliance with the law. The entire scheme—a franchise, in essence—established commitments between the central firm, its branch affiliates, and their local publics. The local sponsors were pledging their reputations as security for the responsible operation of their branches, while the owners of the parent firm risked their investment in the viability of the national project.

Neither salary lenders nor industrial loan companies remain with us today. The former went out of business in the face of legal difficulties and superior competition, while the latter merged with commercial banks after the law diminished the differences between the two. Their stories, however, offer a case study in the challenges of establishing financial markets. Lending to salaried employees required not only the emergence of new sources of supply, but a reconfiguration in the relationships between consumers, the private sector, and the state. Many attempts occurred, but stability took two decades to establish.

MICHAEL EASTERLY, Financial Crisis Inquiry Commission

REFERENCES

• Carruthers, Bruce G., Timothy W. Guinnane, and Yoonseok Lee. “The Passage of the Uniform Small Loan Law.” Working Paper, 2005. Available at http://www-personal.umich.edu/~yoolee/research/CGL.pdf (accessed 22 Aug. 2008).

• Cowan, Robin, and Dominique Foray. “The Economics of Codification and the Diffusion of Knowledge.” Industrial and Corporate Change 6 (September 1997): 595–622.

• Greif, Avner. “Commitment, Coercion, and Markets: The Nature and Dynamics of Institutions Supporting Exchange.” In Handbook of New Institutional Economics, edited by Claude Menard and Mary Shirley, 727–86. Dordrecht: Springer, 2005.

• Guinnane, Timothy W. “Trust: A Concept Too Many.” Yale University Economic Growth Center Discussion Paper No. 907, 2005. Available at http://ssrn.com/abstract=680744 (accessed 26 Nov. 2007).

• ______. “What's So Bad About Pawnshops? Some Thoughts on Micro-Lending.” Yale University Economics Department Working Paper, 2001. Available at http://pantheon.yale.edu/~guinnane/pawnshops.PDF (accessed 22 Aug. 2008).

• Handy, Femida. “Reputation as Collateral: An Economic Analysis of the Role of Trustees in Nonprofits.” Nonprofit and Voluntary Sector Quarterly 24 (Winter 1995): 293–306.

• Klein, Benjamin. “The Economics of Franchise Contracts.” Journal of Corporate Finance 2 (October 1995): 9–37.

• Latour, Bruno. “Drawing Things Together.” In Representation in Scientific Practice, edited by Michael Lynch and Steve Woolgar, 19–68. Cambridge, MA: MIT Press, 1990.

• Leff, Arthur Allen. “Injury, Ignorance, and Spite: The Dynamics of Coercive Collection.” Yale Law Review 80 (November 1970): 1–46.

• Mann, Ronald J. “Strategy and Force in the Liquidation of Secured Debt.” Michigan Law Review 96 (November 1997): 159–244.

• ______. “Verification Institutions in Financing Transactions.” Georgetown Law Journal 87 (July 1999): 2225–72.

• Porter, Theodore M. “Making Things Quantitative.” Science in Context 7 (1994): 389–407.

• Radner, Roy. “Hierarchy: The Economics of Managing.” Journal of Economic Literature 30 (September 1992): 1382–1415.

• Robson, Keith. “Accounting Numbers as ‘Inscription’: Action at a Distance and the Development of Accounting.” Accounting, Organizations, and Society 17 (October 1992): 685–708.

• Schelling, Thomas C. The Strategy of Conflict. Cambridge, MA: Harvard University Press, 1960.

• Scott, Robert E. “Rethinking the Regulation of Coercive Creditor Remedies.” Columbia Law Review 89 (May 1989): 730–88.

• Stein, Jeremy C. “Information Production and Capital Allocation: Decentralized versus Hierarchical Firms.” Journal of Finance 57 (October 2002): 1891–1921.

Essays on Race and the Persistence of Economic Inequality

But the vision of “forty acres and a mule”—the righteous and reasonable ambition to become a landholder, which the nation had all but categorically promised the freedmen—was destined in most cases to bitter disappointment.-W.E.B. DuBoisFootnote ¹

In January of 1865 General William Tecumseh Sherman authorized freed slaves to establish forty acre farms along parts of the South Carolina and Georgia coasts.Footnote ² He later provisioned them with broken down military mules. Although rumors of “forty acres and a mule” quickly spread among the South's newly emancipated slaves, Sherman's order was soon revoked, and the land was largely restored to its Confederate owners.Footnote ³ Both Reconstruction era policymakers and modern scholars have argued that the large and persistent gap between black and white income and wealth could have been reduced or eliminated if each freed slave family had been allocated “forty acres and a mule” following the Civil War. Other scholars, however, have questioned this conventional wisdom that land alone would have altered the economic conditions of former slaves.Footnote ⁴ No previous quantitative investigation of these competing claims exists, primarily because researchers have thought there was little variation in policy toward freed slaves. Without a group of former slaves who were treated with access to free land, the construction of a counterfactual to analyze the economic impact of a postbellum land distribution policy is difficult.

In this dissertation, I develop an empirical strategy that exploits a plausibly exogenous variation in policies of the Cherokee Nation and the southern United States to identify the impact of free land on the economic outcomes of former slave families.Footnote ⁵ The Cherokee Nation, located in what is now the northeastern corner of Oklahoma, joined the Confederacy in 1861 and was forced during postwar negotiations to declare its former slaves, who were of African descent, citizens with “all the rights of native Cherokees.”Footnote ⁶ According to the laws of the nation, all citizens, including the freed slaves, were guaranteed the right to claim and improve any unused land in the Nation's public domain—be it forty acres or 400 acres.Footnote ⁷ The Cherokee freedmen, most of whom had initially entered into sharecropping contracts, stopped working the land of others.Footnote ⁸ By the next crop season, many of the former Cherokee slaves had established their own farms. This variation in the treatment of former slaves provides a compelling way to assess both the shorter- and longer-term effects of land distribution as a policy to ameliorate economic inequality.

To explore these effects, I compiled, encoded, and developed documentation for three data sets: a 60 percent sample of the 1880 Cherokee Census, a 100 percent sample of the 1860 Cherokee Nation slave schedules, and a sample of individuals linked from the 1880 Cherokee Census to the 1900 United States Census. I use these samples to assess how the availability of free land affected the Cherokee freedmen in 1880 and in 1900.

The first chapter provides a brief history of the Cherokee Nation, focusing primarily on its economic organization and the role of slave labor during the eighteenth and nineteenth centuries. This section serves to provide a context for the exploration of the Cherokee freedmen's conditions and economic status that occurs in the following chapters.

In chapter 2, I assess the impact of free land on former slaves in 1880. I begin by utilizing a variety of statistical and historical sources to verify that southern freedmen can serve as a proper comparison group for the Cherokee's former slaves. First, I provide evidence that the differences in land policy between the southern United States and the Cherokee Nation did not reflect differences in underlying attitudes towards freedmen. Instead, it reflected the Cherokee's lack of bargaining power during postwar negotiations with the United States. Second, I document similarities in the experiences of slaves in the Cherokee Nation and the South. The Cherokee Nation and the southern states had comparable slave laws, similar methods of organizing slave labor, and grew the same types of crops. The median slaveholding was five in both the Cherokee Nation and the South. Furthermore, I provide evidence that Cherokee and southern slaves entered freedom with comparable levels of human and physical capital. Next, I demonstrate similarities in the institutions, economic activities, social organization, and wartime experiences of the Cherokee Nation and the South. Additionally, available agricultural data on acreage and livestock ownership are comparable on a per capita basis. Finally, I show that the introduction of different land policies did not alter the composition of former slaves in either location.

Using my sample drawn from the 1880 Cherokee Nation Census and previously collected data from the southern United States in 1880, I then compare various measures of the economic status of Cherokee and southern freedmen along two dimensions.Footnote ⁹ First, I examine the absolute outcomes of each group. Second, I use a difference-in-difference specification to estimate the relative levels of racial inequality in the Cherokee Nation and the South. I find that the Cherokee freedmen were consistently and significantly better off than southern freedmen with respect to various measures of wealth and income. Just over two-thirds of all Cherokee black male heads of households owned a farm in 1880, which was approximately five times that rate in the South. Furthermore, the racial gap in land ownership rates was smaller in the Cherokee Nation than in the southern United States. Black farmers in the Cherokee Nation, on average, owned farms that were closer in size to those of non-black farmers and were more likely to undertake long-term capital investments in their land. They additionally had higher absolute levels of wealth and income than southern black farmers.Footnote ¹⁰ These advantages translated into significantly lower levels of racial inequality in the Cherokee Nation than in the South. The estimated difference in the racial wealth gap was substantial and ranged from 46 percent to 75 percent. For income, the estimated difference in the racial gap was between 20 to 56 percent. These results suggest that if Reconstruction era plans to provide the newly freed slaves with “forty acres and a mule” had been implemented, the level of American racial inequality could have been greatly diminished—at least in the short run.

In chapter 3, I extend my analysis to 1900 by using a sample of Cherokee freedmen linked from the 1880 Cherokee Census to the 1900 United States Census. I initially focus solely on the Cherokee Nation and construct measures of intra- and intergenerational occupational mobility. I find high degrees of occupational persistence and upward mobility for Cherokee freedmen males. Of male heads of households who were farmers in 1880 and survived until 1899, 90 percent remained farmers, and 80 percent of those who could transition to a higher occupational class did. Their children also displayed a high degree of intergenerational occupation persistence and upward mobility. 65.9 percent of all sons had the same occupation as their father, and a majority of the observed occupational mobility is upward. These high degrees of occupational persistence and upward mobility suggest that at least some of the beneficial effects of free land persisted until the beginning of the twentieth century.

I next combine my sample of linked Cherokee freedmen with the 1 Percent Public Use Microdata Sample of the 1900 United States Census with American Indian Oversample to test for a potential convergence in the outcomes of Cherokee and southern freedmen. I first compare the outcomes of Cherokee freedmen to blacks in the South and the Oklahoma and Indian Territories. Cherokee freedmen adults tend to have higher literacy rates, are more likely to own their own homes, and are more likely to be farmers than blacks in either the South or the territories. By incorporating whites and Cherokees into the analysis, I measure the levels of racial inequality in the Cherokee Nation and the South. As in the previous chapter, I find that the level of racial inequality is smaller in the Cherokee Nation than in the South. These results all suggest that the Cherokee freedmen's income and wealth advantages persisted until 1900. Additionally, Cherokee freedmen children have higher levels of human capital accumulation than black children in both the South and the territories. They are more likely to attend school, attend for a greater number of months, and have higher literacy rates. When controls for parents' economic status are included, the effect of being in the Cherokee Nation largely dissipates. This result is consistent with the Cherokee freedmen children's educational advantages deriving from their parents' higher income levels and not something inherent to the Cherokee Nation itself.

In chapter 4, I provide a description of two of the data sets I collected—the 1880 Cherokee Census and the sample of individuals linked to the 1900 United States Census.Footnote ¹¹ Detailed data collection descriptions, sample selection procedures, contents of the data set, summary statistics, and other relevant information are included.

The experiences of the Cherokee freedmen provide insight into the potential effects of distributing the fabled “forty acres and a mule” to former slaves. After receiving access to free land, the Cherokee freedmen were significantly better off than southern freedmen in both 1880 and 1900. Their large wealth and income advantages seem to leave little doubt that, for at least this one group of slaves, forty acres (if no mule) had a long-lasting and beneficial impact on their economic well-being. Not only were the Cherokee freedmen successfully able to maintain ownership of their land, but, within fifteen years of emancipation, they also were able to invest in their farms and thrive in the agricultural society. Their children's high rates of literacy and school attendance in 1900 indicate that their higher levels of wealth contributed not only to their economic well-being, but also that the Cherokee freedmen were able to pass their advantage onto their children. Their success certainly raises the possibility that racial inequality in the decades following the Civil War could have been reduced if different policies had been implemented following the Civil War.

MELINDA MILLER, U.S. Naval Academy

REFERENCES

• Bloom, Khaled J. “An American Tragedy of the Commons: Land and Labor in the Cherokee Nation, 1870–1900.” Agricultural History 76, no. 3 (2002): 497–523.

• Du Bois, W. E. Burghardt. “The Freedmen's Bureau.” Atlantic Monthly 87 (1901): 354–65.

• Engerman, Stanley L. Review of The Roots of Black Poverty: The Southern Plantation Economy after the Civil War by Jay R. Mandle, The Wheel of Servitude: Black Forced Labor after Slavery by Daniel A. Novak, Forty Acres and a Mule: The Freedmen's Bureau and Black Land Ownership by Claude F. Oubre, and Social Origins of the New South: Alabama, 1860-1885 by Jonathan M. Wiener. Journal of Social History 13, no. 3 (1980): 490–97.

• ______. “Economic Adjustments to Emancipation in the United States and British West Indies.” Journal of Interdisciplinary History 13, no. 2 (1982): 191–220.

• Foner, Eric. Reconstruction, America's Unfinished Revolution, 1863–1877. New York: Harper Collins, 1988.

• Higgs, Robert. Competition and Coercion: Blacks in the American Economy, 1865–1914. New York: Cambridge University Press, 1977.

• Ransom, Roger L. “Reconstructing Reconstruction: Options and Limitations to Federal Policies on Land Distribution in 1866/67.” Civil War History 51, no. 4 (2005): 364–77.

• Ruggles, Steven, et al. Integrated Public Use Microdata Series: Version 3.0 [Machine-readable database]. Minneapolis: Minnesota Population Center [producer and distributor], 2004.

• Sherman, William Tecumseh. Memoirs of General W. T. Sherman. Scituate, MA: Digital Scanning, Inc, 2000.

• Sutch, Richard, and Roger Ransom. Southern Agricultural Households in the United States, 1880 [Computer file]. Berkeley: University of California, Berkeley, Institute of Business and Economic Research and Center for Research in Management Science [producers], 1990. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2007-09-19.

• Woodman, H. D. “Sequel to Slavery: The New History Views the Postbellum South.” The Journal of Southern History 43, no. 4 (1977): 523–54.

• ______. “One Kind of Freedom After 20 Years.” Explorations in Economic History 38, no. 1 (2001): 48–57.

Comments on Slavin, Studer, and Yan

I regard it as a great honor to be asked to convene the dissertation session for what we Europeans like to call the un-American section. A great honor but also a daunting task to select three from the twelve dissertations received, and an even more difficult problem to select just one as the winner. The twelve dissertations covered a very wide range of topics, periods and countries, and also a wide range of methodologies.

Faced with such a difficult choice, I was forced to think about what it is that makes an outstanding dissertation in economic history. I have drawn inspiration from a quote by Joseph Schumpeter, who noted that: “There are such things as historical and theoretical temperaments. That is to say, there are types of minds that take delight in all the colors of historical processes and of individual cultural patterns. There are other types that prefer a neat theorem to everything else. We have use for both. But they were not made to appreciate one another.”Footnote ¹

It may be difficult to do both, but that is what is needed rather than merely narrative history without a clear economic framework or a thesis designed merely to impress economists without a clear appreciation of the historical context. I began my search for a thesis that would strike a balance between the rich colors of historical processes and the analytical elegance of a satisfying economic framework.

Comments on Carter, Easterly, and Miller

It was an honor to be asked by Rick Steckel to serve as the Nevins Prize convener. In keeping with custom, let me note three things before discussing the finalists. First, the non-finalist dissertations were also, uniformly, very well done. I received eight dissertations in total. Each provided a considerable contribution to scholarship. Forecasting off this small sample of graduate research and writing provides solid grounds for optimism regarding the future of economic history.

Second, a word about Allan Nevins. Nevins was a prolific historian of the United States. A recent biography lists 32 sole-authored books, six coauthored books, and 17 edited volumes.Footnote ¹ His sole contribution to this JOURNAL was a review in 1956 of Ralph and Muriel Hidy's history of Standard Oil of New Jersey, which he praised for its diligent use of sources but criticized for its lack of “literary touches.”Footnote ² Nevins cared about those touches because he made his living from writing more than most academic historians. Although he was most closely associated with the history department at Columbia University, he actually began his career as a journalist at the New York Evening Post, and later went on to develop American Heritage magazine. In his presidential address to the American Historical Association, Nevins proposed that “the historian should narrate the past in terms of living men and women seen as individuals, groups, or communities.” None of this year's Nevins Prize entries read as though its main subject were game theoretic models, econometric fireworks, or a newly discovered cache of manuscripts. Each work emphasized those once “living men and women” whom Nevins placed in the forefront.Footnote ³

Third, a few words about the standards I employed to winnow the entrants down to the finalists. Ideally, a dissertation would (1) approach a big problem, (2) with new evidence (3) that was clearly analyzed, and then (4) be written in a direct and engaging fashion. These criteria should encompass both history and economics dissertations. A dissertation could achieve, say, three out of these four points and do well, and perhaps asking for all four would seem to be a pipedream. But the three dissertations presented here essentially did just that. Their authors should be proud of their achievement, but not so proud as to slack off on their efforts to publish, lest our colleagues be deprived of the pleasure of reading their findings. I discuss each dissertation in alphabetical order.