CHAPTER 1
The Debate about Guilds
. . . [this ordinance is] for the good and profit of the craft and thecommonality of the people.
—Hosiers’ guild ordinance, Paris, c. 1268
. . . they have enacted such ordinances . . . and have conspired to maintain and defend them, contrary to the regulations made for the common good of the city.
—Judges describing the fishmongers’ guild, London, 1321
. . . [these privileges are] to bring utility and honour and piety not only to us, but also to the common weal.
—Hatters’ guild privileges, Middle Rhine towns, 1477
. . . [under cover of their fraternity they act] to augment their craft at the expense of the Republic.
—Complaint against the velvet-weavers’ guild, Toledo, 1562
. . . [the guild assembly and ordinances] are most important for thepublic good and utility.
—Silk-twisters’ guild ordinance, Toledo, 1627
. . . for private or peculiar profit is the chief foundation (tho’ it always goes under the notion of a general advantage) of all those restrictions and burdens imposed on the citizens by corporations or guilds.
—Pieter de la Court, complaint against the guilds of Holland, 1662
People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices.
—Adam Smith, observation on urban guilds, 1776
What kind of institution makes the entire economy work better, and what kind moves resources into the hands of special-interest groups, at the expense of everyone else? In the terms people used to describe pre-modern guilds, what serves “the common weal”, and what “ends in a conspiracy against the public”? This question is central to improving people’s lives in rich and poor economies alike. Many factors help economies flourish, but a growing body of research suggests that one key cause is institutions – “the rules of the game in a society . . . the humanly devised constraints that shape human interaction”.1 The origins of these “humanly devised constraints” often lie centuries in the past, and most theories of economic growth make assumptions about the historical institutions that set economies on an upward path. But although institutions clearly matter, what kind of institution should a poor economy aim for?
Every institution is multi-faceted and none can be boiled down to a single, all-important characteristic—as this book will argue. But one key feature is whether an institution is “generalized” or “particularized”.2 Generalized institutions are ones whose rules apply uniformly to everyone in society, regardless of that person’s identity or group membership. The rules of the game established by such institutions apply to any economic agent impartially. In principle, a generalized institution is equally accessible to all participants, and thus facilitates and constrains everyone’s transactions in the same way. An example of a generalized institution is a state in which a clear rule of law is established, or a competitive market with open entry. In the words of the Declaration of the Rights of Man, issued by France’s National Constituent Assembly in 1789, “laws should be clear, exact, and alike for all citizens”.3
Particularized institutions, in contrast, have rules that apply differentially to different people. The rules of the game differ according to whether a person has particular characteristics. These characteristics are not related to the transaction in question. Instead, they relate to identity: gender, religion, ethnicity, social stratum, parentage, ancestry, legal privileges, or group membership. The rules and entitlements of an occupational guild, for instance, applied only to persons with membership in that guild. Non-members of the guild were treated quite differently. Guild membership in turn depended—as we shall see in this book—on many non-occupational criteria. The rules of a guild might guarantee your property rights or enforce your contracts, but only because of your particular identity, privileges, and entitlements as a member of a particular subset of economic agents, defined according to transaction-unrelated criteria.
Generalized institutions, some might claim, are idealized constructs that cannot be applied to pre-modern economies. True, no economy is regulated by perfectly generalized or perfectly particularized institutions. Instead, the two types of institution appear in different combinations in each society. But there is no technical reason why generalized institutions were impossible in pre-modern economies. Empirically, some pre-modern economies had quite generalized institutions, just as some modern ones have quite particularized ones. The county of Champagne in the thirteenth century, for instance, did not have perfectly impartial governmental institutions or perfectly impersonal markets; but it did offer a much more generalized institutional framework than most other societies at the time (and for centuries afterwards). These generalized “rules of the game” attracted hundreds of merchants from many lands, making the Champagne fairs the fulcrum of long-distance trade in western Europe from c. 1180 to c. 1300.4 Similarly, neither Venice and Bruges in the fourteenth century, Antwerp in the fifteenth, Amsterdam in the sixteenth and seventeenth, nor London in the seventeenth and eighteenth had perfectly impartial governments or perfectly impersonal markets; but each of them in turn offered an institutional framework that moved away from delegating power to corporate groups to providing public institutions that guaranteed the property rights and enforced the contracts of all comers. These generalized institutions made each city in turn a successful centre of trade and industry which, for a time, outperformed other urban centres.5 Likewise, as we shall see in the course of this book, Flanders in the fifteenth century, the Northern Netherlands between 1560 and 1670, England between 1600 and 1800, and the guild-free jurisdictional enclaves inside the cities of eighteenth-century France, did not have perfectly impartial governments or perfectly impersonal markets. But they did offer an institutional framework that curbed the particularized privileges of guild masters, creating a much more level playing field for outsiders to enter, compete, and engage in disruptive innovation. As this book will show, every society in pre-modern Europe had a different combination of “rules of the game”, each offering different gradations between the particularized and the generalized.
The coexistence of particularized and generalized institutions in all economies raises the question of the relationship between the two. Do highly particularized institutions such as guilds, which enable closely knit groups to cooperate to serve their members’ collective interests, facilitate or hinder the operation of more generalized institutions such as markets or states, whose rules in principle constrain and facilitate economic activity by all members of society impartially? Could a guild serve, as the Paris hosiers claimed in their 1268 privileges, both “the good and profit of the craft” and “the commonality of the people”?6 How does the balance between particularized and generalized institutions affect economic growth and human well-being? Pre-modern European societies, like modern ones, were often characterized by market failure and state failure, and thus by poorly functioning generalized institutions. But they also suffered from poorly functioning particularized institutions, such as the guilds this book investigates. The question is how the relationship between these two types of institution played itself out for individuals and the wider economy.
Guilds are classic exemplars of particularized institutions. A guild is an association of people engaging in the same activities and wishing to pursue shared purposes. But, as the epigraphs to this chapter show, guilds declared that they served both the interests of their members and “the commonality of the people”, “the common weal”, “the public good”. Thus although guilds acknowledged that they were particularized institutions serving their own members, they also claimed to create generalized benefits for society as a whole. But as the other epigraphs to this chapter show, contemporaries took a darker view, arguing that guilds “conspired . . . contrary to the regulations made for the common good of the city”, acting “to augment their craft at the expense of the Republic”, and imposing “restrictions and burdens on the citizens”.
The interplay between the purposes guilds served for their members and the effects they had on the wider economy has been central to guilds ever since they first arose. It was crucial for guilds in securing their own survival. It was important for outsiders in challenging guilds’ special privileges. And it was a key issue for policy makers in deciding how much to enforce or constrain guild entitlements. It is also the central question of this book. In pre-modern Europe, how did the particularized institution of the guild affect the special interests of guild members and the general interests of the economy as a whole?
WHY IS THIS IMPORTANT?
Does this really matter? Establishing the facts is important for their own sake. But do these particular facts have wider implications? Why should we care about an institution that, as we shall see, has not existed anywhere in Europe since 1883?7 Does it really matter what we think about European guilds? Yes—as this book will show. Guilds are important for understanding wider economic and social questions: the sources of sustained economic growth, the relationship between market and non-market institutions, the benefits and costs of social capital, the economic effects of networks, the causes of social exclusion and inequality, the economics of discrimination, and the determinants of institutions themselves.
For one thing, guilds are important for understanding the historical sources of economic growth. The first transition to sustained economic growth relied on economic transformations in the pre-industrial period. During the eight centuries before European industrialization, guilds were central institutions setting the rules of the game for economic activity. Understanding how economies first achieved sustained economic growth requires analyzing how their institutions worked—for good or ill.
Guilds also shed light on how non-market institutions facilitate and constrain markets. Every market, in rich and poor economies alike, suffers from failures—situations in which a market fails to allocate resources efficiently because of public goods, information asymmetries, externalities, principal-agent problems, or lack of competition. Since markets are necessary for economies to grow, institutions that ameliorate or exacerbate market failure are central to understanding growth. Guilds, as we shall see in this book, affected markets in many ways: through entry barriers, price and supply distortions, gender discrimination, product quality, labour skills, and innovation. By changing the working of markets to serve their own members, they inevitably changed markets for everyone else. The effect of a particularized institution such as a guild on a generalized institution such as a market is central to understanding economic efficiency, distribution, and long-term growth.
Third, guilds are exemplars of institutions that generate “social capital”—the stocks of shared norms, information, sanctions, and collective action that accumulate inside closely knit groups.8 Social capital is supposed to enhance trust in ways that make markets and governments work better. Robert Putnam, for instance, thought the social capital created by northern Italy’s medieval guild tradition was a major determinant of its modern economic success compared to the Italian south; this led him to conclude that social capital generally fosters dynamic economies and well-functioning polities.9 Surveys of social capital and economic development commonly refer to guilds as networks whose social capital aided European growth in the past and hold positive lessons for poor economies today.10 If this is true, then at least some particularized institutions might benefit not just special-interest groups but everyone in society. Social capital can be bad as well as good, however. To benefit its members, a closely knit group may use its shared norms, information, and sanctions to organize collective action that corrupts the state and distorts the market. Guilds provided institutional mechanisms for individuals to organize collectively to monitor and influence governments and markets, and this made it possible for them to curb abuses and failures in these institutions. But as we shall see, guilds also provided mechanisms to organize collusive action in markets in collaboration with governments, offering political elites favours in return for market privileges, even when this harmed the rest of society. Guilds shed light on the interactions between closely knit special-interest groups and political institutions in the run-up to industrialization and sustained economic growth.
The social capital guilds generated is closely related to a fourth key issue. Guilds were social networks—indeed, arguably the most widespread and long-lasting examples of such networks. Economists increasingly recognize that economies consist not just of individuals but of wider networks in which individuals are linked with each other in more or less intense clusters. Such networks can generate positive externalities by facilitating communication, diffusion, and cooperation that would not take place if all individuals transacted independently. But networks can also generate negative externalities via conspiracy, groupthink, and corruption. If network links can amplify good decisions by individuals, they can also amplify bad ones. Guilds had two features which made them particularly effective networks: they were closely knit, so everyone knew who was a member and who was not; and they typically generated, and often mandated, multi-stranded relationships among their members. A guild, as we shall see shortly, often favoured shared religious observance, cultural expression, sociability, and even intermarriage. Moreover, closure and multiplexity enabled guilds to generate both positive and negative externalities. This makes the guild an excellent context for exploring how social networks affect economies over a period of centuries.
Guilds also cast light on an issue of urgent concern: the determinants of inequality. A guild typically claimed to ensure equality among all its members —“the utility of both rich and poor”, as the Middle Rhine bakers’ guilds put it in 1436.11 As we shall see in Chapter 4, guilds often imposed elaborate restrictions on competition, which they justified in terms of internal equality. However, guilds also erected elaborate entry barriers and labour market regulations which reduced the opportunities and earnings of wide swathes of society, disproportionately afflicting the poorest and most marginal groups. The result was increased inequality between the small group of guild members and the large population of outsiders. This raises a key question. What sort of institution exacerbates inequality: a particularized institution such as a guild, or a generalized institution such as a market or a state?
Guilds can also shed light on economic discrimination. Economists are puzzled by discrimination. They cannot agree even on whether it is consistent with economic rationality, let alone what causes it and what policies might address it. Guilds, as we shall see, discriminated against women, poor men, Jews, Slavs, gypsies, migrants, people they defined as “dishonourable” or “untouchable”, and members of minority ethnic, linguistic, and religious groups. Did this merely reflect underlying cultural values to which institutions were irrelevant? Did it serve economic efficiency since outsiders were more likely to violate trust? Or was inefficient discrimination facilitated by particular institutions? Guilds, because they engaged in systematic discrimination over a period of centuries, offer a laboratory for exploring rival theories of discrimination. This in turn has policy implications, an important issue given how discrimination affects efficiency and equity in most economies.
Finally, guilds shed light on a very basic question: the determinants of institutions themselves. A guild was an institution that directly affected economic activity not just in the occupation it claimed as its own, but in all the factor and product markets its members used. Although guilds were usually based in towns, they affected the rural economy by constraining peasant crafts and services, shaping work opportunities for country people, and regulating markets in rural raw materials: grain for bakers and brewers, meat for butchers, wool and flax for weavers, leather for tanners and shoemakers, wood for builders. Beyond the purely economic realm, guilds organized sociability, religion, cultural expression, migration, marriage, and death. Guilds thus affected, directly or indirectly, nearly every facet of economy and society. This makes it even more important to understand what caused them to arise, survive, and decline. The literature on economic institutions has long debated whether institutions are just epiphenomena of exogenous natural and geographical factors, are efficient solutions to economic problems, are expressions of cultural beliefs and values, or result from political conflicts over distribution.12 Guilds existed in so many societies over so many centuries that they offer a rich context for testing theories about why institutions exist at all, and thus about what underlies their effects on economic growth.
A BRIEF HISTORY OF GUILDS
Guilds have been observed for thousands of years in many economies: ancient Egypt, Greece, and Rome; medieval and early modern India, Japan, China, Persia, Byzantium, and Europe; and nineteenth-century Latin America and the Ottoman Empire.13 But although guilds have existed for millennia in economies across the world, the analysis of guilds as economic institutions is based largely on Europe between about 1000 and about 1880. This is partly because evidence on guilds is richest there, as shown by the sparse research on non-European guilds.14 Partly, too, it is because guilds showed interesting variation across pre-industrial Europe, gradually weakening after 1500 in some societies but surviving long past 1800 in others. Above all, the focus on European guilds arises from the fact that Europe is where sustained economic growth first arose, raising obvious questions about the relationship between guilds and growth. For these reasons, this book focuses on guilds in Europe between the Middle Ages and the Industrial Revolution.
Guilds existed in European antiquity in the cities of ancient Greece and across the ancient Roman empire, not just in Italy but also in what is now France, Spain, and German-speaking central Europe. They even surfaced occasionally during the so-called Dark Ages (c. 400–c. 1000), although with tantalizing rarity and meagre detail. European guilds came definitively back into view with the resurgence of trade and industry, together with public record-keeping, after about 1000, and they became virtually universal across Europe in the thirteenth and fourteenth centuries.15 After 1500, they gradually declined in some places while becoming more entrenched in others. The last guilds in Europe were not abolished until 1883.16
Guilds were referred to using a variety of terms. The many Latin documents used words handed down from the occupational associations of Roman antiquity, such as collegium, ministerium, universitas, and officium, but also craft-related designations such as magisterium and artificium. The proliferating records in vernacular languages introduced a whole array of new terms. The English called them gilds, crafts, livery companies, brotherhoods, fraternities, and mysteries. The French spoke of métiers, ghildes, corps des métiers, corps jurés, communautés, and corporations. German terms included Amt, Bruderschaft, Gilde, Handwerk, Innung, Zunft, and Zeche—from which also the Czech cechu, Hungarian céh, and Polish cech. In Italian, guilds were arti, gremi, corporazioni; in Spanish, gremios, cofradías, consulados. Other European languages introduced their own terms: the Swedish skrå, Dutch ambacht, Romanian breaslă. Even within the same linguistic area, guilds commonly had multiple designations, each with a slightly different shade of meaning, reflecting differences in their origins, legal status, occupational coverage (craftsmen, merchants, or both), and sometimes even inceptive purposes (occupation, religion, collecting money, laying down rules, getting together to go drinking).
Even more fundamentally, behind the façade of the guild lay a wide array of variegated institutional mechanisms. One aim of this book is to make sense of this variety and see what it meant for how the economy worked. A first key distinction is between “merchant guilds” and “craft guilds”. Merchant guilds were organizations of wholesale traders. Their members specialized in selling merchandise mainly to industrial, commercial, institutional, or other professional business users, rather than to ordinary consumers. Merchant guilds are analytically distinguished from other occupational guilds because of the distinctive challenges of wholesale trading: the gap in space and time between delivery and payment, the need to deal with multiple political regimes, the lack of information about alien markets, the direct confrontation with price and supply shocks. So analysis of merchant guilds has focused mainly on how they affected commercial security, contract enforcement, principal-agent relations, information transmission, and price volatility. These issues are important, so much so that I explored them systematically in Institutions and European Trade (2011). In the present book, therefore, I do not analyze merchant guilds except where they impinged on other occupational guilds.
These “other” occupational guilds are usually called “craft guilds”. This shorthand term is popular but imprecise.17 For our purposes, the main snag is that it leaves out of account the many occupational guilds formed outside merchant trading, but also outside traditional crafts. Almost all urban crafts were guilded, certainly, but so too were many non-craft occupations. The service sector was full of guilds, and throughout Europe we find guilds of retailers, carters, porters, boatmen, painters, sculptors, musicians, physicians, surgeons, public-bath-operators, and chimney-sweeps. Many places also had guilds of primary-sector producers, including farmers, agricultural labourers, gardeners, wine-growers, shepherds, miners, and fishermen. In the Northern Netherlands at the end of the eighteenth century, just 62.5 per cent of guilds were in crafts, while 15 per cent were in commercial occupations (mainly retailing), and 20 per cent in transportation and other services.18 Despite their many individual differences, practitioners of these crafts and “other” guilded occupations shared one feature that distinguished them from wholesale merchants: they produced goods and services that were destined for (and often directly sold to) consumers. Both for this reason and because the term is established in the literature, this book uses “craft guilds” to refer to all guilds other than those of long-distance wholesale merchants.
Of course, merchant guilds and craft guilds also shared many features, impinged on each other, and sometimes wholly coalesced. With the growth of regional and international trade during the medieval and early modern Commercial Revolutions, urban craftsmen and rural cottage workers increasingly expanded beyond producing directly for local consumers to selling to wholesale merchants who exported their goods to consumer markets beyond the locality—what has been called “proto-industry”.19 Some societies, especially Italy, had “sectoral” guilds that combined the wholesale merchants, retail sellers, and craftsmen of a particular branch of industry in a single, overarching organization, encompassing everyone who worked in the city’s silk or wool sector, for instance. So the distinction between merchant and craft guilds should not be drawn too sharply, and this book does not do so. Analytically, this book focuses on those issues that are regarded as most salient to consumer-oriented occupations: guaranteeing the quality of wares and services, ensuring and certifying human capital investment, and inventing, adopting, and diffusing new products and techniques. But this book also examines several analytical issues which craft guilds shared with merchant guilds: their relationships with the political authorities, their entry barriers, and their market manipulations.
Merchant and craft guilds also differed in their historical trajectory. Local guilds of wholesale merchants appeared (or re-appeared) in most European societies from the early eleventh century onwards. A bit later, as long-distance trade expanded during the medieval Commercial Revolution, some local merchant guilds formed branches abroad as alien merchant guilds or “merchant communities” in foreign trading centres. Sometimes the merchant guilds of a group of towns formed a long-distance trading association, a guild of guilds called a universitas or a hansa. The most famous was the German Hansa, which by around 1300 encompassed merchant guilds from a core group of 70 north German, Dutch, and Baltic cities, and a penumbra of about 100 smaller towns. After c. 1500, hansas and guilds of alien merchants broke down in some European societies, were replaced in others by regulated, chartered, or proto-industrial “companies”, but survived in still others in something close to their original form. Spain and Portugal even exported their merchant guilds overseas, establishing powerful consulados which survived in Latin America into the nineteenth century.20
Craft guilds—those formed by craftsmen and other non-wholesale-oriented occupations—followed a different trajectory. Craft guilds reappear in the written sources somewhat later than merchant guilds, typically from around 1100 onwards.21 The time they emerged (or re-emerged) varied greatly across Europe, and the dates are often confused by the accidents of document survival. But by the thirteenth century, guilds of craftsmen and other consumer-oriented occupations were to be found across much of Europe. For the next half millennium or more, to practise most craft or service occupations in most European towns, you had to get a license from the relevant guild.
This is not to say that the guild landscape was homogeneous across European societies or unchanging over time between 1100 and 1850. For one thing, the sheer number of guilds in different towns varied hugely. Table 1.1 illustrates the broad spectrum found in different European cities. Some major cities had numerous guilds: Paris had 100 in 1270; London 92 in 1503; Madrid 113 in 1659; Rome 101 in 1708; and Vienna 150 in 1820. But other large cities had very few: Florence, one of the largest cities in Europe, had only 21 guilds in 1300; Augsburg in 1548 had only 17; Amsterdam in 1551 only 25. For a city, having a large population did not necessarily mean having a large number of guilds.
TABLE 1.1: Guild and Population Numbers, Various European Towns, Twelfth to Nineteenth Century |
|||||
Country |
Town |
Date |
No. Guilds |
Population |
Inhabitants per Guild |
Austria |
Vienna |
1470 |
45 |
41,000 |
911 |
Austria |
Vienna |
1820 |
150 |
294,000 |
1,960 |
Bulgaria |
Sofia |
early 19th century |
63 |
11,694 |
186 |
England |
Exeter |
1586 |
14 |
13,300 |
950 |
England |
London |
1179-80 |
21 |
32,900 |
1,567 |
England |
London |
1503/40 |
92 |
50,000 |
543 |
England |
York |
1415 |
57 |
12,000 |
211 |
England |
York |
1600–88 |
55 |
10,000 |
182 |
France |
Bordeaux |
1762 |
49 |
80,000 |
1,633 |
France |
Dijon |
18th century |
80 |
22,000 |
275 |
France |
Lille |
pre-1791 |
50 |
66,000 |
1,320 |
France |
Lyon |
1789 |
72 |
143,000 |
1,986 |
France |
Montpellier |
12th century |
72 |
6,750 |
94 |
France |
Paris |
1261–71 |
100 |
160,000 |
1,600 |
France |
Paris |
1766 |
133 |
600,000 |
4,511 |
France |
Rouen |
1775 |
112 |
74,000 |
661 |
Germany |
Aachen |
c. 1798 |
27 |
23,000 |
852 |
Germany |
Augsburg |
1368 |
13 |
19,300 |
1,485 |
Germany |
Augsburg |
1548 |
17 |
30,000 |
1,765 |
Germany |
Cologne |
15th century |
45 |
31,000 |
689 |
Germany |
Frankfurt a. M. |
1570 |
26 |
20,000 |
769 |
Germany |
Frankfurt a. M. |
1631 |
34 |
20,000 |
588 |
Germany |
Fulda |
1780 |
21 |
8,500 |
405 |
Germany |
Vörden |
c. 1800 |
2 |
500 |
250 |
Italy |
Florence |
1218–1330 |
21 |
100,000 |
4,762 |
Italy |
Milan |
1400 |
13 |
100,000 |
7,692 |
Italy |
Milan |
1627 |
44 |
120,000 |
2,727 |
Italy |
Rome |
c. 1600 |
71 |
100,000 |
1,408 |
Italy |
Rome |
1708 |
101 |
135,000 |
1,337 |
Italy |
Venice |
1268 |
16 |
110,000 |
6,875 |
Italy |
Venice |
17th century |
100 |
158,772 |
1,588 |
N. Netherlands |
Amsterdam |
1551 |
25 |
30,000 |
1,200 |
N. Netherlands |
Amsterdam |
1798 |
45 |
200,000 |
4,444 |
N. Netherlands |
Dordrecht |
1400 |
28 |
10,000 |
357 |
N. Netherlands |
Dordrecht |
1798 |
42 |
18,000 |
429 |
Scotland |
Edinburgh |
16th–17th century |
14 |
36,000 |
2,571 |
S. Netherlands |
Antwerp |
late 18th century |
60 |
50,700 |
845 |
S. Netherlands |
Bruges |
1477 |
54 |
40,000 |
741 |
S. Netherlands |
Bruges |
late 18th century |
74 |
30,846 |
417 |
S. Netherlands |
Leuven |
1360 |
45 |
40,000 |
889 |
Spain |
Barcelona |
1395 |
45 |
20,000 |
444 |
Spain |
Barcelona |
late 16th century |
75 |
32,000 |
427 |
Spain |
Madrid |
1659 |
113 |
126,500 |
1,119 |
Spain |
Seville |
1527 |
66 |
128,000 |
1,939 |
Spain |
Toledo |
1625 |
65 |
20,000 |
308 |
Spain |
Valencia |
end 14th century |
24 |
40,000 |
1,667 |
Spain |
Valencia |
1523 |
49 |
80,000 |
1,633 |
Spain |
Zaragoza |
1787 |
75 |
42,000 |
560 |
Sweden |
Gävle |
1750 |
5 |
2,983 |
597 |
Sweden |
Gävle |
1790 |
20 |
3,000 |
150 |
Sweden |
Karlshamn |
1803 |
15 |
2,142 |
143 |
Sweden |
Linköping |
1661 |
2 |
2,000 |
1,000 |
Sweden |
Linköping |
1840 |
24 |
5,000 |
208 |
Sweden |
Stockholm |
c. 1500 |
13 |
6,000 |
462 |
Sweden |
Stockholm |
c. 1700 |
20 |
52,500 |
2,625 |
Source: Quantitative guilds database (see text). |
|||||
Some guilds had only a handful of members. In seventeenth-century Paris, with nearly half a million inhabitants, the metal-engravers’ guild permitted a maximum of 20 masters, the clockmakers just 72.22 In Vienna, the largest city in central Europe in 1800, with 230,000 inhabitants, the pastry-bakers’ guild was limited to 14 members and the chimney-sweeps to 18.23 Other guilds did not have a formal upper limit, but nonetheless restricted entry via a mandatory career track of apprenticeship, journeymanship, and mastership with strict conditions for admission, as we shall see in Chapter 3. Even in Florence, with 100,000 inhabitants in 1300, each of the 21 guilds averaged only about 350 masters, ranging from 100 in the smallest to 1,600 in the largest.24 In the small German town of Fulda, by contrast, with just 8,500 inhabitants in 1784, the 21 guilds averaged only 13 masters apiece, ranging from the 4 dyers to the 60 shoemakers.25
The ratio of town inhabitants to guilds also covered a wide spectrum. Among the guilds shown in Table 1.1, it ranged from one guild for every 94 inhabitants in twelfth-century Montpellier to one for every 7,692 inhabitants in fifteenth-century Milan. The ratio varied greatly even within the same society, with much greater guild density in York than London, in Dijon than Lyon, in Rome than Venice, in Dordrecht than Amsterdam, in Toledo than Madrid. There was also huge variation within a given time-period. In the Middle Ages, for instance, we observe very high guild densities, with fewer than 900 inhabitants per guild in Montpellier, Cologne, Dordrecht, Bruges, Leuven, and Barcelona, but also very low guild densities, with more than 1,500 inhabitants per guild in London, Paris, Florence, Venice, Milan, and Valencia. The early modern period also showed guild densities as low as 182 inhabitants per guild in York but as high as 4,000 in Paris or Amsterdam. In the nineteenth century, there were still places in Bulgaria, Germany, and Sweden with a guild for every 200 inhabitants, but also cities such as Vienna with only one for every 2,000.
No matter how high the density of guilds relative to the population, guild membership was typically for the few, not the many. Guild masters—those with full guild membership—typically made up only a minority of town inhabitants. Half the population was excluded almost automatically since, as we shall see in Chapter 5, very few guilds admitted female masters. Many men were also excluded from guilds because of the entry barriers discussed in Chapter 3. As a result, guild mastership was reserved for a privileged minority of the urban population.
This can be seen from Table 1.2, which shows that guild masters typically comprised less than 10 per cent of inhabitants and less than half of all household heads. On the liberal end of the spectrum lay early modern London, Aachen, Augsburg, and Danzig, where guild masters made up between 50 and 60 per cent of household heads and between 12 and 13 per cent of inhabitants. Outliers were medieval Coventry and early modern Nördlingen, where c. 80 per cent of household heads and between 14 and 18 per cent of inhabitants were guild masters. In the middle range lay places like early modern Barcelona, Lyon, Rouen, Danzig, Nuremberg, Venice, and Ghent, with guild masters comprising between 40 and 50 per cent of household heads and between 9 and 11 per cent of inhabitants. But in Vienna, Bayonne, Bordeaux, Montpellier, Nantes, Florence, Padua, Rome, Turin, Amsterdam, Malmö,and Simrishamn, guild masters accounted for between 10 and 30 per cent of householders and no more than 2 to 7 per cent of inhabitants. Overall, across the 34 towns in the table, guild masters—the only people who had full membership—made up just 36 per cent of household heads and 8 per cent of inhabitants. These statistics are consistent with other estimates, such as those advanced by Jan Lucassen and Pieter Lourens, according to which guild members made up just 20 per cent of the male labour force of Amsterdam in 1700.26
Guild membership was therefore reserved for the privileged few. Guilds were small relative to the consumer markets they monopolized. They were also small relative to the wider labour market, whose members they largely excluded. Guilds were not all-encompassing workers’ associations analogous to twentieth-century labour unions, but exclusive organizations for relatively well-off, middle-class men.27 The question was not whether a guild was exclusive, but precisely how exclusive it was—which, as we shall see in Chapters 3 and 5, varied across European societies.
TABLE 1.2: Guild Masters as a Percentage of Inhabitants and Householders, 1300–1800 |
|||||||
|
|
|
|
Number of: |
Masters as % of: |
||
Country |
Town |
Date |
Masters |
Inhabitants |
House-holders |
Inhabitants |
House-holders |
Austria |
Vienna |
1736 |
4,115 |
149,908 |
33,313 |
2.7 |
12.4 |
Austria |
Vienna |
1742 |
4,773 |
158,425 |
35,206 |
3.0 |
13.6 |
Austria |
Vienna |
1770 |
4,850 |
208,970 |
46,438 |
2.3 |
10.4 |
England |
Coventry |
c. 1500 |
– |
8,500 |
1,889 |
c. 17.8 |
80.0 |
England |
London |
c. 1300 |
– |
80,000 |
17,778 |
c. 5.6 |
25.0 |
England |
London |
c. 1550 |
8,880 |
70,000 |
15,556 |
12.7 |
57.1 |
England |
London |
c. 1600 |
18,000 |
150,000 |
33,333 |
12.0 |
54.0 |
France |
Bayonne |
1760s |
843 |
14,500 |
3,222 |
5.8 |
26.2 |
France |
Bordeaux |
1760s |
1,838 |
80,000 |
17,778 |
2.3 |
10.3 |
France |
Dijon |
1464 |
851 |
10,638 |
2,364 |
8.0 |
39.7 |
France |
Dijon |
1556 |
2,463 |
12,690 |
2,820 |
19.4 |
36.9 |
France |
Dijon |
1750 |
1,183 |
24,000 |
4,647 |
4.9 |
25.5 |
France |
Lyons |
late 18th c. |
14,820 |
146,740 |
32,609 |
10.1 |
45.4 |
France |
Montpellier |
1549 |
350 |
7,394 |
1,643 |
4.7 |
21.3 |
France |
Montpellier |
1640 |
588 |
20,000 |
2,343 |
2.9 |
25.1 |
France |
Nantes |
1720 |
2,250 |
50,000 |
11,111 |
4.5 |
20.3 |
France |
Rouen |
18th c. |
8,571 |
85,000 |
18,889 |
10.1 |
45.4 |
France |
Toulouse |
1773 |
3,907 |
49,000 |
10,889 |
8.0 |
35.9 |
Germany |
Aachen |
c. 1798 |
3,000 |
23,000 |
5,111 |
13.0 |
58.7 |
Germany |
Augsburg |
1536 |
4,000 |
30,000 |
6,667 |
13.3 |
60.0 |
Germany |
Danzig |
end 15th c. |
– |
30,000 |
6,667 |
c. 11.1 |
50.0 |
Germany |
Frankfurt a.M. |
1587 |
1,247 |
20,000 |
4,444 |
6.2 |
56.1 |
Germany |
Fulda |
1780 |
268 |
8,559 |
783 |
3.1 |
34.2 |
Germany |
Mainz |
1785 |
582 |
22,000 |
4,889 |
2.6 |
30.5 |
Germany |
Nördlingen |
1579 |
1,054 |
7,484 |
1,663 |
14.1 |
83.3 |
Germany |
Nördlingen |
1724 |
878 |
5,954 |
1,323 |
14.7 |
78.9 |
Germany |
Nuremberg |
1621-2 |
3,385 |
40,000 |
10,069 |
8.5 |
33.6 |
Germany |
Nuremberg |
1797 |
2,401 |
25,176 |
5,595 |
9.5 |
42.9 |
Italy |
Florence |
c. 1300 |
7,500 |
100,000 |
22,222 |
7.5 |
33.8 |
Italy |
Florence |
1343 |
3,500 |
75,000 |
16,667 |
4.7 |
21.0 |
Italy |
Padua |
1562 |
1,120 |
35,953 |
7,990 |
3.1 |
14.0 |
Italy |
Padua |
1668 |
1,038 |
33,000 |
7,333 |
3.1 |
14.2 |
Italy |
Rome |
1526-7 |
1,760 |
55,035 |
12,230 |
3.2 |
55.4 |
Italy |
Rome |
1708 |
8,693 |
135,000 |
30,000 |
6.4 |
29.0 |
Italy |
Turin |
1792 |
2,572 |
73,500 |
16,333 |
3.5 |
15.7 |
Italy |
Venice |
1660 |
10,110 |
158,772 |
35,283 |
6.4 |
28.7 |
Italy |
Venice |
1762 |
13,008 |
140,256 |
31,168 |
9.3 |
41.7 |
N. Netherlands |
Amsterdam |
1688 |
12,000 |
200,000 |
44,444 |
6.0 |
27.0 |
S. Netherlands |
Ghent |
1572-84 |
4,000 |
42,000 |
9,333 |
9.5 |
42.9 |
Spain |
Barcelona |
1516-7 |
– |
35,000 |
7,778 |
< 10.8 |
< 48.5 |
Spain |
Cuenca |
1561 |
2,007 |
14,000 |
3,111 |
14.3 |
58.0 |
Spain |
Cuenca |
1771 |
598 |
6,000 |
1,333 |
10.0 |
35.1 |
Spain |
Madrid |
1757 |
16,731 |
109,000 |
24,222 |
15.3 |
39.4 |
Sweden |
Malmö |
1720s |
200 |
4,000 |
889 |
5.0 |
22.5 |
Sweden |
Malmö |
c. 1800 |
150 |
4,000 |
889 |
3.8 |
16.9 |
Sweden |
Simrishhamn |
1780 |
42 |
644 |
143 |
6.5 |
29.3 |
Sweden |
Swedish towns |
end 18th c. |
– |
– |
– |
< 10.0 |
< 10.0 |
Total (n=47) |
|
|
|
|
|
7.9 |
36.1 |
Source: Quantitative guilds database (see text). |
|||||||
Guilds also differed across Europe in other respects. In some societies there were guild-free towns and quasi-urban agglomerations; in others there were guild-free “liberties” inside towns; and many had expanding industrial countrysides where guild regulation was spotty or absent. In some parts of Europe, particularly the Northern Netherlands and England, guilds began to weaken after c. 1500, while in central, eastern-central and southern Europe they retained their strength until 1800 or even later. In some parts of Europe, when industry moved to the countryside urban guilds relaxed their restrictions to remain competitive; in others, guilds lobbied successfully for government protection against rural competitors. New guilds continued to form during the eighteenth century in many parts of central and southern Europe, as medieval guilds split and merged or practitioners of new occupations formed their own corporative organizations. Many European guilds only broke down in the wake of the French Revolution, as France abolished its own guilds in 1791 and exported this institutional reform to neighbouring polities—especially to the Low Countries, parts of western Germany, and northern Italy. In other European societies, such as the Austrian Habsburg Empire, Iberia, and Scandinavia, guilds survived well into the nineteenth century, breaking down finally only after 1860.28 Guilds thus manifested substantial variation across societies and time-periods, and this can help us assess their economic impact.
THE DEBATE SO FAR
The effects of guilds on economy and society have always attracted controversy.29 Contemporaries held strong views about them, with guild members and their political allies extolling their virtues, while customers, employees, and competitors lamented their misdeeds. Many early economic thinkers praised guilds, as, for example, the French government minister Jean-Baptiste Colbert, who ordered all French crafts to form guilds, “so as compose by this means a group and organization of capable persons, and close the door to the ignorant”,30 and the Austrian imperial councilor Johann Joachim Becher, who argued that the authorities in past eras had wisely invented the guilds because “competition weakens the livelihood of the community”.31 Others censured guilds, as did Adam Smith when he called them “a conspiracy against the public”,32 and Anne-Robert-Jacques Turgot, when he told the King of France: “I do not believe that one can seriously and in good faith hold that these guilds, their exclusive privileges, the barriers they impose to work, emulation, and progress in the arts, are of any utility. . .. The total removal of the obstacles that this system imposes on industry and on the poor and laborious sections of your subjects [is] one of the greatest steps to be taken towards the betterment, or rather the regeneration, of the realm”.33 Ordinary people who had to deal with guilds in everyday life held even stronger views, as we shall see in the chapters that follow.
Modern scholars are also deeply divided on guilds. Some claim that guilds were so widespread and long-lived that they must have generated economic benefits. They might, for example, have solved information asymmetries between producers and consumers, overcome imperfections in markets for human capital, created incentives favouring innovation, put pressure on governments to be business-friendly, or generated social harmony by reducing competition, conflict, and inequality.34 Other scholars take a darker view. Guilds, they hold, were in a position to extract benefits for their own members by acting as cartels, exploiting consumers, rationing access to human capital investment, stifling innovation, bribing governments for favours, harming outsiders such as women, Jews, and the poor, and redistributing resources to their members at the expense of the wider economy.35
As this book will show, my own reading of the evidence is that a common theme underlies guilds’ activities: guilds tended to do what was best for guild members. In some cases, what guilds did brought certain benefits for the broader public. But overall, the actions guilds took mainly had the effect of protecting and enriching their members at the expense of consumers and non-members; reducing threats from innovators, competitors, and audacious upstarts; and generating sufficient rents to pay off the political elites that enforced guilds’ privileges and might otherwise have interfered with them.
LEVERAGING SOCIAL CAPITAL
Guilds took action not just in politics, but also in many other spheres beyond the purely economic—conviviality, religion, charity, and forms of cultural expression such as processions, plays, and art patronage.36 Guild members were also often linked by shared citizenship, kinship, intermarriage, and even language and ethnicity, as emerges from the guild entry barriers examined in Chapter 3. These non-economic activities and relationships helped guilds to motivate and inspire their members, as well as to persuade outsiders that guilds deserved their privileged position.
The connection between guilds’ economic and non-economic activities is best understood by thinking of guilds as social networks. As already mentioned, guilds are adduced as historical exemplars of social networks generating a social capital of shared norms, information, sanctions, and collective action.37 Social networks generate social capital, it is argued, because they have two key characteristics—closure and multiplex links.38 “Closure”, when it is clear exactly who belongs to the network, intensifies the quality and reliability of the shared norms, information, and sanctions needed to enforce cooperation. European guilds succeeded in maintaining their special economic privileges by creating norms that were shared by their members, by conveying information swiftly between them, punishing members who violated these norms, and organizing collective action among their members to secure and maintain their privileges. Such closure was viscerally important to European guilds, as we shall see in looking at guild entry barriers in Chapter 3 and guilds’ treatment of women in Chapter 5.
Multiplex links among network members make up the second crucial feature of an effective social network. Social capital is more likely to be generated when members of a network transact with one another in a range of different spheres—economic, social, political, religious, cultural, demographic. These multi-stranded ties make relationships “appropriable”: the resources of one relationship can be brought into play in other relationships with the same person. This gives network members multiple means to reinforce shared norms, convey and receive information about one another, inflict penalties on each other, and efficiently organize collective action.39 Ties inside European guilds were visibly multi-stranded. A guild was formed around a shared set of economic activities and focused mainly on furthering its members’ economic interests. But a guild typically also engaged in social, religious, cultural, and political activities, which created multiplex internal ties that members could use for other purposes.40
Shared sociability was an important norm for many craft guilds, as it was for the religious confraternities that often preceded and overlapped with them.41 The earliest surviving Austrian craft ordinance, the 1368 charter of the Viennese tailors, mandated compulsory attendance at guild assemblies.42 One guild in the Swedish town of Hellestad stressed the central role of sociability in 1404 when it ordained that the guild banquet “has been instituted not for drinking or greed, but for mutual support, assistance and friendship”.43 In late fifteenth-century Bristol, the weavers invited the town mayor and his brethren to the guild’s St Katherine’s feast every year, thereby securing “amity and affection”, not just inside the guild but between the guild masters and the city’s political leaders.44 The worsted-weavers’ guild of the Württemberg district of Wildberg held assemblies on average every seven months between 1598 and 1760, attended by nearly 100 per cent of masters and involving not only economic decisions, but political strategizing and the festive consumption of bread and wine at the guild tavern.45 Most guilds in early modern Vilnius, despite being confessionally mixed, required the entire membership to attend the funeral of any guild member or any member of his household, including females and servants. Money fines were levied from violators.46 Up to the end of the eighteenth century in German city-states such as Aachen, citizens’ social lives centred around the guild headquarters, where masters discussed industrial issues, but also took part in communal festivities, religious observances, welfare allocation, conflict resolution, the election of town council delegates, and political strategizing.47 Even the much less closely knit London guilds where, as we shall see in Chapter 9, attendance at assemblies was lower and declined drastically before 1600, regarded corporate commensality as increasing “love and amytie withe better knowledge amonge the Bretherne”.48 Sociability generated individual enjoyment, of course, but it also fostered the multiplex ties by which guild members shared information, punished opportunism, and organized collective action.
Religious observance created a second set of multi-stranded ties. As craft guilds emerge into view in Europe from the eleventh century onwards, religious confraternities are not always clearly distinguished from occupational guilds.49 Devotional confraternities were often organized around shared occupations, while occupational guilds often bore the name of a patron saint, employed a chaplain or priest, engaged in good works, formed ties with religious houses, and amassed “ghostly treasure”—in the evocative phrase of the London tailors’ guild, referring to the indulgences and other religious privileges the guild procured and hoarded.50 Contemporaries recognized the importance of shared religious observance for motivating cooperation, reducing transaction costs, and organizing collective action, as in the 1320s when the Florentine city council forbade the woolworkers
to make constitutions or statutes . . . within the guise of a fraternity or otherwise, and under the pretext or cover of religion, or of providing for funerals or religious offerings, or for any other purpose . . . except by special licence of the consuls of that [officially organized] craft under whose authority they stand . . .. And they are disallowed from having or carrying any banner.51
Assembling in a particular place of worship could also assist guild members in market collusion, as in 1321, when the London weavers’ guild was accused of engaging in cartelistic price-setting, “through an agreement reached in the church of St Margaret Pattens”.52
After the sixteenth-century Protestant Reformation, craft guilds continued to pursue religious activities, though in different ways depending on confession. In Catholic societies, guild devotion continued as before. In Spain, Fernández Navarrete complained in 1626 that there were so many guild fraternities that “artisans spend half the year vying with one another in display rather than in devotion”.53 In early-seventeenth-century Granada, Henríquez de Jorquera’s chronicle describes a town blazed with images and crosses in the particular locations where each craft guild kept its chapel, worshipped, and held its assemblies.54 In Piedmont, well into the eighteenth century, craft guilds continued to focus their activities around particular churches and saints.55 A survey of 264 guild assemblies in eighteenth-century Spain found that 45 per cent were held in religious and public buildings, such as churches, convents, and hospitals.56
But even in Protestant Europe, guilds engaged in religious politics, struggled over religious artefacts, held religious services, and mandated attendance at members’ weddings and funerals. In sixteenth-century Augsburg, the weavers’ guild supported the Protestants; the butchers’ guild clove to the Catholics; and the foreman and masters of the shoemakers’ guild carted the guild’s ceremonial candles away from the city cathedral to prevent their being used in “idolatrous worship” by the old clergy.57 In Scotland, guilds retained a strong religious element under Protestantism: in post-Reformation Dundee, the baxters’ (bakers’) guild erected a church pew for their members with the inscription “Bread is the staff of life”, only to be countered by the fleshers’ (butchers’) guild with the pew inscription, “Man shall not live by bread alone”.58 In York, when the tailors’ fraternity of St John the Baptist was dissolved during the Reformation, the name was discreetly dropped, but most of its religious infrastructure and activities, including the meeting-hall, alms-house, and holy-day feasts, were retained by the guild.59 In London in the mid-1590s, the tallow-chandlers’ guild collected money for a new hearse cloth, with voluntary contributions from half the active members.60 In Dublin in the 1660s, clashes between dissenters and conformists divided the shoemakers’ guild, which petitioned the government to remove one of its officers for his religious views.61 In German Protestant territories, many guilds continued to hold religious services at their annual assemblies.62 In Lutheran Württemberg in the seventeenth and eighteenth centuries, the worsted-weavers’ guild kept hearse-cloths for members’ family funerals,63 and at the annual assembly of the shepherds’ guilds in Markgröningen on St Bartholomew’s day, every master was required to attend two church sermons, each immediately followed by a reading of the guild ordinance.64 In Lutheran Mömpelgardt (modern Montbéliard) to the end of the eighteenth century, guilds mandated compulsory attendance at members’ funerals and held assemblies on the days of their former patron saints.65 In Lutheran Sweden in the eighteenth century, guilds played a central role in organizing craftsmen’s funerals.66 And, as we shall see in Chapter 3, virtually all European guilds used religious affiliation as a barrier to entry, excluding from membership anyone outside the local majority confession. Religion thus contributed to the closure and multiplex internal bonds that helped generate guild-specific social capital. Indeed, statistical analysis of 340 guilds in the late medieval Venetian Republic suggests that those guilds with origins in religious confraternities imposed significantly stronger regulations on entry and competition than guilds with no religious origins.67
Multi-stranded ties among guild members were also fostered by processions and festivals, although after the early sixteenth century this was more common in Catholic than Protestant places. Many Italian guilds organized displays for religious festivals.68 Sometimes guild spending on religious display was so lavish that it threatened solidarity rather than cementing it, as in 1770 when several Piedmontese printers complained that their guild’s “mutual solidarity” funds were being wasted on celebrating its patron saint, and this prompted a devastating external investigation of the guild’s internal organization.69 In Austria, by contrast, the huge spate of new guild formations throughout the eighteenth century is thought to have come about precisely because “during the Catholic Counter-Reformation, collective display in the numerous processions and pilgrimages increased in importance”.70
Multiplex social relationships inside guilds were sometimes enhanced by common geographical origins, language, and culture. The Viennese chimney-sweeps’ guild, founded in 1664, was dominated by a group of twelve to fourteen families originating in Ticino, Grisons, and northern Lombardy; in the late eighteenth century the guild still recruited two-thirds of its apprentices from that region. The Viennese silk-cloth-weavers’ guild, founded in 1710, numbered among its first eleven members five men from the same family in Bergamo and four from other parts of Italy; Italians continued to dominate the guild over the following decades and in the 1720s the government guild commissioner still had to master the Italian language to understand proceedings at the guild assemblies.71 In many parts of eastern-central Europe, as we shall see in Chapter 3, most guilds made admission conditional on German ethnicity, thus excluding Estonians, Czechs, Poles, and Wends (west Slavs).
Multi-stranded ties inside guilds were often cemented by family bonds. Between 1650 and 1790, two-thirds of the masters in the rural-urban scythe-smiths’ guild of Upper Austria married masters’ daughters, and in 1784 five families owned half of all workshops.72 The twelve to fourteen families that monopolized mastership niches in the Viennese chimney-sweeps’ guild also practised strong endogamy throughout the eighteenth and early nineteenth centuries.73 The same was true of the gold-drawers’ guild in the town of Trévoux (now in France), where in the early eighteenth century almost all marriages took place among families of existing guild members.74 Familial bonds inside guilds were often encouraged by guild entry rules which, as Chapter 3 discusses, favoured masters’ sons, sons-in-law, and other relatives.
Craft guilds thus engaged in many activities that went far beyond the economic. Debate sometimes arises over whether these non-economic activities were pursued for their own sake or served economic ends.75 But the two were not mutually exclusive. Religious, sociable, cultural, and familial ties could simultaneously satisfy the preferences of guild members and, for that very reason, intensify the multiplex ties that generated social capital. Shared sociability, worship, festivity, funerary observance, language, intermarriage, and kinship generated multiple means by which members could foster norms, exchange information, punish deviants, and organize collective action. Joint non-economic activities were undoubtedly pursued for their own sake, but they also enhanced guilds’ ability to motivate and inspire their members to organize economic cooperation.76
As this book will show, the social capital generated by multi-stranded ties facilitated the capacity of guild members to engage in collective action. This might have good economic effects, for instance if it imposed social penalties on poor-quality work, helped apprentices find good masters, transmitted information about innovations, or facilitated resistance to harmful government policies. It could also cause harm, for instance by easing surveillance of members who adopted disruptive innovations, imposing ostracism on members who failed to discriminate against women or Jews, or facilitating market collusion. All these are theoretically possible. But which predominated empirically?
GETTING DATA ABOUT GUILDS
Systematic information on guild behaviour is scarce. There are a few great compilations of guild ordinances—for Paris in 1268 and 1766; for the Middle Rhine towns between 1300 and the Thirty Years War; for Italian cities between 1220 and 1800; for Hungarian towns between 1600 and 1883. But such compilations are rare. For many societies, even to look at guild rules we must turn to individual ordinances for specific guilds in particular places and times.
How guilds actually behaved, as opposed to how the rules ordered them to behave, is even harder to get at. What guilds did in reality can only be teased out through painstaking analyses of variegated, local-level sources, such as guild accounts, mastership admissions, journeyman registrations, apprenticeship enrolments, assembly resolutions, guild court minutes, petitions, and records of legal conflicts. These sources survive only for a minority of guilds in a few time-periods. We only know as much as we do about European guilds because hundreds of devoted scholars have carried out labour-intensive micro-studies of guilds in specific places and periods.
Case studies such as these have proved the mainstay of most accounts of medieval and early modern guilds. Without such dedicated work, our knowledge of guilds in Europe would be as tentative as that for other continents.77 But precious though these European case studies are, they can only reveal the activities of specific guilds in particular places and times. Some are fortunate enough to exploit documentary sources that can be analyzed quantitatively, but most rely on piecing together scattered evidence from purely qualitative sources. Through no fault of their own, these studies can only draw limited conclusions. An individual case can illustrate what is possible, and this can be a strong way of refuting an erroneous hypothesis. But one example cannot show what is typical. Nor do scattered qualitative references enable systematic comparisons among guilds of different occupations, societies, or periods. Almost any theory of guilds can find at least one guild that did something consistent with that hypothesis.
Detailed micro-studies of specific guilds and industries are hugely valuable, and this book uses them to the full. But in an effort to transcend some of their limitations, this study also sought to bring together observations of guild activities from hundreds of studies of guilds in different European societies. This exceptionally labour-intensive and time-consuming data compilation exercise yielded two databases, one qualitative and the other quantitative.
The first database consists of 12,051 qualitative observations, each showing a guild or group of guilds in a particular period and place engaging in a particular activity of interest. Examples of these activities include engaging in collective devotion, paying rulers for privileges, restricting entry to an occupation, fixing prices and output, limiting women’s work, inspecting quality, regulating training, forbidding or permitting innovations, holding guild assemblies, and dozens more. The observations come from a wide range of different societies, corresponding to 23 countries of modern Europe, as can be seen in Table 1.3. The earliest observation dates from 1095 (the admissions policy of the Pisa smiths’ guild), the latest from 1862 (the sale of a mastership license in the Willstätt raftsmen’s guild, for three times the value of a cow).78
By assembling qualitative observations on many variables of interest, this data compilation makes it possible to transcend the boundaries of the individual case study. With care, the data can be used to check how typical or unusual a particular guild activity was across places and times. Some activities were virtually universal: thus nearly every guild in Europe excluded Jews; the fact that four guilds in eighteenth-century Amsterdam admitted them can thus be set in a wider context, as we shall see in Chapter 3. Other activities were rare in one time-period but universal in another: thus in the medieval period very few guilds conducted examinations of skills, as we shall see in Chapter 7; by the seventeenth century many more guilds did so. Still other activities were rare in some societies but widespread in others: thus most German and Spanish guilds excluded individuals with contact to a “dishonourable” activity (such as skinning or healing animals), whereas guilds elsewhere in Europe hardly ever did this. In a few cases we can even analyse the qualitative observations quantitatively—for instance, by assembling every reference to any all-female guild and analyzing their characteristics, as in Chapter 5. But for the most part, this book uses this qualitative database to gain a sense of what is usual or unusual for guild activities in different societies and periods.
TABLE 1.3: Guild Observations and Population by Country, c. 995–c. 1899 |
|||||||
|
Qualitative observations |
Quantitative observations |
Total observations |
Population 1000–1900 |
|||
Country |
no. |
% |
no. |
% |
no. |
% |
% |
Austria |
371 |
3.1 |
159 |
3.0 |
530 |
3.0 |
2.0hh |
Bohemia (Czech Republic) |
202 |
1.7 |
48 |
0.9 |
250 |
1.4 |
4.1ll |
Bulgaria |
177 |
1.5 |
12 |
0.2 |
189 |
1.1 |
1.5ll |
Denmark |
65 |
0.5 |
15 |
0.3 |
80 |
0.5 |
0.9ll |
England |
1,133 |
9.4 |
534 |
10.0 |
1,667 |
9.6 |
– |
Scotland |
26 |
0.2 |
6 |
0.1 |
32 |
0.2 |
– |
England + Scotland (GB) |
1,159 |
9.6 |
540 |
10.1 |
1,699 |
9.8 |
13.2ll |
Estonia |
78 |
0.6 |
3 |
0.1 |
81 |
0.5 |
– |
Finland |
11 |
0.1 |
13 |
0.2 |
24 |
0.1 |
0.8ll |
France |
1,717 |
14.2 |
1,149 |
21.5 |
2,866 |
16.5 |
14.4hh |
Germany |
3,369 |
28.0 |
1,430 |
26.8 |
4,799 |
27.6 |
17.8hh |
Greece |
2 |
0.0 |
2 |
0.0 |
4 |
0.0 |
1.8ll |
Hungary |
43 |
0.4 |
14 |
0.3 |
57 |
0.3 |
2.3ll |
Ireland |
2 |
0.0 |
— |
0.0 |
2 |
0.0 |
1.5ll |
Italy |
2,116 |
17.6 |
487 |
9.1 |
2,603 |
15.0 |
11.9hh |
N. Netherlands |
548 |
4.5 |
405 |
7.6 |
953 |
5.5 |
1.7hh |
Norway |
26 |
0.2 |
10 |
0.2 |
36 |
0.2 |
0.7ll |
Poland |
245 |
2.0 |
9 |
0.2 |
254 |
1.5 |
8.0ll |
Portugal |
26 |
0.2 |
2 |
0.0 |
28 |
0.2 |
1.8ll |
Romania |
24 |
0.2 |
3 |
0.1 |
27 |
0.2 |
3.6ll |
S. Netherlands (Belgium) |
485 |
4.0 |
626 |
11.7 |
1,111 |
6.4 |
2.2hh |
Spain |
961 |
8.0 |
156 |
2.9 |
1,117 |
6.4 |
6.9ll |
Sweden |
121 |
1.0 |
193 |
3.6 |
314 |
1.8 |
1.7s |
Switzerland |
303 |
2.5 |
57 |
1.1 |
360 |
2.1 |
1.1hh |
Total |
12,051 |
100.0 |
5,333 |
100.0 |
17,384 |
100.0 |
100.0 |
Source: Guilds: quantitative and qualitative guilds databases (see text). Population: Maddison data [http://www.ggdc.net/maddison/maddison-project/home.htm]. Notes: Population shares calculated in terms of average between estimate of 1000 and estimate of 1900. Estonia: no population estimate available. Ireland: Maddison reports no population estimate for 1000; the current calculation imputes an estimate of 350,000 for that year, based on average 230% increase in population for other European countries between 1000 and 1500. s = percentage of guilds observations and percentage of population not significantly different at 0.10 level. ll = percentage of guilds observations significantly lower than percentage of population at 0.05 level. hh = percentage of guilds observations significantly higher than percentage of population at 0.05 level. |
|||||||
The second database is quantitative. It assembles 5,333 observations of particular types of guild behaviour that can be measured in numbers. Examples include the share of guild expenditures allocated to lobbying the public authorities, the proportion of masters’ sons among new guild members, the share of women operating guild workshops, the percentage of guild fines imposed for quality violations, the share of guilds with apprenticeship or journeymanship requirements, the number of guilds formed in particular periods and places, the attendance rate at guild assemblies, the date of guild abolition in different polities. These quantitative observations again come from a wide range of different societies, corresponding to 22 countries of modern Europe over a period of more than nine centuries.79 The earliest observation dates from the late tenth century (the numerus clausus imposed by the Pavia tanners’ guild), the latest from 1859 (the percentage of female apprentices in the Vienna silk-weavers’ guild). The observations in this quantitative database do not cover all guilds activities. They can only be assembled for things guilds did that lend themselves to quantitative analysis, mattered to those who kept written records, and have attracted the interest of more than one scholar. But for the subset of guild behaviour that satisfies those conditions, it is illuminating to compare quantitative observations across multiple places and periods.
How representative are these data? Table 1.3 shows the distribution of observations across the 23 European societies in the database, relative to the distribution of population in those societies according to the most widely used demographic estimates for pre-industrial Europe.80 As the table shows, for most European societies, their share of observations in the guilds database is either significantly higher or lower than their share of the total average European population during this period.81 For 13 of the 20 societies in the database for which separate population estimates are available, the share of observations in the guilds database lies within two percentage points of their estimated share of the European population.82 Thus France accounts for 16.5 per cent of observations of guilds and 14.4 per cent of average European population across the period. Spain accounts for 6.4 per cent of guild observations and 6.9 per cent of European population. A similar congruence between the share of guild observations and the share of European population holds for most of the smaller European societies in the database.
But there are also societies whose representation in the guilds database differs from their share of the European population by more than two percentage points. Four societies are under-represented in the guilds databases to this degree: Bohemia (the modern Czech lands), Britain (England and Scotland combined), Poland, and the territory of modern Romania. In the case of Bohemia, Poland, and Romania, this reflects the fact that these eastern-central European societies were subject to serfdom, with a predominantly agricultural occupational structure and low urbanization.83 The same goes for Scotland, a predominantly rural society in which guilds only began to form in the fifteenth century, and then weakened rapidly after 1672. For England, under-representation in the guilds database may result from the fact that, as we shall see, English guilds were very active up to the sixteenth century but then gradually ceased to do many of the things guilds continued to do elsewhere in Europe.
Conversely, four societies are over-represented, in the sense that their share of guilds observations is more than two percentage points higher than their share of European population: Germany, Italy, the Northern Netherlands, and the Southern Netherlands. Guilds generated a particularly notable wealth of observations for Germany, which comprised 18 per cent of Europe’s population but 28 per cent of guild observations. This is due partly to the existence of separate guild systems in each of the 350 separate sovereign German territories before the end of the Old Empire in 1805, partly to the rich tradition of German guild historiography, and partly to the extraordinary strength of German guilds (discussed in later chapters). Italy, too, is overrepresented, with 15 per cent of guilds observations but only 12 per cent of Europe’s population. Again, this arises partly from the existence of separate guild systems in each Italian territory before national unification and partly from the rich Italian guild historiography, although, as we shall see, Italian guilds were only of intermediate strength by European standards. The relative wealth of information on guilds in the two parts of the Low Countries, each of which accounts for only about 2 per cent of Europe’s population but 4 to 6.5 per cent of the guilds database, also arises from a rich guild historiography, encouraged by the relatively small size of these societies which has facilitated admirably comprehensive analyses of their guild systems.
Although, therefore, the guilds database is not perfectly representative in a demographic weighting, more than half of the 23 European societies it records are represented within two percentage points of their demographic weight in medieval and early modern Europe. Those societies that were under-represented are primarily those that had either a relatively low share of non-agricultural activity (eastern-central Europe, Scotland) or whose guilds suffered from early stagnation (England and Scotland). Those societies that were over-represented are mainly those with early and high levels of non-agricultural activity and a particularly rich guilds historiography.
TABLE 1.4: Guild Observations and Population by Period, Europe c. 995–c. 1899 |
||||||||
Period |
Qualitative observations |
Quantitative observations |
Total observations |
Population (thousands) |
||||
|
no. |
% |
no. |
% |
no. |
% |
average |
per guild obs. |
900–99 |
— |
0.0 |
1 |
0.0 |
1 |
0.0 |
30,200 |
0 |
1000–99 |
2 |
0.0 |
– |
0.0 |
2 |
0.0 |
30,200 |
15,100 |
1100–99 |
53 |
0.4 |
4 |
0.1 |
57 |
0.3 |
49,521 |
869 |
1200–99 |
821 |
6.8 |
259 |
4.9 |
1,080 |
6.2 |
49,521 |
46 |
1300–99 |
765 |
6.3 |
178 |
3.3 |
943 |
5.4 |
49,521 |
53 |
1400–99 |
1,321 |
11.0 |
432 |
8.1 |
1,753 |
10.1 |
68,842 |
39 |
“medieval” |
810 |
6.7 |
84 |
1.6 |
894 |
5.1 |
|
|
Total medieval |
3,772 |
31.3 |
958 |
18.0 |
4,730 |
27.2 |
|
|
1500–99 |
2,082 |
17.3 |
732 |
13.7 |
2,814 |
16.2 |
68,842 |
24 |
1600–99 |
2,257 |
18.7 |
822 |
15.4 |
3,079 |
17.7 |
88,420 |
29 |
1700–99 |
2,437 |
20.2 |
2,185 |
41.0 |
4,622 |
26.6 |
98,741 |
21 |
1800–99 |
361 |
3.0 |
299 |
5.6 |
660 |
3.8 |
210,137 |
318 |
“early modern” |
1,085 |
9.0 |
265 |
5.0 |
1,350 |
7.8 |
|
|
Total early modern |
8,222 |
68.2 |
4,303 |
80.7 |
12,525 |
72.0 |
|
|
Medieval & early modern |
57 |
0.5 |
72 |
1.4 |
129 |
0.7 |
|
|
Total |
12,051 |
100.0 |
5,333 |
100.0 |
17,384 |
100.0 |
|
|
Source: Guilds: quantitative and qualitative guilds databases (see text). Population: Maddison data [http://www.ggdc.net/maddison/maddison-project/home.htm]. Notes: “medieval”: unspecified date or spanning multiple centuries before 1500. “early modern”: unspecified date or spanning multiple centuries from 1500 onwards. Medieval & early modern: spanning multiple centuries pre- and post-1500. Population: estimated, in thousands. Figure for nineteenth century is for 1850. |
||||||||
What about chronological representation? Table 1.4 shows the distribution of observations in the guilds database across the nine centuries covered by this book. In compiling the database, every effort was made to disaggregate observations to specific dates and centuries since, as later chapters will show, the guild was not a static institution but one that developed and metamorphosed over time. But some observations could not be assigned to a particular century, either because the underlying research study from which it was drawn did not give this information or, more often, because the observation itself spanned several centuries. This means that about 5 per cent of observations in the database could only be assigned generally to the medieval period, about 8 per cent generally to the early modern period (including the nineteenth century), and about 1 per cent spanned the medieval and early modern periods.
The other 86 per cent of observations, however, could be assigned to particular centuries, and their distribution is consistent with what we know of the trajectory of development of craft (and other non-merchant) guilds in Europe: small numbers from 1000 to 1200, a remarkable increase between 1200 and 1500, another big rise from 1500 to 1700, a huge proliferation in the eighteenth century, and then an abrupt decline after 1800 to below the level of the thirteenth century. Of course, this time-path reflects changes in record-keeping as well as economic reality. The growth of the modern state and bureaucratic government after c. 1500 caused all forms of record-keeping to proliferate, which makes it unsurprising that 72 per cent of observations in the guilds database come from the early modern period. The greater density of guild observations after 1500 than before holds even when controlling for population, as shown by the final column of Table 1.4. Excluding those observations that cannot be assigned to a specific century, there were between 39,000 and 53,000 inhabitants per guild observation from 1200 to 1500, and only 24,000 to 29,000 inhabitants per guild observation in the period between 1500 and 1800.
The observations of guilds in this database can be categorized in many ways, as would be expected of more than seventeen thousand observations spanning twenty-three countries and nine centuries. Table 1.5 shows how they are distributed across the spheres of guild activity analyzed in the chapters that follow. The central spheres of economic activity by guilds are plentifully covered, with 2,500 to 5,000 observations apiece concerning such matters as entry barriers, market manipulation, women’s economic participation, and human capital investment, and 400 to 1,300 observations apiece on political activity, quality regulation, innovation, and institutional influences on growth. Only on “social capital”— those activities of guilds that were not directly economic, such as devotional or cultural expression—are there fewer than 400 observations, and it is to be hoped that future research will enhance this component of the database.
TABLE 1.5: Guild Observations by Sphere of Activity, Europe, c. 995–c. 1899 |
|||||||
|
|
Qualitative observations |
Quantitative observations |
Total observations |
|||
Sphere of activity |
Chapter |
no. |
% |
no. |
% |
no. |
% |
Social capital |
1 |
350 |
2.9 |
0 |
0.0 |
350 |
2.0 |
Political activity |
2 |
851 |
7.1 |
207 |
3.9 |
1,058 |
6.1 |
Entry barriers |
3 |
3,073 |
25.5 |
1,846 |
34.6 |
4,919 |
28.3 |
Market manipulation |
4 |
2,529 |
21.0 |
11 |
0.2 |
2,540 |
14.6 |
Women’s economic participation |
5 |
2,503 |
20.8 |
992 |
18.6 |
3,495 |
20.1 |
Quality regulation |
6 |
440 |
3.7 |
33 |
0.6 |
473 |
2.7 |
Human capital investment |
7 |
969 |
8.0 |
1,657 |
31.1 |
2,626 |
15.1 |
Innovation |
8 |
691 |
5.7 |
15 |
0.3 |
706 |
4.1 |
Institutional influences on growth |
9 |
645 |
5.4 |
572 |
10.7 |
1,217 |
7.0 |
Total |
|
12,051 |
100.0 |
5,333 |
100.0 |
17,384 |
100.0 |
Source: Qualitative and quantitative guilds databases (see text). Notes: Spheres of guild activity correspond to chapters of present book. |
|||||||
These data compilations are a work in progress. They cannot, and do not claim to, encompass remotely all of the tens of thousands of craft guilds that existed in Europe between the end of the Dark Ages and the Industrial Revolution. That would be beyond the capacities of a single scholar. Even in their current state, they depend on the dedicated work of hundreds of scholars, and before them on the record-keeping of thousands of guildsmen, public officials, and ordinary people in cities, towns and villages across the continent for a period of nine centuries. By its very nature, a data compilation such as this cannot represent all activities undertaken by guilds in all occupations, societies and time-periods, even in Europe. It is hoped that the database will be expanded to cover other occupations, societies and time-periods, other aspects of guild activity, and the many occupational associations formed in non-European societies. To encourage this collaborative process and to facilitate analyses, additions, and improvements by future researchers, both the qualitative and the quantitative databases assembled for the present book have been made available on open access.84 This book uses these databases in their open-access form to address the questions posed at the beginning of this chapter about how guilds affected the economy.
THE STRUCTURE OF THIS BOOK
Guilds did many things, so they provide an excellent demonstration of the principle that in analyzing the net effect of an institution, it is imprudent to focus on any one of its activities in isolation.85 The chapters that follow focus in turn on seven guild activities that affected key parts of the economy: doing deals with governments, limiting entry, manipulating markets, restricting women, supervising quality, regulating training, and controlling innovation.
Chapter 2 begins this exploration by looking at how a guild got—and kept—privileges over particular economic activities. Though guilds are sometimes called private-order institutions, in practice they were recognized by the public authorities. They enjoyed legal entitlements and coercive power, in return for which they channelled favours to political elites—so much so that the French controller-general described guild affairs in 1776 as “one of the most abundant sources of profits for the people of the Palace”.86 Guilds thus provide an excellent context for investigating how social networks deploy social capital to shape government action. Guilds, as Chapter 2 reveals, gave people practising a particular occupation an institutional mechanism enabling them to exercise more effective political pressure than they could by acting independently as individuals.
A first entitlement sought by any guild was to decide who could practise certain economic activities. Chapter 3 explores how guilds defined and enforced this entitlement. Every guild aimed to secure the exclusive right for its members to do specific kinds of work in a particular place—as the Wiener-Neustadt butchers’ guild put it in 1310, officially “no-one other than they alone may supply meat by the penny or by the farthing”.87 A guild also secured the right to decide who could gain admission to the guild “mysteries”; as the London tailors put it in 1326, guild admission was controlled “by the honest and lawful men of the misteries”.88 To enforce these privileges, as Chapter 3 shows, guilds erected an elaborate array of entry barriers, making admission costly or impossible for any applicant who could not satisfy conditions relating to citizenship, ethnicity, religion, occupation, wealth, property, fees, marriage, age, legitimate birth, parental occupation, ancestral “purity”, reputation, or approval by existing guild members. Chapter 3 also examines the rich evidence, both direct and indirect, bearing on how guild entry barriers were interpreted and enforced. By limiting the number of practitioners, guilds sought, in the words of the Burgdorf shoemakers’ guild in 1785, “to create a better and surer livelihood for the remainder”.89 But, as Chapter 3 shows, guild entry controls had knock-on consequences for the wider economy.
A guild controlled not only who could legally practise an occupation, but what he could do once he got in—the “usages and customs” of the trade, in the phrase used in Etienne Boileau’s famous 1268 compilation for the 101 guilds of Paris.90 The guilds’ major preoccupation in these collectively agreed regulations, as Chapter 4 shows, was to manipulate markets in what their members produced and in the inputs they used to produce it. As the sole legitimate practitioners of a particular occupation, it made sense for guild masters to use their guilds to restrict internal rivalry and external competition that might push down prices and push up costs. In the words of the London founders in 1507, the guild had first to suppress outsiders who sold the same wares “much more cheaply”, and then to set minimum selling prices for guild members, “so they might have a comfortable living by doing so”.91 Chapter 4 investigates the many ways guilds limited competition in order that their members might have a more comfortable living, at the expense of customers, employees, rivals, and the wider economy.
One of the most far-reaching ways in which guilds manipulated markets, as Chapter 5 shows, was by restricting the economic options of half the population— women. Not all guilds adopted the hard-line position of the German jurist Adrian Beier who declared in 1683 that “[m]asculine sex is one of the indispensable basic preconditions for admission to a guild”.92 True, some guilds absolutely forbade any activity by any females. But most guilds granted masters’ wives and widows the right to work, at least in a conditional and limited way; some extended this to daughters and other dependent female family members; and a few admitted female masters. There were even a few guilds set up by women themselves, though usually under male tutelage. Chapter 5 examines how guilds balanced the benefits their members derived from cheap and productive female workers against the threat posed by female competitors. As this chapter shows, pre-industrial guilds provide a fruitful context for analysing the cultural, technological, and institutional determinants of economic discrimination—and its potential costs for economic performance.
Guilds intervened in markets to benefit their own members, but did they also intervene to correct market failures and benefit the wider economy? Chapter 6 examines one major market failure guilds might have helped solve: the potential for information asymmetries between producers and consumers about the quality of goods and services. Many guilds erected market regulations, like the Paris tallow-chandlers in 1268 who petitioned the authorities to confirm their guild privileges “for good and loyalty and for the profit of all, because false work in tallow-candles is too damaging and villainous a thing for poor and rich”.93 To address such concerns, guilds required producers and products to be guild-certified, inspected workshops or wares, and penalized quality violations. Guilds also engaged in many unrelated activities which affected quality unintentionally. Chapter 6 explores the evidence on information asymmetries about quality, the institutional mechanisms available to solve them, and the outcomes in different sectors of the European economy. Guilds, it finds, shed light on the balance between market failures, state failures, and the failure of particularized institutions in intermediating between producers and consumers.
Chapter 7 examines how guilds dealt with human capital investment, a second sphere in which markets are often thought to fail. Many guilds imposed comprehensive training requirements, as in the case of the seventeenth-century Toledo silk-twisters who justified their exclusive legal privileges on the grounds that “the art of silk is useful and has many secrets . . . and because of the quality and secrets of the said art, no matter how deft a man may be, he cannot master them except through long practice and the passage of time”.94 Chapter 7 examines the externalities and information asymmetries which could cause markets in human capital investment to fail, and the ways in which guilds regulated training in pre-industrial crafts and services.
Innovation is a final sphere in which market failures are widespread in pre-modern economies, as in modern ones. Information—including about new and better ways of making things—is what economists call a public good. You cannot prevent an idea from spreading without charge, and once you tell it to one person you can tell others at no extra cost. So ideas will be under-provided if people try to trade them in markets. Either people will not devise an innovative way of making something, since they cannot profit from their own efforts. Or they will devise it but tell it only to a few paying customers, even though it could benefit society more widely if it were free. Non-market institutions such as guilds might be able to provide better incentives than markets for the innovation and diffusion of new ideas. Chapter 8 examines how guilds dealt with technological innovation. On the one hand, contemporaries frequently complained that guilds blocked new techniques and practices, as in 1604 when the Middle Rhine bakers’ guilds instructed bakers of white bread to “refrain completely from all innovation in baking”.95 On the other, guilds were in a position to generate cartel rents, and this might have encouraged their members to incur the costs of invention. Guilds might also have encouraged diffusion of technological knowledge through compulsory apprenticeship, mandatory travelling by journeymen, or the spatial clustering of practitioners. Guilds could also affect innovation unintentionally by things they did for other reasons. Guilds thus provide a rich context for investigating the role of different institutional mechanisms in encouraging the invention and diffusion of innovations.
A basic question about any economic institution is how it affects the performance of the economy. Were strong guilds associated with stronger or weaker economic growth? Chapter 9 investigates this question. It examines different measures of guild strength, in terms of guild numbers, producer-merchant relations, guilds’ internal cohesiveness, their relationship with the state, characteristics of towns, interaction with the countryside, and the role of guild-free enclaves. The chapter then examines how guild strength and weakness were associated with economic performance across pre-industrial Europe.
Chapter 10 draws together the analyses from earlier chapters to address the questions that started this book. What do guilds tell us about the kind of institution that makes economies work well or poorly? Did guilds use the trust and social capital they generated as closely knit networks to correct market and state failures? Or did they use it to extract resources for their members at the expense of the rest of society? What were the unintended effects of guild activities for the economy at large? And why did guilds themselves exist so widely, survive for so long, and take so long to disappear? Did they serve economic efficiency, as some have claimed, or did they enable powerful and privileged groups to redistribute resources coercively? Guilds, it concludes, help us understand not only how institutions affect economies but why institutions themselves exist.
Epigraph sources: Hosiers’ guild ordinance, Depping 1837, 142; Judges describing velvet-weavers’ guild, Unwin 1908, 41–42; Hatters’ guild privileges, Göttmann 1977 [Handwerk], 97–98 n. 46; Complaint against the velvet-weavers’ guild, Montemayor 1996, 230; Silk-twisters’ guild ordinance, MacKay 2006, 32; Complaint against the guilds of Holland, De Witt [De la Court] 1743 [1662], 74; Adam Smith’s observation on urban guilds, Smith 1776, Book IV, Chapter VIII, p. 145, para. c27.
1 For this famous definition, North 1990, 3.
2 Ogilvie 2005; Ogilvie 2011; Puttevils 2009; Hillmann 2013; Ogilvie and Carus 2014.
3 Quoted in Neumeyer 1956, 145.
4 Edwards and Ogilvie 2012, 132–36.
5 Edwards and Ogilvie 2012, 146; Gelderblom 2013.
6 Quoted in Depping 1837, 142.
7 See Table 9.4 on the dates of guild abolition in different parts of Europe.
8 Ogilvie 2004 [How]; Ogilvie 2004 [Guilds]; Ogilvie 2005.
9 Putnam et al. 1993, 163–85; Putnam 2000, 319, 322–23, 325, 346–47; Coleman 1988, S97–S99.
10 Stiglitz and Ellerman 2000; Dasgupta 2000, 351–52; Raiser 2001, 231.
11 Göttmann 1977 [Handwerk], 97–98 n. 46.
12 For recent surveys of alternative theoretical approaches to explaining institutions, see Acemoglu, Johnson, and Robinson 2005; Ogilvie 2007 [Whatever]; Ogilvie and Carus 2014.
13 Ehmer 2001 [Artisans], 818.
14 For surveys of Chinese and Indian guilds, revealing the still exiguous factual basis but opening up stimulating questions for future research, see Moll-Murata 2008, 2013 on China; and Roy 2013 on India.
15 Ehmer 2001 [Artisans], 818.
17 So much so that some scholars have questioned its use at all; see Thrupp 1942, 165.
18 Van den Heuvel 2007, 148.
19 See the survey in Ogilvie and Cerman 1996.
20 On merchant guilds, see Ogilvie 2011.
21 Ehmer 2005, 62–63.
22 Saint-Léon 1922, 381, 430, 472, 490.
23 Goedde 2000, 42; Ehmer 2001 [Family], 194.
24 Najemy 1979, 59–60 n. 21.
25 Walker 1971, 99.
26 Lourens and Lucassen 1998, 145.
27 Ehmer 2005, 70–71.
28 For precise dates of guild abolition, see Table 9.4.
29 For surveys of the debate, see Gustafsson 1987, 1–7, 12–13; Hickson and Thompson 1991, 127–31; Bräuer 1993/4; Ogilvie 1997, 308–10, 339–66; Ehmer 1998 [Traditionelles Denken], 19–35; Epstein 1998, 685–86; Pfister 1998, 12–14; Reith 2000, 21–28; Ogilvie 2005; Ogilvie 2007 [Can]; Epstein and Prak 2008; Lucassen, De Moor, and Van Zanden 2008; Ogilvie 2011; Prak and Van Zanden 2013; Davids and De Munck 2014; Ogilvie 2014 [Economics].
30 Quoted in Cole 1939, 419.
31 Becher 1688, 113.
32 Smith 1776, Book IV Chapter VIII, p. 145, para. c27.
33 Quoted in Schelle 1913–23, vol. 5, 159.
34 For representative examples, see Gustafsson 1987; Hickson and Thompson 1991; Putnam et al. 1993, 163–85; Epstein 1998; Reith 2000; and a number of the essays in Epstein et al. 1998; Haupt 2002; and Prak et al. 2006.
35 For examples, see La Force 1965; Walker 1971; Kisch 1989; Rosenband 1997; Wiesner 2000; Trivellato 2006; Van den Heuvel 2007; Hafter 2007; Davids 2008; Lindberg 2008; Lindberg 2009; Boldorf 2009; Caracausi 2014; Horn 2015.
36 Ehmer 2001 [Artisans], 818.
37 Putnam et al. 1993.
38 For the original insight, see Coleman 1988, S104–S110. For a more rigorous development, see Sobel 2002, esp. 151.
39 Coleman 1988, S104–S110.
40 Ehmer 2005, 63.
41 Reininghaus 1981, 66–67; Schulz 1985, 165; Rosser 2015, 119–20.
42 Eulenburg 1893, 282.
43 Quoted in Wilda 1831, 33–34.
44 Sacks 1986, 151.
45 Ogilvie 1997, 316–23, 341.
46 Frick 2013, 390–91.
47 Sobania 1991, 195–96.
48 Quoted in Archer 1991, 116.
49 Maréchal 1953, 13; Vander Linden1896, 1–11; Eulenburg 1893, 305.
50 Gross 1890, I:34–35; Slessarev 1967; Brigden 1984, 94–96; Davies and Saunders 2004, 19 (quotation).
51 Quoted in Rosser 2015, 182.
52 Quoted in Mickwitz 1936, 145.
53 Quoted in Casey 1999, 244.
54 Casey 1999, 115.
55 Allio 1998, 444–55.
56 MacKay 2006, 42 n. 89.
57 Broadhead 1996, 577, 583, 587–88, 591–93, 596.
58 Whyte 1995, 199; MacDonald 2007, 160.
59 Palliser 1979, 152.
60 Archer 1991, 112.
61 Whelan 2012, 36.
62 Kluge 2007, 319.
63 Ogilvie 1997, 341.
64 Reyscher 1858ff., 13:104.
65 Faivre 1949, 58–60.
66 Edgren 1997 [Craftsmen], esp. 144.
67 Comino, Galasso, and Graziano 2017, 3, 20–22.
68 Allio 1998, 444.
69 Allio 1998, 444.
70 Ehmer 2000, 204.
71 Ehmer 2001 [Family], 194; Ehmer 2000, 200.
72 Ehmer 2001 [Family], 193.
73 Ehmer 2001 [Family], 194.
74 Benoit 1986, 389.
75 See, e.g., Lambert and Stabel 2005, 15, 22.
76 Coleman 1988, S104–S110.
77 See, for instance, the tentative account of Chinese guilds in Moll-Murata 2008, 2013; and Indian guilds in Roy 2008.
78 Hetzel 1996, 246. The database also contains some observations that span time-periods up to 1899.
79 Sadly, the database contains no quantitative observations for guilds in Ireland.
80 See the data and documentation at http://www.ggdc.net/maddison/maddison-project/home.htm.
81 Throughout this book, “significant” means that the null hypothesis cannot be rejected at the 0.05 level; “borderline” statistical significance means that the null hypothesis can be rejected at the 0.05 level but not at the 0.10 level.
82 Separate population estimates are not available for England, Scotland, or Estonia, although population estimates for “Great Britain” roughly correspond to England and Scotland combined.
83 See Klein and Ogilvie 2016.
84 The Qualitative Guilds Database can be downloaded from http://www.econ.cam.ac.uk/people-files/faculty/sco2/Ogilvie-Guilds-Qualitative-Database-1-Nov-2018.xlsx, and the Quantitative Guilds Database from http://www.econ.cam.ac.uk/people-files/faculty/sco2/Ogilvie-Guilds-Quantitative-Database-1-Nov-2018.xlsx. These databases consist of all observations that could be located up to October 2017. They make no claim to include all data sources published before that date, and they do not include sources published after it.
85 Ogilvie and Carus 2014, 406, 461–69, 487; Ogilvie 2014 [Economics], 173.
86 Quoted in Schelle 1913–23, vol. 5, 159.
87 Quoted in Mickwitz 1936, 135.
88 Quoted in Lipson 1915, I: 382.
89 Quoted in Kluge 2007, 232.
90 Compiled in Lespinasse and Bonnardot 1879.
91 Quoted in Lipson 1915, I:302.
92 Beier 1683 [1717], 35ff. The quotation is from the 1717 edition.
93 Depping 1837, 165.
94 MacKay 2006, 133.
95 Quoted in Göttmann 1977 [Handwerk], 118.