THE COFFEE COMMODITY CHAIN IN THE WORLD-ECONOMY: 
ARRIGHl'S SYSTEMIC CYCLES AND BRAUD EL'S LAYERS OF ANALYSIS 

John M. Talbot 
Department of Sociology 
Universi1y of the West Indies, Mona 
john.talbot@uwimona.edu.jm 

ABSTRACT 

This article presents a history of coffee in the modern world-economy, w;ing an analytical 
framework synthesized from Arrighi's concept of systemic cycles of accumulation and Braudel's 
notion of three levels of economic analysis: material life, the market economy, and capitalism. It 
takes the commodity chain as the unit of analysis, and argues that this choice helps to illuminate 
the caw;al connections between Braudel 's three layers. The method of incorporated comparison 
is w;ed to compare restructurings of the coffee commodity chain with the restructurings of the 
larger world-economy during each of Arrighi 's systemic cycles. 

INTRODUCTION 

In this paper, I use Arrighi's four "systemic cycles of accumulation" as a framework to describe 
and explain transformations of the world coffee economy, focusing on the use of coffee in 
everyday life and the global trade in coffee.' I link these changes to the recurrent pattern that 
Arrighi identifies for each systemic cycle, moving through alternating periods of "material" and 
"financial" expansion. I attempt to explain how the changes in material life and the market 
economy were shaped by, and in tum helped to shape, the development of each systemic cycle of 
accumulation as a whole. I employ a variation of McMichael's (1990, 1992) method of 
incorporated comparison to compare changes in the structure of the coffee commodity chain with 
changes in the world-economy, because the former is a part of the latter. When a hegemonic 
power restructures the world-economy, it does so by restructuring the commodity chains that 
comprise it. However, each commodity chain also has its own logic and history, which condition 
its further evolution and render it either open, or resistant, to restructuring at particular historical 
conjunctures. Instances where the evolution of the commodity chain does not correspond exactly 
with the evolution of the world-economy may provide important clues about the dynamics 

1 Arrighi describes each systemic cycle as beginning in the financial expansion of the preceding cycle, 
running through the material expansion, and including the subsequent financial expansion. This yields a 
series of overlapping systemic cycles which is probably truer to the development of capitalism as the top 
layer of the system (1994: 214-15). For this analysis, I have dated the beginning of each cycle at the 
beginning of the material expansion to obtain a series of non-overlapping cycles, which makes it easier to 
deal with events in the two lower layers of the system. 

Copyright ©2011, American Sociological Association, Volume XVII, Number 1, Pages 58-88 
ISSN 1076-156X 



59 JOURNAL OF WORLD-SYSTEMS RESEARCH 

underlying the evolution of each. The article concludes with some generalizations drawn from the 
analysis and some reflections on the theoretical framework. 

ANALYTICAL FRAMEWORK 

Giovanni Arrighi's The Long Twentieth Century (1994) stands as a masterpiece of world-systems 
analysis. Arrighi sets out to construct, as he puts it, a "genealogy of modem capitalism ... as a 
succession of systemic cycles of accumulation" (1994: 84). Arrighi's analysis shows how 
capitalism began in the system of northern Italian city-states in the fourteenth century and 
expanded to encompass the entire planet. The development and expansion of capitalism has been 
accomplished through a succession of systemic cycles: the Genoese, Dutch , British , and U.S. 
Each cycle begins as a new hegemonic power, with new organizational capabilities, leads an 
expansion and restructuring of the system. Each cycle has two phases: a material expansion 
followed by a financial expansion. The material expansion begins when the new hegemon 
embarks on a process of restructuring of both the world economy and the interstate system to 
serve its interests; however, these interests are always presented as, and to some extent perceived 
as, the general interests of the system as a whole. During the material expansion, capital is 
invested primarily in the expansion of production and trade in goods. But as the material 
expansion proceeds, it generates more profits than can be profitably re-invested in production and 
trade. As this happens, a growing share of capital is diverted into speculation and financial deals, 
inaugurating the financial expansion. This stimulates competition between states to attract the 
growing volume of mobile capital, causing increasing instability and chaos in the system. At the 
same time, other rising capitalist powers copy the organizational innovation of the hegemonic 
state, undercutting its power and ability to lead. The crisis that follows is eventually resolved by 
the rise of a new hegemonic power, offering a new organizational innovation that initiates a new 
expansion and restructuring of the system. 

Arrighi draws heavily on Fernand Braudel's three-volume magnum opus, Civilization & 
Capitalism J j h -1 ff h Century ( 1981; 1982; 1984). In this work, Braudel conceptualizes the totality 
of economic life as consisting of three intersecting layers, which he calls material life, the market 
economy, and capitalism. The bottom layer in this construction, material life, is described as the 
"infra-economy, the informal other half of economic activity, the world of self-sufficiency and 
barter of goods and services within a very small radius." (1981: 24) The middle layer of the 
market economy consists of all those activities which have traditionally been considered 
economic: exchanges between freely competing buyers and sellers in markets. The top layer, 
capitalism , in Braudel's memorable phrase , is "the zone of the anti-market, where the great 
predators roam and the law of the jungle operates." (1982: 230, quoted in Arrighi 1994: 10) 
However, Arrighi's analysis focuses primarily on this top layer of the three-tiered structure: 

The notion of systemic cycle of accumulation, we have noted, derives directly 
from Braudel's notion of capitalism as the top layer of the hierarchy of the world 
of trade. Our analytical construct, therefore, focuses on that top layer and offers a 
limited view of what goes on in the middle layer of market economy and the 
bottom layer of material life. This is simultaneously the main strength and the 
main weakness of the construct. (24) 



THE COFFEE COMMODITY CHAIN 60 

As Arrighi notes, "we cannot do everything at once." (25) And he has done quite a lot, not only 
setting out a genealogy of capitalism, but also opening up a research program. For, as he notes: 

The logic of the top layer is only relatively autonomous from the logics of the 
lower layers and can be fully understood only in relation to these other logics .... 
[This] requires that we descend again to explore the lower layers of market 
economy and material life with the knowledge and questions brought back from 
the journey into the top layer which this book undertakes. (26) 

In order to advance this research program, it is necessary to understand not only the 
distinctions between these three layers, but also the ways in which they are connected. Brandel 
never presents an extended, systematic discussion of these distinctions and connections; he 
interjects brief sketches of them throughout the three volumes of Civilization and Capitalism. We 
have to attempt to construct a systematic account from the sketches. He starts with the definition 
of material life quoted above, as the informal and subsistence economy. But his discussion in the 
first volume ranges much more broadly than this. Central chapters in this volume discuss food 
and eating habits, beverages and drinking habits, clothing styles, and the construction and 
decoration of housing. He concludes this section by saying that: 

our investigation takes us at this point not simply into the realm of material 
'things', but into a world of 'things and words' - ... languages with everything 
that man contributes or insinuates into them ... And those strange collections of 
commodities, symbols, illusions, fantasms and intellectual schemas that we call 
civilizations must also be invoked at this point. ( 1981: 333) 

Rather than simple, informal, economic activity, then, what he is discussing here might better be 
labeled material civilization, which he seems to use as a synonym for material life. This includes 
all of the material objects of everyday life as well as the cultural meanings attached to them. He 
further clarifies at the beginning of the second volume: 

Between 'material life' (in the sense of an extremely elementary economy) and 
'economic life', the contact surface is not continuous, but takes the form of 
thousands of humble points of intersection: markets, stalls, shops. Each point 
marks a break: on one side is economic life with its commerce, its currencies ... 
on the other 'material life', the non-economy, imprisoned within self-sufficiency. 
The economy begins at the fateful threshold of 'exchange value'. (1982:21) 

He contrasts the mentality of the merchant, who starts with money and buys goods only to sell 
them in order to acquire more money (MGM' ), with the mentality of the peasant, who comes to 
the market with goods to sell for money, but only as a means of acquiring other goods (GMG' ) 
(1982: 62-4). From the peasant's perspective, the transaction closely resembles barter, except that 
it is mediated by money. Thus Brandel's conception of material life includes the common 'things' 
of everyday life that are used or consumed primarily for their use values, even if these things have 



61 JOURNAL OF WORLD-SYSTEMS RESEARCH 

been obtained from the market, as long as the buyer is primarily concerned with their use values 
rather than their exchange values. 

Economic life, or the market, involves the exchange or circulation of goods according to 
set rules between relatively small buyers and sellers: 

The economy, in the sense in which I wish to use the word, was a world of 
transparence and regularity, in which everyone could be sure in advance, with the 
benefit of common experience, how the processes of exchange would operate. 
This was always the case on the town market-place ... And it was the case too, 
even if the distance was greater, for any regular trade of which the origins, 
conditions, routes and markets were fixed. (1982:455) 

He acknowledges that there are two different senses of the word market: "On one hand it is used, 
in a very loose sense, of all types of exchange that go beyond self-sufficiency ... " Thus we have 
urban markets or a gold market (1982:223). "On the other hand, the word 'market' is often 
applied to a rather large broad form of exchange, also known as the market economy ... " 
(1982:223-4). "Historically, one can speak of a market economy, in my view, when prices in the 
markets of a given area fluctuate in unison ... "(1982: 227). This, Brandel thinks, has been the 
case in Europe since the twelfth century. The latter use of the word is what he is referring to when 
he speaks of economic life. Finally, because this market system was freely competitive and run 
by small traders and merchants, only small profits were earned; Brandel refers to it as "micro-
capitalism" ( 1981: 562). 

The top layer ofBraudel's schema is the zone of capitalism: 

[T]he agents and men involved were not the same . . . the rules of the market 
economy regarding, for instance, free competition as described in classical 
economics, although visible at some levels, operated far less frequently in the 
upper sphere, which is that of calculation and speculation. At this level, one 
enters a shadowy zone, a twilight area of activities by the initiated which I 
believe to lie at the very root of what is encompassed by the term capitalism: the 
latter being an accumulation of power ( one that bases exchange on the balance of 
strength, as much as, or more than on the reciprocity of needs) a form of social 
parasitism ... (1982:22) 

Brandel refers to it as the "anti-market" because the laws of the market do not apply here, or they 
apply only rarely. Consequently, this is the sphere in which the large profits are made. 2 A 
defining characteristic of capitalism is its flexibility and adaptability. Opportunities for huge 
profits opened up at different times, in different places, and in different spheres of economic 
activity. Capitalists had to be ready to take advantage of the opportunities when they arose: "the 
characteristic advantage of standing at the commanding heights of the economy . . . consisted 

2 For a detailed discussion of the distinction between the market and capitalism, see Wallerstein (2001: 
207-17). 



THE COFFEE COMMODITY CHAIN 62 

precisely of not having to confine oneself to a single choice, of being able, as today's 
businessmen would put it, to keep one's options open." (1982: 381) During the period from 1400-
1800 that Braudel was analyzing, capitalism's "home ground" was the long-distance trade, 
including the financing of this trade: 

Long-distance trade certainly made super-profits: it was after all based on the 
price differences between two markets very far apart, with supply and demand in 
complete ignorance of each other and brought into contact only by the activities 
of the middleman. . . . [I]f super-profits vanished from one line, it was always 
possible to find them again on another route with different commodities. 
(1982:405) 

Capitalism only occasionally entered into the sphere of production before 1800. However, after 
the Industrial Revolution in the nineteenth century, capitalism found a new home ground in 
industrial production ( 1982: 231-49). 

Since capitalism was based on power, it depended on the state. State power was 
necessary to suspend the rules of the market economy, to guarantee the monopolies, and to 
protect the super-profits of the capitalists. But state power was controlled by the capitalists: "The 
city-states ... were at this time tools completely in the hands of their merchants." (Braudel 1984: 
51) Braudel considers Genoa and Amsterdam to have been city-states; Britain was the first 
territorial state to become a hegemon, but here, "England fell under the domination of its 
merchants fairly soon after the 1688 revolution" (Braudel 1984: 51), that is, well before its rise to 
hegemony. However, Braudel does acknowledge that there is the potential for conflict, because 
states and capitalism do not always have interests in common: 

[T]he state undoubtedly encouraged capitalism and came to its rescue. But the 
formula can be reversed: the state also discouraged capitalism which was capable 
in return of harming the interests of the state .... Capitalist interests, in the past as 
in the present, naturally extend beyond the narrow boundaries of the nation; and 
this prejudices, or at any rate complicates, the dialogue and relationship between 
state and capital. ( 1982: 554) 

If Braudel only sketches the characteristics of, and the distinctions between, his three 
layers of analysis, he has even less to say about the connections between them. As Arrighi notes, 
echoing other critics, Braudel "does indeed describe more than he theorizes." (2001: 120) 
However, it is an exaggeration to say, as Tilly does, of Braudel's three-layer conceptualization: 
"It does not, however, reflect a causal hierarchy. It does not unfold a tight analytical model that 
guides the movement from one analysis to the next." ( 1984: 66) There is a causal analysis in 
Braudel, but like his descriptions of the three layers, it is scattered through his work. It is complex 
and contingent, and maybe not as "tight" as Tilly would like. However, it is not clear whether a 
"tight" causal model can really explain the broad sweep of historical time that both Braudel and 
Arrighi seek to deal with. 

On the one hand, first the market and later capitalism grow out of the base of material 
life. As population increased, a division of labor became necessary, leading to the growth of 
towns and the need to exchange goods. The existence of towns required a supply of food from the 



63 JOURNAL OF WORLD-SYSTEMS RESEARCH 

surrounding countryside, exchanged for goods produced in the towns. Markets were taken over 
by the towns, increased in number and became specialized (1981: 480-1; 1982: 25-44, 600-1). 
Gradually, various forms of private trading developed to evade the tightly controlled town 
markets and make larger speculative profits ( 1982: 45-9). But it was the opportunities for huge 
profits opened up by the long distance trade in the wake of European attempts to conquer the rest 
of the world that led to the birth of capitalism ( 1982:600-1 ). 

On the other hand, once all three layers had been established, influence flowed primarily 
from the top down. The lower level of material life was a world of unthinking routine which was 
highly resistant to change. He calls it "this layer of stagnant history" (1981: 28). As the market 
economy expanded, more and more of the goods which were part of material life passed through 
the market, gradually changing the way people thought about them. A decisive development was 
the creation of labor markets: "The peasant who was paid a wage, ... even if he received so much 
of his pay in kind that he practically never had two coins to rub together at the end of the year, 
had grown accustomed to reckoning in money terms." (1982:59) Thus the market slowly invaded 
and reshaped some of the layer of material life, rolling back its boundaries; but large parts of it 
still remained outside of economic life. However, it was the layer of capitalism that exerted the 
most influence on the other layers. The capitalists: 

Further: 

. . . could manipulate exchange to their advantage and disturb the established 
order .... [T]hey created anomalies, 'zones of turbulence' and conducted their 
affairs in a very individual way. At this exalted level, a few wealthy merchants 
... could throw whole sectors of the European or even world economy into 
confusion, from a distance. ( 1981: 24) 

In a context where other structures were inflexible (those of material life and, no 
less, those of ordinary economic life) capitalism could choose the areas where it 
wished and was able to intervene, and the areas it would leave to their fate, 
rebuilding as it went its own structure from these components, and gradually in 
the process transforming the structures of others. (1981: 562) 

Even the expansion of the market economy was often ( though not always) driven by the top layer. 
During the sixteenth century, fairs spread across Europe. Braudel argues that "progress forward in 
the sixteenth century must have been achieved from above, under the impact of top-level 
circulation of money and credit, from one fair to another. Everything else would have depended 
on these international monetary movements ... " ( 1982: 13 5). 

Braudel does note a couple of exceptions to this top-down causation. First, the layer of material 
life could be a site of active resistance against the forces of capitalism and the market: 

[T]he market economy had . . . been a restrictive order, and, like all restrictive 
orders, . . . it had created an opposition, counter-forces, both above and below 
itself .... [F]ollowing the 1973-4 crisis, we are beginning to see the development 



THE COFFEE COMMODITY CHAIN 64 

of a modem version of the non-market economy: hardly disguised forms of 
barter, the direct exchange of services ... (1981:24-5). 

Nowadays, on the other hand, if there is a substantial sector outside the 
'economy' or outside the market, it is more likely to reflect a refusal from below 
... ( 1982: 28-9). 

Second, economic life is the main source of innovations that are later adopted by capitalism: 

The lowest level, not being paralyzed by the size of its plant or organization, is 
the one readiest to adapt; it is the seedbed of inspiration, improvisation and even 
innovation, although its most brilliant discoveries sooner or later fall into the 
hands of the holders of capital. (1984:631) 

There is a clear causal hierarchy here; capitalism shapes the two lower layers much more than 
they shape capitalism. Further, the market tends to invade and colonize material life more than 
material life alters the functioning of the market. However, this does not preclude some causal 
influence from working in the opposite direction during certain conjunctures. 

In this article, I seek to contribute to the research program set out by Arrighi, by 
venturing down from the heights of capitalism's systemic cycles of accumulation into the aspects 
of material life and economic life connected to the coffee commodity chain. I investigate how the 
coffee commodity chain evolved through Arrighi's four systemic cycles and to explore the 
connections between Braudel's three layers of analysis through this evolution. Coffee is an apt 
commodity for this type of analysis. It was introduced into Western Europe by Venetian traders 
during the Genoese systemic cycle, and has played an important role in the shaping of material 
life, the market economy, and capitalism as a system ever since. The coffee commodity chain is 
the linked sequence of activities involved in growing coffee, processing it, shipping it, roasting it, 
and selling it to consumers. Some of these activities, such as growing coffee or consuming it, may 
be phenomena of material life, while others, such as the transactions that move coffee along the 
chain from the grower to the exporter to the importer to the consumer, are primarily phenomena 
of the market economy. The overall structure and organization of the chain are often decided at 
the level of capitalism. The commodity chain as a unit of analysis thus encompasses activities at 
all three of Braudel's levels, within a structure that links them together so that a change in one 
activity is likely to spark changes in all other linked activities. The commodity chain is therefore 
an ideal unit of analysis with which to explore the connections among Braudel's three levels. 

The introduction of an agricultural commodity into Braudel's conceptualization 
illuminates a fourth layer that serves as the foundation of his three-story house: the ecological 
layer. Although Braudel begins the first volume of the Civilization and Capitalism trilogy by 
introducing the concept of the biological regime, it does not play a major role in his analysis 
thereafter (Moore, 2003:42-5). However, there are several ecological characteristics of coffee 
production that impinge on the market, and through it, on material life and capitalism. First, 
coffee is a tree crop that takes 3-5 years after it is planted to begin producing; this means that 
coffee production responds relatively slowly to price signals emanating from the market 
economy. Thus coffee growers collectively tend to over-plant during periods of high prices, since 
the planting of new trees has no immediate effect on price levels. This over-planting typically 



65 JOURNAL OF WORLD-SYSTEMS RESEARCH 

leads to oversupply and falling prices 3-5 years later. Second, coffee tends to produce cyclically; 
a heavy crop one year is followed by a lighter crop in the following year. These supply 
fluctuations can also cause price fluctuations in the market. In addition, coffee is susceptible to a 
variety of pests and diseases, and it is sensitive to envirornnental variations in rainfall and 
temperature; all of these factors can influence the amount of coffee that a tree will produce. All of 
this means that the amount of coffee produced from year to year and region to region can 
fluctuate rather widely and unpredictably, introducing a measure of uncertainty into the 
calculations of the market economy, and opening opportunities for speculation by capitalists 
(Talbot 2004: 31-5). Capitalism and the market also impinge on the ecological layer. Coffee was 
often planted on forested land cleared by slash and bum techniques; this facilitated erosion and 
rapid exhaustion of soil fertility, driving the growers further into the forest to clear more land to 
meet the demands of the market and the capitalists for more coffee (Tucker 2000). 

THE GENOESE SYSTEMIC CYCLE 

Arrighi's first, or Genoese, systemic cycle of accumulation began in the mid-fifteenth century and 
extended through the 161h century into the early 17th century. It was characterized by a 
relationship of political exchange between Genoese capitalists and Iberian rulers. The Genoese 
controlled European and Atlantic trade and provided the financing, and the Iberians provided 
military power and protection of trade routes, as well as an expanding New World empire from 
which to draw commodities for trade (Arrighi 1994: 109-32). 

During the Genoese cycle, coffee was mainly an Islamic beverage, and its commodity 
chain was located outside the European world-economy. Although the origins of coffee drinking 
are obscure, it is fairly well established that coffee drinking was popular among the Sufis of 
Arabia Felix, now known as Yemen, by the fifteenth century. From there, during the sixteenth 
century, coffee drinking spread north through the Arabian Peninsula and beyond, reaching Cairo 
in the early 1500s and Istanbul around 1550. At the same time, although much less well 
documented, the coffee drinking habit was spreading east through Persia to India and west 
through northern Africa (Topik 2003: 26-7; Ukers 1935: 17-8). At first, the coffee probably came 
from Ethiopia, the place where arabica coffee is believed to have originated, where it was picked 
from wild bushes. As demand for coffee grew, peasant farmers in Yemen began to cultivate the 
crop in the mid sixteenth century (Wrigley 1988: 1-4; Hattox 1985: 12-26; Tuchscherer 2003: 50-
2). Thus, during this time, coffee was an Islamic beverage, a commodity that circulated within a 
world-system separate from the European one (Braudel, 1984: 467-84). 

The early coffee trade centered on the Red Sea port of Mocha, near the main area of 
coffee cultivation; Indian merchants carried it east from there, Egyptian merchants carried it north 
through the Red Sea, and Arab and Turkish merchants carried it overland through the Ottoman 
Empire (Wrigley 1988: 16-7; Hattox 1985:26-8; Tuchscherer 2003: 52-5). Although we know 
very little about these early non-European traders, it is likely that they, like the European long-
distance traders, were motivated by large profits that could be made by carrying coffee from 
Mocha to distant cities where it was in demand (Hattox 1985: 72-3, 79; Tuscherer 2003: 53-4). 
However, the merchants in Mocha who sold the coffee probably also made considerable profits, 
because they enjoyed a monopoly on its supply, which they jealously guarded (Wild 2004: 75-6; 
Wrigley 1988: 38; Tuscherer 2003: 54-5). 



THE COFFEE COMMODITY CHAIN 66 

In those early Islamic cities where coffee drinking became popular, it sparked a change in 
the material lives of the citizens: the rise of the coffeehouse. Hattox (1985) argues that coffee 
drinking spread not only because of the nature of the beverage itself, but because of the social 
rituals that grew up surrounding its consumption. "One went to the coffeehouse not merely 
because one wished to drink coffee. One went to the coffeehouse because one wished to go out, 
to spend the evening in the society of his fellows, to be entertained, to see and be seen." (89-90) 
Hattox argues that this was a new phenomenon in Islamic cultures, where all meals were prepared 
and eaten in the home, and the only public venue available for socializing was the tavern, where 
no devout Muslim could afford to be seen. This social aspect of coffee drinking was also what led 
to periodic attempts by various rulers to ban the beverage as unacceptable under Islamic law; 
what the rulers were really concerned about, Hattox (1985: 91, 100-3) argues, was what those 
gathered at the coffeehouse were talking about, and where that talk might lead. 

Once coffee drinking had become established in cities like Cairo and Istanbul, it was 
encountered by European travelers, who brought back stories of the hot, dark brew consumed by 
the Arabs (Wrigley 1988: 12; Ukers 1935: 18-9). Venetian merchants, who controlled the eastern 
Mediterranean trade (Arrighi 1994: 91, ll5), were probably the first to bring coffee into Western 
Europe, around 1600, from Alexandria or Istanbul. The Dutch first imported coffee from Mocha 
into Amsterdam in 1616 (Wrigley 1988: 16, 24). During the latter half of the seventeenth century, 
coffee drinking would spread through the major urban centers of Europe. 

Coffee thus entered the European world-economy toward the end of the Genoese cycle, 
as one of the many exotic products being brought in by explorers and traders. It was a novelty, 
probably consumed by a few small groups of very wealthy individuals in the major cities. 
However, there was a thriving coffee commodity chain during this cycle, which connected 
production in Ethiopia and Y ernen with consumption located throughout the Islamic world and 
beyond. From what little we know about the chain during this period, it seems probable that its 
growth was driven by changes in material life. The cultural innovation of the coffeehouse led to 
increasing demand for coffee. Once the coffee drinking habit had spread beyond Yemen, coffee 
necessarily became a marketed commodity, since it was not grown in the other places where it 
was consumed. However, there was not yet a market economy in Brandel's sense, with prices 
moving in unison in the different urban coffee markets. This opened the opportunity for large 
profits to be obtained by the Mocha merchants and the long distance traders who served as the 
middlemen, both capitalist in the Braudelian sense of using market power to obtain super-profits. 
Where the European world-system of the Genoese cycle intersected with this non-European trade 
circuit, in the eastern Mediterranean, Europeans "discovered" coffee and began to use it. Its 
consumption would spread rapidly through Europe during the Dutch cycle, and the major world 
powers would begin to restructure the commodity chain and incorporate it into the expanding 
European world-system. 

THE DUTCH SYSTEMIC CYCLE 

Arrighi dates the second, Dutch-led systemic cycle of accumulation from c. 1640, after the United 
Provinces won independence from Spain and used their control over the Baltic trade in grains and 
naval stores to consolidate their position as the hegemonic power in the world-economy. Dutch 
hegemony was again based on an alliance between political/military and economic powers, but 



67 JOURNAL OF WORLD-SYSTEMS RESEARCH 

this time within the territory of the United Provinces. The organizational innovation of the Dutch 
that allowed them to surpass the Genoese was the internalization of the protection costs involved 
in protecting their trade routes. The material expansion of the Dutch cycle lasted through the mid-
l 700s, followed by a financial expansion through the rest of the century. By the 1790s, Britain 
had taken over as the hegemonic power in the world-economy (Arrighi 1994: 132-44). 

Once coffee had been "discovered" during the Genoese cycle, its use spread rapidly 
during the Dutch-led material expansion. One chronology of the first openings of coffeehouses in 
major cities illustrates this: Venice 1645, London 1652, Marseilles 1659, Amsterdam 1663, Paris 
1675, Hamburg 1679, and Vienna 1683 (Burgin 1978, cited in Wrigley 1988: 32). These 
coffeehouses were meeting places for political, economic, and literary elites, but were also 
frequented by not-so-wealthy middle classes and professionals (Wrigley 1988: 24, 28-9; Heise 
1987: 37). As coffee drinking spread through Europe, it also spread to new and larger segments of 
the population. Demand for coffee therefore must have grown fairly rapidly during the Dutch 
period of material expansion. 3 

At the beginning of this period, Yemen was still the only place in the world where coffee 
was grown commercially ( although wild Ethiopian coffee was also being imported into Yemen 
for local consumption and for re-export), and its merchants enjoyed a monopoly (Tuscherer 
2003:54-5; Wild 2004: 75-6). But the growing demand for coffee, and the opportunities to profit 
from its trade, led the Dutch and the French to attempt to break Mocha's monopoly on the trade, 
by creating new sources of supply more directly under their control. 

The first important commercial plantings of coffee outside Yemen were made by the 
Dutch in the 1690s, in Java and Ceylon. 4 During the 1720s, Java eclipsed Mocha as the major 
source of coffee for the Amsterdam market, Europe's most important (Wrigley 1988: 16; Topik 
2003: 28). However, the Dutch strategy was also to focus on trade, not production ( Arrighi 1994: 
150-8). They devised a system to force Javanese peasants to produce coffee by using local rulers 
as their agents. They signed agreements with these rulers obliging them to deliver a set amount of 
coffee to the VOC (the Dutch East Indies Company) at a fixed price. These local rulers then 
forced peasants under their control to grow the coffee (Fernando 2003: 159; Ukers 1935: 40). The 
Dutch also attempted to enhance their trading profits by creating a monopoly of supply. They 
were able to do this forcibly for the spice trade in the East Indies (Arrighi 1994: 150-8), but were 
unable to eliminate Mocha as an alternate source of supply for the coffee trade. However, they 
did try to use their control over the supply from Java to augment their control over the Indian 
Ocean trade, undercutting the Indian merchants who had previously controlled it' (Topik 2003: 
27-8; Wild 2004: 98-100). 

Dutch attempts at controlling the supply of coffee were short-lived. Once the Dutch broke 
the Arab monopoly, another rising mercantilist power, France, also got into the business of 
globalizing coffee cultivation. They introduced coffee to the island of Bourbon (now Reunion) in 
1715, and it rapidly, if briefly, became a major coffee producer. Here coffee was grown directly 

3 This was rapid growth from a very small base, so that even at the end of the Dutch cycle, total 
consumption was fairly low by modem standards, and it was found only among the urban upper and middle 
classes. 
4 Some coffee had probably been planted earlier along the southwest (Malabar) coast of India, but it is 
unclear by whom, and this coffee was not an important factor in world trade at this time. 
5 Even before coffee drinking took off in Europe, the VOC was profiting from the Indian Ocean trade, 
carrying coffee from Mocha to India. 



THE COFFEE COMMODITY CHAIN 68 

by French planters using slave labor (Campbell 2003: 68-9). This exemplified the French and 
British mercantilist strategy, which was soon to overtake the Dutch, of exerting direct control 
over production (Arrighi 1994: 176-7). The labor necessary to pursue this strategy was supplied 
by slavery. 

From these beginnings in Java and Bourbon', coffee quickly spread to the New World, 
but via Europe. Coffee was raised in the Amsterdam and Paris botanical gardens, and seedlings 
from these plants were carried to Surinam by the Dutch in the early 1700s, and to Martinique by 
the French in the 1720s. Surinam and Martinique would be the source for most, if not all, the 
coffee trees that diffused throughout the Americas in subsequent centuries (Wrigley 1988: 40-7; 
Ukers 1935:2-5). From Martinique, coffee was introduced into Saint-Domingue (Haiti), which 
became a major producer in the second half of the century, until the 1791 revolution (Trouillot 
1982: 337-8). The British also introduced it into Jamaica in about 1730 (Ukers 1935: 70), but 
Jamaica only became a major producer, briefly, after a slump in world sugar prices and the arrival 
of Saint-Domingue planters driven out by the revolution, in the early 19th century (Delle 1998: 
50-2). Coffee from Surinam or French Guiana was introduced into Brazil in the 1720s, and coffee 
from these and other sources spread through other Caribbean, Central and South American 
colonies in the latter half of the 18th century (Wrigley 1988: 46-50; Ukers 1935: 5). Latin 
Americans would become the world's major producers only after their countries achieved 
independence in the 1820s. All of these developments during the Dutch cycle brought the coffee 
commodity chain more firmly under European control. Consumption spread through Europe, and 
most production was now within the boundaries of the expanding European world-economy 
(Trouillot 1982: 339-40). But the spread of production itself was also a part of the outward 
expansion of the boundaries of the system, as new areas of the world outside of Europe were 
linked to the core of the system, as suppliers of coffee. 7 

As was the case for the Islamic-controlled commodity chain during the Genoese cycle, 
the main driving force behind the development of this new, European-controlled coffee 
commodity chain was at the level of material life. Just as the coffeehouse filled a social need in 
Islamic societies, so it did in European cities. As coffeehouses spread through Europe in the 
second half of the 17th century, demand increased, but Mocha was still the only source of coffee. 
The long-distance traders were the key connections between these export and import markets. 
European coffeehouse owners had no direct knowledge of coffee production or of the supply 
available in Mocha; and the Arab merchants in Mocha who sold it could only infer information 
about fmal demand for coffee in Europe from the amounts purchased by the traders. Further, 
there were no regular markets for coffee in the early days of this trade. The arrival in Mocha of a 
European trader wanting to buy coffee raised local prices (Wrigley 1988: 16). Conversely, the 
arrival of a new load of coffee in Amsterdam probably depressed local prices (Topik 2003: 29). 

6 The names of these earliest coffee production sites were used to denote types of coffee in the early trade, 
and they have become a part of contemporary coffee lingo. Mocha is still used to denote the type of coffee 
grown in Yemen, and is often applied to coffees with similar taste characteristics. Java became an 
American slang word meaning coffee in general, as well as denoting coffee produced on the island of Java, 
which is a constituent of the oldest and best known coffee blend, Mocha Java. Bourbon today is used to 
denote a specific variety of Arabi ca coffee tree, supposedly a direct descendant of the trees that were 
brought from Bourbon to Martinique by way of Paris. 
7 This process involved other commodities besides coffee; foremost among them sugar, which was the 
main plantation crop of both Saint-Domingue and Jamaica. 



69 JOURNAL OF WORLD-SYSTEMS RESEARCH 

But there was little correlation between prices in Mocha and Amsterdam. This created a situation 
in which the traders, the only ones who had some information about both supply and demand, 
were able to make large profits by maximizing the spread between the purchase price at Mocha 
and the selling price in Amsterdam. 

All of this made the coffee trade part of the "home ground" of capitalism, in Braudel's 
terms. This trade also provided the impetus for capitalism to venture into the reahn of production. 
By gaining control over the supply of coffee, long distance traders could lower the purchase price 
and eliminate the uncertainties associated with buying coffee from Mocha. While the Dutch 
capitalists did not invest directly in production, but worked through Javanese intermediaries, the 
French and British capitalists did move into production, initially in Bourbon and Jamaica, 
respectively. However, this spread of production through the newly colonized areas would not 
have been profitable without expanding European demand. As Europeans gained more control 
over production and the supply of coffee increased, the trade became regularized, competition 
increased, and the rate of profit declined. Topik (2003: 29) argues that the world coffee market 
became more integrated by the second half of the 18th century, and that prices in Amsterdam 
became more stable. In addition, with rising consumption, the volume of coffee flowing into 
Europe increased substantially. All these factors combined meant that the Amsterdam coffee 
market was a more competitive market by the end of the 181h century. In Braudel's terms, then, 
the coffee trade became a part of the market economy. This shift of coffee from the realm of 
capitalism down to the level of the market economy roughly coincides with the transition from 
the material expansion of the Dutch cycle to its financial expansion. 

The social aspects of coffee consumption embodied in the coffeehouses seem to place 
them at the level of material life, according to Braudel's definition. But the relationship between 
capitalism, the market, and material life in the coffeehouse phenomenon was much more 
complex. It illustrates the difficulty of separating the three layers in Braudel's schema, even at 
this relatively early period in the history of the European world-economy. This is best 
documented for the London coffeehouses, which I discuss presently, although similar social 
processes were probably happening in other major European cities at this time as well (Heise 
1987: 130-40; Wrigley 1988: 33; Ukers 1935: 93-100). First, coffee, as a tropical crop, could 
never be a part of Europe's informal, subsistence economy. If the Europeans wanted coffee, they 
had to venture to the tropics to get it. This meant that European coffee drinkers were integrated 
into a world market. Even so, we can still place coffee drinking firmly within the level of material 
life, since the stimulative effect of the brew and the sociability of the coffeehouses were the main 
reasons that people consumed coffee. They did not value it for its exchange value, but for its use 
value. 

Coffeehouses also became meeting places for merchants and others, where important 
business was conducted. Perhaps most famous was Edward Lloyd's coffeehouse, which was 
frequented by merchants engaged in long-distance trade, ship owners, and insurance underwriters. 
The ad hoc deals arranged in this coffeehouse would in time evolve into Lloyd's of London 
(Ukers 1935: 80-1; Wrigley 1988: 28; Schivelbusch 1992: 49-51). Jonathan's and Garraway's 
became meeting places for stock brokers, and the birthplaces of the London Stock Exchange 
(Wrigley 1988: 28). The Jerusalem Coffee House was a gathering place for agents of the East 
India Company (Wild 2004: 86). Heise (1987: 128) argues that "the coffee-house lent major 
impetus to the evolution of the bourgeoisie from a class in itself to a class for itself, and to the 
latter's emancipation from the aristocracy and the clergy." While this may be a bit of hyperbole, it 



THE COFFEE COMMODITY CHAIN 70 

is clear that major institutions of the market economy, as well as strictly capitalist ventures, were 
organized through the coffeehouses. Here we see all three of Braudel's layers interwoven in a 
way that is very difficult to separate, even analytically. When a ship owner and a merchant who 
were planning a voyage to China went to Lloyd's, bought cups of coffee, and sat down with an 
underwriter to negotiate insurance for the trip, as well as to discuss current affairs, was this 
material life, the market, or capitalism? 

The impact of coffeehouses on society went well beyond the economic. Political, literary, 
and scientific elites all had their favorite, "specialized," meeting places. The Royal Society was 
founded in Tillyard's Coffeehouse, which was frequented by scientists such as Edmund Halley 
and Isaac Newton (Wild 2004: 89-90; Wrigley 1988: 25). Richard Steele's Tatler and Joseph 
Addison's Spectator emerged from Button's, one of the leading literary coffeehouses, and they 
gathered material for their journals at various other coffeehouses (Heise 1987: 131; Wrigley 
1988: 29; Ukers 1935: 75-6). Republicans met at the Turk's Head Coffeehouse, while the 
Royalists met at Tillyard's; Whigs met at St James', while Tories met at the Cocoa Tree (Wild 
2004: 88-92; Ukers 1935: 55-6; Wrigley 1988: 30). Coffeehouses became known as "Penny 
Universities," because the price of admission was often one penny, while a cup ( or dish, as it was 
called) of coffee could be had for one or two pennies more. For this price, one could listen to, or 
participate in, discussions of the great ideas of the time, often involving some of the leading 
minds of the time (Ukers 1935: 69-70; Wild 2004: 87). It could be argued that the coffeehouses 
were incubators of what today would be called civil society, a potential counterweight against the 
forces of the market, hence, a part of material life for Braudel. Material life does not just roll back 
and weaken under the onslaught of capitalism and the market economy, but retains the ability to 
rejuvenate and expand in new ways. 

While the development of a European-controlled coffee commodity chain had many 
positive effects on material life in Europe, the effects on the material lives of those peoples 
incorporated into the expanding world-system through coffee production were certainly 
disastrous. The penetration of capitalism into their material lives led to the imposition of coffee 
cultivation in Java, in the more benign case. Javanese peasants surely resented the imposition, but 
little is known about their reactions to it during this early period. The evidence suggests that their 
resistance was mostly passive. Yet the Dutch were able to obtain significant amounts of coffee 
through this system (Fernando 2003). 

In the other early colonies where coffee was grown with slave labor, the effects are 
widely known. Africans were ripped from their old material lives and transplanted to completely 
different ones in plantation-slave economies in Reunion and the Caribbean, and later Brazil, in 
order to produce coffee for the European markets. This was the epitome of capitalism, throwing 
"whole sectors of the ... world economy into confusion, from a distance." (Braudel 1981: 24) 
There were revolts and rebellions against this destruction of material life by capitalist forces, 
most notably the Haitian revolution. Sugar was a more important force behind the development of 
these slave economies and the rebellions against them, but coffee also played an important role. 
Trouillot (1982) argues that it was the introduction of coffee into Haiti that destabilized the sugar-
plantation slavery system and made the revolution possible. 

Coffee consumption appears to have grown rapidly during the material expansion of 
1640-1740. This opened the possibility for super-profits to be earned by capitalists, first by 
controlling trade, and later by controlling production. This led to the global expansion of coffee 
production and the incorporation of new regions into the expanding world-economy. During the 



71 JOURNAL OF WORLD-SYSTEMS RESEARCH 

financial expansion of the late 18th century, this growth appears to have tapered off. Coffee 
consumption had probably saturated the markets among the urban upper and middle classes in 
Europe, and the slowing of growth in demand probably made it less urgent to secure new sources 
of supply. Profitability in the coffee trade was also lower, removing incentives to bring new areas 
into production. Profits continued to be made from the coffee trade, but they were probably 
increasingly invested in other financial dealings. It would take the spread of coffee consumption 
to the working classes during the British cycle to set off a new round of expansion. 

THE BRITISH SYSTEMIC CYCLE 

By the 1790s, Britain had displaced the United Provinces as the hegemonic power in the world-
economy. The organizational innovation that propelled the British forward was the internalization 
of production costs, achieved by bringing the peripheral areas where the raw materials for British 
industry were produced within the expanding boundaries of the British Empire. The material 
expansion of the British cycle stretched through the 1870s, with two distinct phases. For the first 
50 years, the "endless accumulation of capital" was achieved through investment in expanding 
industrial production in Britain. However, beginning in the 1840s, there was a massive export of 
British capital to peripheral areas. This in tum stimulated a massive increase in raw material 
flows back to Britain. The financial expansion began in the Great Depression of 1873-96 and 
ended in the Great Depression of the 1930s (Arrighi 1994: 159-74). Latin America was one of the 
major destinations for the exported British capital, which touched off an export boom of raw 
materials from Latin America to Europe and North America. Coffee was the most important 
export of Latin America during this boom (Topik and Wells 1998: 8; Topik 1998: 38). 

Once again during the British cycle, the uses of coffee in material life seem to have been 
the driving force behind the expansion and reorganization of the coffee commodity chain. Coffee 
consumption spread from the middle and upper classes to the new industrial working classes 
being created by the industrial revolution. Ironically, despite the central role of British capital in 
the Latin American export boom, most of the coffee exports did not go to London. British 
mercantilists and the state had built their economic and military power on the trade of tea from 
China', sugar from the Caribbean, and slaves from Africa to produce the sugar. When the 
Industrial Revolution created a large working class in Britain in the early 191h century, tea with 
sugar provided an important part of their sustenance (Mintz 1985: 110-9). 9 However, the situation 
in the other industrializing core powers was different, and coffee became the beverage that kept 
workers awake for long hours in the factories. Most important among these was the U.S., whose 
break from Britain was symbolized by the Boston Tea Party, confirming its national identity as a 
coffee-drinking country (Ukers 1935: 102-3; Wild 2004: 127-38). During its early 
industrialization, between 1830 and 1900, U.S. per capita coffee consumption rose from three to 

8 The British began planting tea in India only after they had achieved hegemonic status, in the 1830s. And 
although Britain had been a major coffee consuming country in the second half of the seventeenth century, 
by the early eighteenth century, tea had replaced coffee as the hot beverage of choice. This change was 
driven by the East India Company and the state, which both derived far more revenues from tea than from 
coffee (Mintz 1985:110-9; Schivelbusch 1992: 79-84: Ukers 1935: 70). 
9 In a parallel with coffee, this expansion of tea consumption encouraged the British East India Company to 
break China's monopoly on the supply of tea by planting it in India. 



THE COFFEE COMMODITY CHAIN 72 

thirteen pounds annually (Jimenez 1995: 39-40). France and Germany were the major European 
coffee importers (Samper and Fernando 2003: 443). The British still traded in coffee, and London 
was a major coffee port because it was the fmancial center of the world, but it served mainly as an 
entrep6t for the European trade (Topik 2003: 39). 

During the Dutch cycle, coffee had been introduced into most of the major regions of the 
world where it could be grown. During the British cycle, the late entrants surpassed the older, 
established producing regions. The Caribbean had become the most important coffee-growing 
region toward the end of the Dutch cycle, with Saint Domingue being the largest producer up to 
the time of the revolution. After the revolution, the center of Caribbean coffee production briefly 
shifted to Jamaica and Cuba, before Haiti recovered its position as a major producer by the 1820s 
(Samper and Fernando 2003: 428)." Coffee had been introduced into many Latin American 
countries from the Caribbean during the late 18th and early 19th centuries, but had not become a 
major product. The most important early producers were Brazil, Venezuela, and Costa Rica, but 
none were major exporters until the 1820s. Colombia, Guatemala, El Salvador, Nicaragua, and 
Mexico became significant producers toward the end of the 19th century (Roseberry 1995: 6, 11; 
Samper and Fernando 2003: 424-35). It was not until the countries of Latin America won 
independence from Spain and Portugal that the center of world coffee production shifted there. 
Brazil rapidly emerged as the dominant producer. 

The mass consumption of coffee in the U.S. and Europe, and the mass production of 
coffee in Brazil, were mutually reinforcing developments. Brazil's supply of huge volumes of 
relatively cheap coffee was the precondition for the rapid expansion of coffee consumption by the 
working classes. At the same time, Brazil could not have expanded production so rapidly, and 
achieved economies of scale in production and transportation, if not for the growing demand for 
coffee among European and American working classes (Topik 2003: 31-8). While U.S. per capita 
coffee consumption rose from three to thirteen pounds, Brazilian coffee exports increased from 
30,000 to 600,000 metric tons (Samper and Fernando 2003: 432-3). 

Brazilian production was based on slave labor for most of the 19th century. The 
importance of coffee to the economy, and the importance of slave labor to coffee production, may 
have delayed the abolition of slavery, which didn't happen until 1888 (Topik 1998: 43). This was 
one factor in the cheapness of Brazilian coffee. But an equally important factor was the 
availability of huge expanses of virgin land that could be cleared and planted in coffee ( after the 
indigenous inhabitants had been removed). These new plantings were considerably more 
productive than land that had been under coffee cultivation for many years. 11 Here British capital 
played the crucial role, supplying the capital necessary to build an extensive railway network. The 
new railroads opened up virgin lands in the interior to coffee cultivation and made the transport of 

10 Jamaican production declined drastically in the 1830s; Cuban production did the same in the 1840s. 
11 An additional factor was the way Brazilian coffee was harvested and processed. It was harvested by 
stripping the branches of the trees, collecting ripe and umipe cherries together, rather than picking only the 
ripe cherries. This method meant that all the coffee could be picked with one pass through the trees, rather 
than the several passes required by the method of picking only the ripe cherries. It also meant that less 
skilled labor was required. The coffee was processed by the dry method, in which the cherries were dried 
in the sun and then the cherry, parchment and silverskin coatings were all removed in one hulling 
operation. This was cheaper than the wet processing method, in which the fresh cherries were depulped, 
fermented, washed, and then dried, and the parchment and silverskin coatings removed in a later, separate 
hulling operation. The Brazilian method produced coffee of lower quality, but at a significantly lower price. 



73 JOURNAL OF WORLD-SYSTEMS RESEARCH 

massive quantities of coffee to the ports economically feasible. Beginning in the middle of the 
19th century, the steamship revolution also made transportation from Brazilian ports to Europe 
and the U.S. somewhat cheaper, much quicker, and more reliable (Topik 2003: 31-6; 1987: 59-
63). 

The coffee commodity chain shifted geographically in this cycle. The archetypal coffee 
trade route during the Dutch cycle had been Java to Amsterdam; under the British cycle it was 
Brazil to New York. The form of transactions also shifted. Mercantilist transactions that operated 
within imperial spheres gave way to market transactions between U.S. and European trading 
firms and Latin American exporters. 12 It seems that rising hegernons are already testing and 
perfecting their organizational innovations well before they attain hegemonic status. The British 
(and French) were already internalizing production costs through colonial coffee production 
during the Dutch cycle, and during the British cycle, the U.S. was perfecting a new form of 
neocolonial domination of nominally independent nations. Already, there was heavy U.S. and 
European participation in the Latin American export firms (Topik 1998: 58-60). This allowed the 
traders to drive down export prices and maximize the price spreads between export and import 
markets, thus earning large profits. Finally, the volume of coffee flowing through the chain also 
expanded dramatically as coffee went from a luxury item to a mass consumption good. 

The British cycle represents a major change from the Dutch in the relative importance of 
material life, the market, and capitalism. Capitalism moves into new home grounds, the layer of 
the market greatly expands, and consequently, the layer of material life recedes. With the corning 
of the Industrial Revolution, capitalism began to radically restructure the material lives of the vast 
majority of the population that had been confrned to that "layer of stagnant history" (Braudel 
1981:28). As peasants were drawn or driven off the land and into the factories to become the new 
working class, they needed a stimulant to get them up in the morning and keep them awake 
through a long working day. While tea provided that stimulation in Britain, coffee provided it in 
Continental Europe and the U.S .. Consequently, there are important differences between the ways 
that coffee was used by the upper classes in the coffeehouses of the Dutch cycle and by the 
working classes of the British cycle. Coffee began to move out of the coffeehouse and into the 
home (Jimenez 1995: 39-41). As a result, the focus of coffee consumption moved from the social 
rituals surrounding it to the physical effects of the beverage. This changed the way people thought 
about coffee, from a beverage that promoted alertness and mental agility for conversation and 
debate, to a much more instrumental orientation: coffee was an aid that helped them to earn a 
wage. In other words, coffee drinking began to move out of the realm of material life and into the 
realm of the market economy. 

Capitalism had already relinquished the long distance trade in coffee by the frnancial 
expansion of the Dutch cycle, as the trade became more regular and competitive. During the 
British cycle, a few capitalists who managed to gain market power in both exporting and 
importing markets were able to manipulate the markets to earn large profits, but trade was no 
longer the main locus of capitalism. It had been involved in the spread of colonial production 
during the Dutch cycle, but this option also began to decline during the British cycle. Capitalism 
now moved directly into production, in Brazil. The increasing demand for coffee by the growing 
U.S. working class was expressed at the level of the market economy. Brazilian planters read 

12 
Some rnercantilist transactions between European colonial powers and their colonies continued, but they 

were no longer the most important ones. 



THE COFFEE COMMODITY CHAIN 74 

these market signals, realized that there was money to be made from coffee, and greatly expanded 
their production. In the process, they created a demand for British capital goods in the form of 
iron to build the railroads, at a time when British capitalists were looking for investment 
opportunities outside Britain in order to continue their accumulation of capital. I refer to the 
Brazilian planters as capitalists in Braudel's sense of the word, because they based their 
production on slave labor to minimize costs, and because they took over and cleared massive 
tracts of new land to plant coffee. Both required the exercise of power ( in alliance with the state) 
and enabled the planters to earn profits larger than they would have on a competitive market 
(Dean 1969: 6). However, these were peripheral capitalists; they were producing for a world 
market controlled by Brazilian export firms (many foreign controlled) and U.S. and European 
import firms. They could not set their own prices for the coffee, and hence probably did not earn 
super-profits. Further, the financing required for this venture was beyond the capacity of local 
capital and the state, hence the necessity of drawing in British capitalists to help build the 
railroads, and share in the profits. However, as a result of this process, the Brazilian planters were 
able, by the late 1800s, to construct a near-monopoly in coffee production, through which they 
would attempt to exert their market power. 

Once again, these changes at the level of the market economy, as well as at the level of 
capitalism as a system, had impacts on the material lives of the people growing the coffee. In the 
case of Brazil, coffee was grown by slave labor until abolition in 1888, although the importation 
of colonos had begun in the 1870s (Dean 1969: 35-6). Under the colono system, the state 
subsidized the importation of European workers to work the plantations under one-year contracts. 
This resulted in "the only instance in history in which massive numbers of Europeans crossed the 
Atlantic to work on semitropical plantations." (Topik 1998: 45) Since the colonos were free to 
move after the ends of their contracts, and since the rapid expansion of coffee created severe 
labor shortages, the colonos were much better treated than the slaves had been (Topik 1998: 48-
50; Dean 1969: 42). Many colonos worked the plantations for only as long as it took them to 
acquire their own land; thus the colono system led to the growth of a sector of independent, 
small-to-medium size, market-oriented farmers alongside the plantation sector 13 (Font 1990: 11-
22). 

As Brazilian production grew to over 50% of total world production by the 1850s, events 
in the Brazilian coffee regions became more closely linked with developments in the coffee 
markets of the U.S. and Europe. The use of steamships to transport the coffee and the use of 
transatlantic cables to transmit information about coffee supplies and prices made this linkage 
much tighter than it had been during the Dutch cycle (Topik 2003: 32-4). Under these conditions, 
the ecology of coffee production, which caused wide variations in Brazilian production, could 
have effects similar to capitalism, throwing "whole sectors of the . . . world economy into 
confusion, from a distance" (Braudel 1981: 24). The instability of supply helped to make coffee 
an object for speculation by capitalists in the main consuming market, the U.S .. An alliance of 
three large import firms, known as the Trinity Syndicate, manipulated prices through the 1870s 

13 However, the small-to-medium sized farms in Brazil would have been considered large to very large in 
most other coffee growing countries. Font (1990: 17, 282-3) defines smallholders as owners ofless than 
20,000 coffee trees. He estimates that this translates into a farm of less than 60 hectares. And these 
"smallholders" only produced 30% of Brazil's coffee by 1930, so in the late 1800s, coffee was still a 
plantation crop. 



75 JOURNAL OF WORLD-SYSTEMS RESEARCH 

by buying up large amounts of coffee to create artificial shortages. When the syndicate collapsed 
in 1880-81, its losses were variously estimated at between $5 and $10 million; this suggests the 
magnitude of the speculative profits that had been made. This episode led directly to the founding 
of the New York Coffee Exchange, a futures market to help regulate and stabilize the coffee 
market. However, speculators still attempted to manipulate both the Coffee Exchange and the 
market for physical coffee (Jimenez 1995: 42; Pendergrast 1999: 63-9). 

Ecological effects also magnified the rise of Brazil to dominance in coffee production in 
the late 1800s. Coffee leaf rust, a fungus that attacks the trees, spread through Ceylon and Java, 
the second and third largest producers after Brazil, in the 1870s and 1880s (Samper and Fernando 
2003: 424-40; Clarence-Smith 2003: 101-11). In Ceylon, planters gave up on coffee and started 
planting tea or rubber, while some coffee production continued on Java and the other islands of 
Dutch Indonesia (Clarence-Smith 2003: 110-3). Declining Asian production combined with a 
frost in Brazil in 1887 caused a shortage of coffee on world markets and a period of high prices 
through the mid 1890s (Topik 2003: 35; Samper and Fernando 2003: 451). Brazilian growers 
responded to the high prices by planting massive numbers of new coffee trees. The number of 
trees in Sao Paulo state alone increased fourfold in fourteen years (Topik 2003: 35). As the new 
trees began to produce, Brazilian exports began to grow rapidly in the mid 1890s, causing prices 
to fall by 50% between 1895 and 1897 (Samper and Fernando 2003: 451). Some growers realized 
that overproduction was the cause, but others blamed the U.S., British, and German exporters 
who dominated the Brazilian export trade for manipulating prices. Growers and their political 
representatives floated numerous plans to rescue the coffee sector, but the Brazilian state, 
committed to a liberal, non-interventionist stance, refused to intervene (Topik 1987: 62-7). 

Prices had recovered only slightly by 1906, when the forecast of a record crop sent panic 
through the coffee regions ( the actual 1906 crop turned out to be 50% higher than the previous 
record) (Topik 1987: 67). Brazil at this point produced about three-quarters of the world's coffee, 
and the world price was already low due to the persistent overproduction. A crop that large in 
Brazil could have caused the price to fall to near zero, ruining many large Brazilian growers. 
Since the federal state had declined to intervene previously, the state governments of the main 
coffee-producing states decided to go ahead on their own. 14 They agreed to buy surplus coffee 
and hold it off the market to prevent the price from collapsing (Holloway 1975: 44-55; Topik 
1987: 67-8). They could then gradually sell it off in succeeding years in such a way as to not 
undercut the price. 15 However, they needed financial backing in order to buy the coffee and hold 
it until it could be sold. The Rothschilds, the most powerful London financial house and the main 
source of foreign loans for the Brazilian state, refused to get involved because they didn't think 
the plan would work. Herman Sielcken, head of one of the largest U.S. coffee import firms, put 
together the necessary financing. He ended up controlling the surplus coffee and using it to 
squeeze the market to make huge profits. But the intervention worked, in the sense that it did 
prevent the price of coffee from collapsing. This first valorization plan, as it was called in Brazil, 
set the pattern for future interventions by the Brazilian state. Once the Sao Paulo state 

14 Sao Paulo, Minas Gerais, and Rio states originally agreed to participate in the valorization, but ultimately 
only Sao Paulo, which stood to be the main beneficiary in any event, carried it out. Thanks to Steven Topik 
for this clarification. 
15 Because of the two-year bearing cycle, Brazilian production was virtually guaranteed to be significantly 
lower in 1907, so that at least some of the retained surplus could be sold relatively quickly. 



THE COFFEE COMMODITY CHAIN 76 

government had proved that intervention could work, the federal government got involved in 
subsequent valorizations, as did the Rothschilds, and this eventually led to the Permanent Defense 
of Coffee, which institutionalized state regulation of the coffee sector (Holloway 1975: 56-75; 
Topik 1987: 71-92). 

This story illustrates the complexity of the relations between state and capitalism, and 
adds the dimension of core-periphery relations. Peripheral capitalism, in the person of the 
Brazilian planters, faced ruin in the market. They appealed to the state to subvert the operation of 
the market. Sao Paulo state, more directly under the control of the planters, complied, but the 
federal state was drawn in only reluctantly. However, neither the peripheral state nor peripheral 
capitalism could command the funds to finance the operation. They had to rely on capitalists in 
the core. Even they were hesitant to take the risk, but Sielcken and those who were willing made 
the super-profits; all the peripheral capitalists got from the deal was their own survival. After 
Sielcken demonstrated the viability of the scheme, other core capitalists were more willing to 
participate in future valorizations, but they almost certainly earned much lower profits. Sielcken's 
super-profits came not from his investment in the valorization, but from subverting the operation 
of the market - using his control over the valorization coffee to manipulate coffee markets in the 
U.S .. This manipulation forced coffee consumers to pay higher prices, and thus eventually 
attracted the attention of the U.S. state, which forced Sielcken to sell off the remaining 
valorization coffee, but not until 1913 (Holloway 1975: 69-75). 

The U.S. speculators who made super-profits on coffee, represented by the Trinity 
Syndicate and Herman Sielcken, were the capitalist heirs of the long distance traders of the Dutch 
cycle. They were coffee importers, in a position to possess the most up-to-date information about 
production and exports from Brazil, and from other producing regions, information that the 
people who bought coffee from them were probably not privy to. However, they did not earn their 
super-profits by buying and selling coffee on the market, but by manipulating it; that is, they were 
operating at the level of the anti-market, as Braudel would say. Further, this manipulation 
required access to huge amounts of liquid capital, and its growth coincides with the beginning of 
the fmancial expansion of the British cycle, around 1870. The New York Coffee Exchange itself 
was a creature of the fmancial expansion, a new market where those who possessed huge amounts 
of capital could attempt to multiply it through financial dealings. 

Once again during this cycle, it seems that a rapid increase in coffee consumption during 
the material expansion triggered not only an increase in production, but a geographic shift in its 
center. During the Dutch cycle, the center had shifted from Mocha to Asia (Java, Ceylon, India). 
During the transition to the British cycle, it shifted briefly to the Caribbean (Haiti, Jamaica, 
Cuba), and during the British cycle, to Latin America, particularly Brazil. In this case, there was a 
lag in the shift because the British were tea drinkers. Therefore the biggest increase in 
consumption and production came in the second half of the 19th century, as the U.S. working 
class, and its coffee consumption, grew rapidly. This means that the expansion of consumption 
and production continued well into the period of financial expansion, whereas it had probably 
slowed during the fmancial expansion of the Dutch cycle. However, consumption stagnated after 
1900 (Jimenez 1995: 42-5), and probably declined during the world depression of the 1930s. It 
would take the initiation of a new systemic cycle, under U.S. leadership, to revive it. 



77 JOURNAL OF WORLD-SYSTEMS RESEARCH 

THE U.S. SYSTEMIC CYCLE" 

The fourth (U.S.) cycle began in the 1940s, with a very brief period of material expansion lasting 
until the early 1970s, and a financial expansion which may now be experiencing its terminal 
crisis. The organizational innovation of the U.S. cycle was a diversified and integrated national 
economy (as opposed to the British economy's integration with its colonies) and the 
internalization of transaction costs within large vertically integrated transnational corporations 
(TNCs) (Arrighi 1994: 269-300). The U.S. oversaw the decolonization of most of the former 
European colonies and their integration into the capitalist bloc during the period of material 
expansion, and the integrated U.S. economy became the development model for these new 
nations (McMichael 2004: 14-36). 

Brazil's Permanent Defense of Coffee led logically to efforts to enlist major Latin 
American coffee producing countries into efforts to control the supply of coffee on the world 
market. These other Latin American countries had benefited from Brazil's unilateral efforts to 
support the world market price and had expanded their production during the early 1900s, but 
their growing market shares undermined the effectiveness of Brazil's efforts. However, they 
refused to cooperate in supply management, and during the 1930s, the Brazilian state was forced 
to destroy massive amounts of coffee that it had stockpiled to support the price, because that was 
cheaper than continuing to hold it. As war engulfed Europe, Latin American producers began to 
lose access to the European coffee market, leaving them with huge surpluses and threatening a 
collapse of the world market worse than had been anticipated in 1906. U.S. officials feared that 
the economic devastation caused by such a collapse might drive the Latin Americans, particularly 
Brazil, into alliances with the Axis powers. They proposed an Inter-American Coffee Agreement. 
Each Latin American country would be given an export quota for the U.S. market. This would 
control the flow of coffee into the U.S. market and head off potential cutthroat competition that 
could lead to a price collapse. Producing countries would receive reasonable prices for the coffee 
they exported, and they could then afford to store their surplus coffee and hold it off the market to 
support the price. This agreement would provide a model for future coffee agreements. In this 
case, we have an alliance of the U.S. and peripheral states acting to regulate the market, in the 
geopolitical interest of the U.S., but also in the long-term interests of the U.S. coffee trading and 
roasting companies, which would be assured of reasonably stable future supplies and prices. 
Coffee growers and peripheral capitalists who controlled coffee processing and export also 
benefited. 

After World War II, as the European market again became accessible, and as coffee 
consumption increased after wartime controls were lifted in the U.S., world demand for coffee 
increased rapidly, and shortages developed. There was also a spread of coffee consumption 
beyond the traditional markets of the U.S. and Europe, which further increased world demand for 
coffee." One impact of the U.S. occupation of Japan was a rapid increase in coffee consumption 

16 Unless otherwise noted, all material in this section is drawn from Talbot (2004). 
17 US. coffee consumption actually began to decline in the 1960s, and coffee was surpassed by soft drinks 
as the favorite US. beverage in the late 1970s. This decline was only arrested by the specialty coffee 
revolution that began in the 1980s, although the decline leveled off in the 1990s and consumption has 
increased slightly in the early 2000s. However, the decline in US. consumption has been more than offset 
by increases elsewhere. 



THE COFFEE COMMODITY CHAIN 78 

there. As living standards and per capita incomes began to rise in parts of what was now known 
as the Third World, middle classes began to take up the coffee habit. Latin American non-
producers such as Argentina and Chile, North African and Middle Eastern countries, and 
Southeast Asia all increased their shares of world coffee imports. And consumption also 
increased in Eastern Europe and the USSR. In many cases, coffee consumption served a 
particular function in the material lives of these new consumers: status symbol. It was a sign that 
their countries were developing and modernizing, that they now had enough disposable income to 
imitate the lifestyles of people in the world's richest, most powerful country. But increasingly, 
under the dominance of U.S. capitalism, embodied in the TNCs, lifestyles themselves became 
objects of consumption, that is, part of the market economy rather than material life. 

The result was a new expansion and relocation of coffee production. The new area of 
expansion was primarily in Africa, and to a lesser extent in Asia. Two different forces drove this 
relocation. On the one hand, there was a rapidly growing demand for instant coffee, which could 
be most profitably produced with robusta, the coffee variety grown in West Africa and most 
Asian producing countries. 18 On the other hand, the European powers and the U.S., as 
decolonization began to seem inevitable, were seeking to create new capitalist-oriented economic 
activities, or expand existing ones, to bring a measure of prosperity to the colonies and give them 
an economic stake in joining the "free" world. A number of African countries were already minor 
coffee producers, and more had the right ecological conditions for its production. With shortages 
of coffee developing, and demand for robusta coffee increasing, expansion of robusta coffee 
production in these African colonies seemed a logical choice. 

The geography of the coffee commodity chain thus became much more complex than it 
had been in previous cycles. Where the archetypal coffee trade route during the British cycle had 
been Brazil to New York, there were now multiple routes, defined by (neo)colonial relations: 
Latin America to New York; Africa to London, Le Havre, and Hamburg; and, a bit later, 
Indonesia to Tokyo. The form of transactions that moved coffee along the chain remained market 
based as they had been during the previous cycle. While the U.S. organizational innovation 
involved vertically integrated TN Cs, this was not characteristic of the coffee chain. The TN Cs 
that came to dominate the coffee chain were large coffee roasting companies (who also produced 
instant coffee). They bought their green coffee from importers, manufactured it, and sold it to 
retailers. There were several reasons for the absence of vertical integration: the risks involved in 
coffee growing, the predominance of small coffee growers, the control exercised over coffee 
production and export by peripheral states, the collective action of the producing states, and the 
prior existence ofTNC importing companies with expertise in handling coffee. 

As Arrighi notes, the U.S. TNCs expanded internationally as soon as they consolidated 
control over the national market (Arrighi 1994: 294-95). But this happened relatively late for 
coffee. At the beginning of the cycle, the one transnational coffee company was Nestle, a Swiss 

18 Latin America produced arabicas almost exclusively at this time, although a number of Latin American 
countries later began to grow robusta. East African countries in the area where coffee originated (Ethiopia, 
Kenya, Tanzania) produced arabica coffee. Robusta coffee was native to West and Central Africa, but was 
not widely commercially cultivated there until the mid-1900s. The only significant Asian coffee producers 
at this time were India, which produced a mix of arabica and robusta, and Indonesia, which produced 
mainly robusta. Java, the original European coffee outpost, began as a producer of arabica. But after the 
coffee plantations were devastated by the coffee leaf rust in the late 1800s, the Javanese switched to 
robusta, which was resistant to the leaf rust. 



79 JOURNAL OF WORLD-SYSTEMS RESEARCH 

company. The major U.S. companies did not really consolidate control over the national market 
until the 1970s, and during the 1980s, they began to go global, quickly gaining dominant 
positions in the major consuming markets of the U.S., Europe, and Japan. The market power 
resulting from this dominance enabled the TNC roasters to get the upper hand on the trading 
companies and squeeze their profit margins to the minimum. This was part of a larger process of 
concentration of commodities trading companies, in which the companies that survived switched 
from specialization in one commodity to dealing in many different ones (Chalmin 1987). 

For a while, during the material expansion and stretching into the beginning of the 
financial expansion, the power of the TN Cs was checked to some extent by the collective power 
of coffee exporting states. A series of International Coffee Agreements (ICAs) between all major 
producing states and most major consuming states regulated the world market through a system 
of export quotas similar to what had existed during World War II. The impetus for these 
agreements came from the interaction of increasing demand and the ecology of coffee production. 
As world demand for coffee grew rapidly after World War II, and shortages developed, world 
market prices spiked in 1954, stimulating a huge increase in planting of coffee, particularly in 
Africa and Asia. Because of the nature of coffee as a tree crop, the output of these new trees did 
not begin to reach the market until the late 1950s; thus there was over-planting. As the new coffee 
began to reach the market, prices fell, coffee growers turned to their states for protection, and the 
major Latin American coffee producers started to make plans to hold coffee off the market in 
order to hold the price up. 

In this context, U.S. geopolitical interests dictated cooperation with the plan. The U.S. 
state had been shocked by the Cuban revolution, and was worried about the spread of 
communism through Latin America, which despite the spread of coffee production to Africa and 
Asia, was still the source of 70% of the world's coffee. At the same time, the European colonial 
powers were worried by the growing militancy of the anti-colonial movements in Africa. Under 
these conditions, the states of the major consuming countries were willing to enter into an 
agreement with peripheral states to regulate the world coffee market and create a price floor. This 
was seen as a disguised form of economic aid to the Third World to stop the spread of 
communism and keep the countries in the capitalist bloc. The series ofICAs regulated the world 
coffee market for most of the period 1962-1989, assuring coffee growers of reasonable prices for 
their coffee despite the growing dominance of TNCs over the chain. The export quotas of the 
ICAs also required coffee producing states to manage their coffee sectors to control exports. The 
ICAs were thus the culmination of a process of state intervention begun by Brazil in 1906. As 
was the case with the Inter-American Coffee Agreement, the geopolitical interest of the U.S. state 
and the long-term interests of U.S. capital (the coffee TNCs) coincided. Now, however, the 
peripheral states had a much more direct interest, because they regulated their domestic coffee 
sectors and obtained significant incomes from coffee exports, sometimes at the expense of their 
own coffee growers. 

By 1989, many of these interests had changed. Under the pressure of the IMF /World 
Bank structural adjustment programs, peripheral states were seeking to expand their exports, 
creating a structural condition of oversupply of coffee. Coffee TNCs began to feel that the ICA 
was preventing them from getting the best coffee at the cheapest price. And with the Soviet bloc 
collapsing, the U.S. state no longer saw such strong geopolitical reasons for supporting the price 
of coffee. Efforts to renew the ICA failed. Under the new neoliberal orthodoxy, the capacities of 
most coffee producing states to manage their coffee sectors and intervene in the world coffee 



THE COFFEE COMMODITY CHAIN 80 

market were quickly dismantled. World market prices crashed, and since 1989, coffee prices have 
been at historic lows, with the exception of a brief price spike in 1994-97. 19 Growers have been 
pressuring their states for protection from the low prices, but under current conditions, in which 
each country must compete for niches in the world market, the coffee producing states have been 
unable to recover the degree of solidarity that allowed them to cooperate in the regulation of the 
market during the 1960s and 1970s. Meanwhile, the tremendous market power of the TNC 
roasters has allowed them to maintain or even increase the wholesale prices they charge for their 
roasted coffee, even as world market prices for green coffee have fallen to historic lows. This 
price spread is one major source of capitalist super-profits in the U.S. cycle. 

The motive force behind the relocation of coffee production and the reorganization of the 
coffee commodity chain seems to have gradually shifted from the level of material life to the 
level of capitalism. During the Genoese and Dutch cycles, the uses of coffee in material life - the 
coffeehouse phenomenon - provided the major impetus. During the British cycle, the spread of 
consumption to the working classes was again the motive force, but behind this spread of 
consumption was the capitalist reorganization of the material lives of the mass of the population. 
And this prompted the working classes themselves to take a more market-oriented view of coffee 
consumption. During the U.S. cycle, although the spread of coffee consumption to new 
consumers again played an important role, the increasing demand for coffee was itself largely a 
creation of the advertising and demand-creation strategies of the TNCs 

First, the demand for the new product, instant coffee, after the war, was largely a creation 
of big capital and the U.S. state. Brazilian coffee officials, seeking new ways to use the surplus 
coffee that was piling up in Brazil during the 1930s, had consulted with Nestle technicians, who 
modified the spray drying machinery they used to produce powdered milk, and produced instant 
coffee. Instant coffee had been produced and sold before, but it had not been very good. This new 
process could produce large quantities of instant coffee relatively cheaply, which were superior in 
taste to previous products. Before Nestle could commercialize the new process, the war 
intervened. But during the war, the U.S. government added instant coffee to the rations provided 
to its troops, simultaneously creating a huge new demand and bringing into existence a new 
industry to meet it. After the war, people who had gotten used to the taste and convenience of 
instant coffee in the armed services formed a ready-made market, and the new instant coffee 
industry advertised vigorously to spread the demand. Second, even the spread of the coffee habit 
to new consumers outside of the U.S. and Europe was driven much more by the advertising 
campaigns of the globalizing TNCs than had ever been the case in previous cycles. They were the 
ones spreading what Leslie Sklair (2002) calls the culture-ideology of consumerism: you are what 
you consume. Coffee became one element of a lifestyle that had to be purchased in the market, 
necessitating increased hours and intensity of work to enable people to purchase their desired 
lifestyle. 

The reorganization of the coffee chain in response to these new increases in demand once 
again had impacts on the material lives of the people who grew the coffee. One major change 

19 Coffee prices began to rise slowly in about 2006, and by October 2010 had reached levels not seen since 
the end of the price spike in 1997. This was a result of prolonged low prices since 1997, causing many 
small growers to lose or abandon their land, or pull out their coffee trees and plant other crops. Under 
present conditions, however, this period of higher prices is likely to also be short-lived, as the rising prices 
will stimulate new planting, causing a new oversupply in a few years. 



81 JOURNAL OF WORLD-SYSTEMS RESEARCH 

grew out of the transition from colonial production to production by independent nations, which 
began with growing Latin American production in the 1820s, and continued with the 
decolonization of Africa and Asia that peaked in the 1960s. This was a shift from plantation 
production to smallholder production. Coffee production is very labor intensive, with few 
economies of scale; thus small-scale producers can be just as efficient as large-scale ones. When 
slaves were freed, or when peasants were released from their obligations to labor on the 
plantations, they much preferred to grow their own coffee on their own land wherever possible. 
Thus by the end of the material expansion of the U.S. cycle, most of the world's coffee was being 
grown by small peasant producers, or by small-to-medium-sized, market-oriented farmers. Coffee 
production was clearly established as part of the economic life of the producers; it was grown for 
sale on the market. This shift had two important implications. First, swings in coffee prices had 
enormous implications for the economies of coffee producing countries, because so many small 
rural producers were involved. Second, because of this, when prices fell and small producers 
began to organize to demand protection from the destructive forces of the market, their states 
were likely to be responsive. This was the dynamic behind the ICAs, but after structural 
adjustment, peripheral states lost the capacity to respond, even if they wanted to. 

During the financial expansion, and particularly since the end of regulation in 1989, 
prices have become more unstable. Coffee consumption and world demand for coffee continue to 
grow slowly. But production, now uncontrolled and driven by pressures from the IMF/World 
Bank debt regime (McMichael 2004), has generally exceeded demand. World market prices have 
been at historic lows, despite brief price spikes, but retail prices in the major consuming markets 
are high. As the TNC roasters reap enormous profits from this price gap, they cannot profitably 
reinvest them in the coffee trade. One symptom of the financial expansion has been the explosion 
of trading on the futures markets, and the introduction of new financial instruments, options on 
futures contracts. The TNC roasters as well as the TNC commodities traders can now reinvest 
some of their profits in these financial instruments. Further, they can coordinate their purchases of 
physical coffee and of fmancial instruments based on coffee, to manipulate markets and further 
increase their profits. But much of the increased trading volume on the futures markets has been 
driven by the involvement of speculators, particularly institutional speculators such as 
commodities funds. As the volume of futures trading has grown, the futures market prices for 
coffee have become the benchmarks used to set the prices used in the trade of physical coffee, the 
tail wagging the dog. These developments have greatly increased the price instability. TN Cs, with 
their financial resources and global information networks, can profit from this instability, and this 
is the other major source of capitalist super-profits. Small coffee growers, unprotected by their 
states, lose. Similar trends can be seen across many other sectors of the world economy. 20 The 
result is a growing polarization of incomes and wealth between countries and regions of the 
world, leading to growing political instability. 

At the opening of the 21' 1 century, there are signs that a counterrnovement against the 
hegemony of the U.S. and its TNCs is beginning to take shape. Some of the signs are the growing 

2° Coffee prices were also influenced by the commodity price bubble that caused rapid increases in food 
prices and then equally rapid price declines during 2008. While there has been much debate over the causes 
of this bubble, most analysts agree that massive flows of speculative capital into commodity futures and 
options, and creation of even more complex financial derivatives based on commodities, played an 
important role. 



THE COFFEE COMMODITY CHAIN 82 

protests against the World Bank/IMF, the protests that have dogged the WTO and slowed 
progress of the millennium round of negotiations since Seattle in 1999, and the World Social 
Forum. In coffee, the most visible sign is the fair trade movement. It is no longer simply a 
movement of coffee growers organizing for protection against the penetration of market forces, 
but a movement combining coffee growers, consumers, NGOs, and small coffee trading and 
roasting companies. Their analysis is that the "free" trading system that has been established by 
goverrnnents, primarily the U.S., in alliance with the TNCs, is inherently unfair. That is, instead 
of focusing their attacks on market forces, they are going after the top layer of the system. And 
although coffee is the main fair trade commodity, the movement is expanding to other tropical 
products such as chocolate and bananas. These movements, as well as the growing political 
instability, may be part of the increasing systemic chaos that signals the corning end of the U.S. 
cycle, if Arrighi's analysis is correct. 

CONCLUSION 

I conclude by drawing out some of the implications of this analysis for Arrighi's systemic cycles 
and their articulation with Brandel's layers of analysis. This excursion into the lower layers of 
Brandel's schema suggests some additional mechanisms underlying the systemic cycles, and it 
also suggests some additional factors that merit consideration. Similarly, viewing Brandel's three 
layers in the context of Arrighi's systemic cycles also suggests new forms of connection between 
the layers. 

Coffee consumption, and therefore the demand for coffee, seems to increase cyclically. It 
grows more rapidly during material expansions, and then slows or stagnates during frnancial 
expansions. This makes sense at two different levels. During material expansions, the production 
and trade of most commodities grows. More people are drawn into the market economy, and so 
more people purchase commodities, such as coffee. More generally, the material expansion is a 
period when commodity production is reorganized on a new and expanded basis. New classes 
emerge, some classes that previously existed grow in size, while still others wither or disappear. 
These new or growing classes are a major source of new demand. The cultural meanings of 
coffee to these classes in the particular historical periods play an important role in 
stimulating demand. During the Genoese cycle, coffee consumption was confined to small 
groups of elites in Europe. With the beginning of the Dutch cycle, the huge expansion in the 
volume and distance of trade brought an increase in the number of merchants and businessmen, as 
well as professionals to serve them, such as lawyers and accountants. These were precisely the 
classes that frequented the early European coffee houses. The British cycle brought the industrial 
revolution and the rise of the working class, and the expansion of coffee consumption during this 
period was among this class (although in the U.S. and not Britain). This generalization may not fit 
the U.S. cycle as well. The U.S. cycle did not really bring a new class into existence in the same 
way as the Dutch and British cycles. Part of the expansion in coffee consumption during the U.S. 
cycle resulted from the rise of professional and middle classes in new countries, outside the core 
of the world-economy. But another part was also created by the advertising efforts of the 
transnational corporations. The massive nature of this demand-creation may be a defrning feature 
of the U.S. cycle. 



83 JOURNAL OF WORLD-SYSTEMS RESEARCH 

During financial expansions, production and trade may continue to grow, but at a slower 
rate, as capital is diverted into financial deals. Coffee consumption may continue to increase, but 
slowly. More generally, during the course of the material expansion, the coffee habit is probably 
taken up by most of the members of the new class that are predisposed towards it. This means 
that once the material expansion runs its course, there is little room for the further spread of 
consumption. The major exception to this generalization is the continued expansion of coffee 
consumption in the U.S. well into the period of financial expansion during the British cycle. This 
can be explained by the different timing of U.S. industrialization, and by the fact that the U.S. 
was the major coffee consuming market, while the British Empire had been built partly on tea. 

While consumption and production are growing during the material expansion, prices 
may be more stable than they are during the financial expansion. This seems to be the case for the 
last two cycles. During the Dutch cycle, price stability increased with the onset of the financial 
expansion, but this was due to the establishment of a routinized world market for coffee. Price 
instability and stagnating demand in the financial expansion reinforce the movement of capital 
away from production and trade and into financial deals and speculation, as the era's big 
capitalists seek to capitalize on the instability. Thus the New York coffee futures market was 
founded during the British fmancial expansion: and new speculative instruments were introduced, 
and trading volumes soared, during the U.S. financial expansion. Further, the Brazilians may have 
been able to pull off their interventions in the market in the early 20th century in part because 
there was a significant volume of mobile capital seeking financial deals, which was available to 
finance the valorizations. 

The increased coffee demand that accompanies material expansion provides the impetus 
for a systemwide restructuring of the coffee commodity chain. This is part of the overall 
reorganization of the world-economy undertaken by the new hegemonic power. This 
reorganization involves the relocation of the main locus of production, the reorganization of 
production systems, changes in the types of transactions that move coffee along the chain, and 
increases in the volumes of coffee flowing through the chain. The Dutch began this process with 
the establishment of Java as a new source of supply, and they were quickly emulated by their 
rivals, most notably France. However, the Dutch were content to control the trade in coffee, while 
the French and British pioneered the control over production by their colonists, and the use of 
slave labor. What had been market transactions between European traders and the Mocha coffee 
merchants located outside the European world-system were shifted to mercantilist transactions 
within an expanding system. During the British cycle, the center of coffee production moved from 
Asia to Latin America, particularly Brazil. Here the British interest was not in controlling the 
trade of coffee, but in seeking profitable investments for exported capital. These investment 
opportunities were found in the building of rail and port facilities in Latin America that initiated 
the export boom of 1850-1930. Production systems began to shift (although Brazil lagged behind) 
from those based on slavery to those based on small farmers. This shift was not completed until 
decolonization was accomplished during the U.S. cycle. The trade also shifted from a mercantilist 
trade, based on colonial production, to a market-based trade between nominally independent 
countries and firms. During the U.S. cycle, production was once again relocated, this time toward 
Africa and Asia, although the center remained in Latin America. This was in line with U.S. 
interests in overseeing decolonization in the context of the Cold War, and the competition with 
the Soviet bloc for the "hearts and minds" of the newly independent nations. Production systems 
based on small farmers were regulated by the peripheral states, until the 1990s, and market 



THE COFFEE COMMODITY CHAIN 84 

transactions predominated along the chain. Huge (although not vertically integrated) transnational 
firms came to control the chain through their control over the major consuming markets. 

Here we see a complex interaction between the reorganization of the world-economy as a 
whole and the reorganization of a specific commodity chain. There is no easy one-to-one 
correspondence; the organization of the coffee chain enjoys a degree of relative autonomy, as 
Arrighi ( 1994: 26) puts it, from the organization of the world-economy. In several instances, the 
reorganization of the coffee chain seems to be in the lead, allowing rising hegemons the 
opportunity to test and perfect their organizational innovations. Thus, the French and British 
pioneered colonial production during the Dutch cycle, and the early decolonization of Latin 
America allowed for the development of market transactions between coffee exporters and 
importers during the mercantilist British cycle. It raises the question: which organizational form 
currently existing in the coffee commodity chain is the new organizational innovation that will 
propel the next hegemon to power? 

In other instances, changes in the coffee chain seem to lag behind developments in the 
world-economy. Thus, a production system based on slavery persisted in Brazil after it had been 
abolished elsewhere, and colonial production persisted into the U.S. cycle, particularly in Africa. 
These holdovers are not too surprising. All social institutions tend to take on a life of their own 
and to resist change. Other discrepancies may have to do with specific interests of the hegemonic 
power, the legacy of previous reorganizations, or the nature of the specific commodity. British 
world power, for example, rested partly on tea rather than coffee, so the U.S. played a key role in 
reorganizing the chain during the British cycle. And because of the nature of coffee, and the 
legacy of previous commodity chain reorganizations, the transnational firms that came to 
dominate the coffee chain during the U.S. cycle were not vertically integrated. While Arrighi's 
periodization of capitalism is probably correct at the "top" level of the system, if we follow his 
research program and descend to the lower levels to look at specific commodities, we are likely to 
find slightly different timing across the systemic diversity of commodity chains. This is not 
surprising, as Arrighi's systemic cycles are general patterns that have been abstracted from the 
specific structural changes that occur across each of the commodity chains that comprise the 
world-economy. 

This excursion into the lower layers of material life and the market economy therefore 
serves to further elaborate some of the mechanisms underlying Arrighi's systemic cycles, but also 
adds complexity and qualification. The interactions among Brandel's three layers are complex, 
and they are not determined solely by the structural logic of the top layer; however, this top layer 
does powerfully shape the interactions of the two lower layers. For example, I have argued that 
the spread of coffee consumption at the beginning of the material expansions of the most recent 
three cycles provided the impetus for the geographical and institutional restructuring of the chain. 
However, it was primarily the structural logic of the top layer that determined where production 
would be relocated, and how the chain would be reorganized. In addition, the periodic shortages 
and surpluses that send shocks through the chain are primarily market phenomena. By disrupting 
the flow of commodities and money along the chain, they create opportunities for subordinate 
players to struggle to change the structure of the commodity chain, and ultimately, the structure 
of the top layer of the system. At the same time, the hegemonic power retains the ability to at 
least shape the outcome of this struggle, if not to entirely defeat it. 

The periodic shortages and surpluses also point to the role of a fourth, ecological, layer 
that should be added to Brandel's three-layer conceptualization. While they are phenomena of the 



85 JOURNAL OF WORLD-SYSTEMS RESEARCH 

market, they arise out of the ecology of the coffee tree: its susceptibility to frosts, pests, and 
diseases; its cyclical bearing pattern, and the lag between planting and production of new coffee. 
In this analysis, I have only been able to sketch in a few of the connections between this layer and 
the other three; much more work is needed to fully integrate it into the analysis. 

This analysis raises another important addition to Arrighi's theory: the concept of 
peripheral capitalism. I have argued that actors located in the periphery of the system can develop 
the same capitalist orientation as the big capitalists located in the core. They can attempt to use 
their market power, in alliance with their states, to subvert the operation of the market in order to 
earn super-profits. Because of their location in the periphery, and because their states are not as 
powerful as the core states, they can rarely hope to compete with the core capitalists and attain the 
same level of super-profits realized by them. They may be forced into alliances with core 
capitalists, who are able to appropriate a share of these profits, but they still probably earn higher 
profits than would be available to them through participation at the level of the market economy. 
And they may play an important role in the evolution of the systemic cycles, as did the Brazilian 
capitalists of the British cycle. It would be interesting to see if there were similar instances of 
peripheral capitalism related to other commodities during earlier systemic cycles. Peripheral 
capitalists, and their states, have certainly played a role, albeit subordinate, in shaping the 
evolution of the U.S. cycle 

Turning now to the implications of this analysis for the causal connections between 
Braudel's three layers, one thing that is immediately apparent is the growing strength and reach of 
capitalism. This is seen in its ability to penetrate and reorganize the two lower layers. During the 
Genoese and Dutch cycles, the vast majority of the population was confined to the lowest layer of 
material life. Capitalism began to incorporate them into the market economy in the British cycle, 
a process that was nearly completed during the U.S. cycle. As this process proceeded, the layer of 
material life shrank correspondingly, and became increasingly shaped by the forces of capitalism, 
which began to influence the consumption choices that people made in the market, commodifying 
even the process of identity formation. Ironically, Braudel, whose analysis focuses on the period 
of the Genoese and Dutch cycles, sees most of the influence flowing from the top layer of 
capitalism down to the other two layers. But this may have been precisely the time when material 
life could actually exert some independent effect on the top layer of capitalism. I have argued that 
it was changes at the level of material life - the coffeehouse phenomenon - that opened 
possibilities for capitalism to reorganize the coffee commodity chain in order to earn super-profits 
during the first two cycles. The subsequent increases in coffee consumption during the British 
and especially the U.S. material expansions, in contrast, were more driven by the independent 
operation of capitalism as the top layer of the world-system. 

However, we should not conclude that material life only shrinks and loses power over 
time; it can also rejuvenate and grow in new directions. Thus, revolts of enslaved people 
against their intolerable material lives helped to force the end of slavery, and people's 
general preference for growing coffee on their own small plots rather than working on 
plantations led to a reorganization of production on a peasant smallholder basis. I have 
also argued that what we now call civil society was, at least in part, an outgrowth of material life 
in the coffeehouses, and it has grown into a counterweight against the worst excesses of 
capitalism. Further, as Braudel notes, the "new" informal economy which began to grow during 
the fmancial expansion of the U.S. cycle can be seen as a sign ofresistance against the incursions 
of market forces into material life (McMichael 2000: 217-20). 



THE COFFEE COMMODITY CHAIN 86 

Once again, it is important to distinguish between core and periphery. While the top layer 
of capitalism may dominate material life in the core, the process lags behind in the periphery. 
Large segments of the populations there have still not been completely incorporated into the 
market economy. This means that the layer of material life in the periphery may still retain some 
capacity to influence the unfolding of the systemic cycles at the top layer of the system. In fact, a 
major impetus behind the fair trade movement in coffee, and behind the movement for 
globalization from below more generally, is the struggle by the poor and marginalized peoples of 
the periphery to prevent their material lives from being colonized and reshaped by the forces of 
capitalism. 

This study of the coffee commodity chain has focused on one small segment of the 
world-economy. Many more studies of this nature will be needed before we can begin to get a 
complete picture of the interactions among all three ( or four) of Brandel's layers and the ways in 
which they have shaped the evolution of the systemic cycles of accumulation. The commodity 
chain as unit of analysis and the method of incorporated comparison seem to be useful tools for 
this type of investigation. The breadth of the research program that flows from Arrighi's The 
Long Twentieth Century is one measure of its enduring importance. 

REFERENCES 

Arrighi, Giovanni. 1994. The Long Twentieth Century: Money, Power, and the Origins of Our 
Times London: Verso. 

__ . "Brandel, Capitalism, and the New Economic Sociology" Review 24(1): 107-23. 
Brandel, Femand. 1981. The Structures of Everyday Life: Civilization & Capitalism 151h-lffh 

Century, Volume 1 New York: Harper and Row. Translated from the French by S. 
Reynolds. 

__ . 1982. The Wheels of Commerce: Civilization & Capitalism l 51h-l ffh Century, Volume 2 
New York: Harper and Row. Translated from the French by S. Reynolds. 

__ . 1984. The Perspective of the World: Civilization & Capitalism 151h-lffh Century, Volume 
3 New York: Harper and Row. Translated from the French by S. Reynolds. 

Campbell, Guryn. 2003. "The Origins and Development of Coffee Production in Reunion and 
Madagascar, 1711-1972" in Steven Topik and William Gervase Clarence-Smith (eds.) 
The Global Coffee Economy in Africa, Asia, and Latin America, 1500-1989 Cambridge: 
Cambridge University Press. 

Chalmin, Philippe. 1987. Traders and Merchants: Panorama of International Commodity 
Trading Chur: Harwood Academic Publishers. 

Clarence-Smith, William Gervase. 2003. "The Coffee Crisis in Asia, Africa, and the Pacific, 
1870-1914" in Steven Topik and William Gervase Clarence-Smith (eds.) The Global 
Coffee Economy in Africa, Asia, and Latin America, 1500-1989 Cambridge: Cambridge 
University Press. 

Dean, Warren. 1969. The Industrialization of Sao Paulo 1880-1945 Austin: University of Texas 
Press. 

Delle, James A. 1998. An Archeology of Social Space: Analyzing Coffee Plantations in Jamaica's 
Blue Mountains New York: Plenum Press. 



87 JOURNAL OF WORLD-SYSTEMS RESEARCH 

Fernando, M. R. 2003. "Coffee Cultivation in Java, 1830-1917" in Steven Topik and William 
Gervase Clarence-Smith (eds.) The Global Coffee Economy in Africa, Asia, and Latin 
America, 1500-1989 Cambridge: Cambridge University Press. 

Font, Mauricio. 1990. Coffee, Contention, and Change in the Making of Modern Brazil 
Cambridge: Basil Blackwell. 

Hattox, Ralph S. 1985. Coffee and Coffeehow;es: The Origins of a Social Beverage in the 
Medieval Near East Seattle: University of Washington Press. 

Heise, Ulla. 1987. Coffee and Coffee Houses. West Chester, PA: Schiffer Publishing Ltd. 
Translated from the German by Paul Roper. 

Holloway, Thomas H. 1975. The Brazilian Coffee Valorization of 1906: Regional Politics and 
Economic Dependence Madison: The State Historical Society of Wisconsin. 

Hopkins, Terrence, and Immanuel Wallerstein. 1986. "Commodity Chains in the World-Economy 
Prior to 1800" Review 10(1):157-70. 

Jimenez, Michael F. 1995. "'From Plantation to Cup': Coffee and Capitalism in the United States, 
1830-1930" in William Roseberry, Lowell Gudmundson, and Mario Samper Kutschbach 
(eds.) Coffee, Society, and Power in Latin America Baltimore: The Johns Hopkins 
University Press. 

McMichael, Philip. 1990. "Incorporating Comparison Within a World-Historical Perspective: An 
Alternative Comparative Method" American Sociological Review 55: 385-97. 

__ . 1992. Rethinking Comparative Analysis in a Post-Developmentalist Context" 
International Social Science Journal 44: 351-65. 

__ . 2004. Development and Social Change: A Global Perspective Thousand Oaks: Pine 
Forge Press. 

Mintz, Sidney. 1985. Sweetness and Power: The Place of Sugar in Modern History New York: 
Penguin Books. 

Moore, Jason W. 2003. "Capitalism as World Ecology: Brandel and Marx on Environmental 
History" Organization and Environment 16( 4): 431-58. 

Pendergrast, Mark. 1999. Uncommon Grounds: The History of Coffee and How It Transformed 
Our World. New York: Basic Books. 

Roseberry, William. 1995. "Introduction" in William Roseberry, Lowell Gudmundson, and Mario 
Samper Kutschbach (eds.) Coffee, Society, and Power in Latin America Baltimore: The 
Johns Hopkins University Press. 

Samper, Mario and Radin Fernando. 2003. "Appendix: Historical Statistics of Coffee Production 
and Trade from 1700 to 1960" in Steven Topik and William Gervase Clarence-Smith 
(eds.) The Global Coffee Economy in Africa, Asia, and Latin America, 1500-1989 
Cambridge: Cambridge University Press. 

Schivelbusch, Wolfgang. 1992. Tastes of Paradise: A Social History of Spices, Stimulants, and 
Intoxicants New York: Vintage Books. 

Sklair, Leslie. 2002. Globalization: Capitalism and Its Alternatives. Third Edition. Oxford: 
Oxford University Press. 

Talbot, John M. 2004. Grounds for Agreement: The Political Economy of the Coffee Commodity 
Chain Lanham: Rowman and Littlefield. 

Tilly, Charles. 1984. Big Structures. Large Processes, Huge Comparisons. New York: Russell 
Sage Foundation. 



THE COFFEE COMMODITY CHAIN 88 

Topik, Steven. 2003. "The Integration of the World Coffee Market" in Steven Topik and William 
Gervase Clarence-Smith (eds.) The Global Coffee Economy in Africa, Asia, and Latin 
America, 1500-1989 Cambridge: Cambridge University Press. 

__ . 1998. "Coffee"" in Topik and Allen Wells (eds.) The Second Conquest of Latin America: 
Coffee, Henequen, and Oil during the Export Boom, 1850-1930 Austin: University of 
Texas Press. 

__ . 1987. The Political Economy of the Brazilian State, 1889-1930 Austin: University of 
Texas Press. 

Topik, Steven and Allen Wells. 1998. "Introduction: Latin America's Response to International 
Markets During the Export Boom" in Topik and Wells (eds.) The Second Conquest of 
Latin America: Coffee, Henequen, and Oil during the Export Boom, 1850-1930 Austin: 
University of Texas Press. 

Trouillot, Michel-Rolph. 1982. "Motion in the System: Coffee, Color, and Slavery in Eighteenth-
Century Saint-Domingue" Review 5(3):331-88. 

Tuchscherer, Michel. 2003. "Coffee in the Red Sea Area from the Sixteenth to the Nineteenth 
Century" in Steven Topik and William Gervase Clarence-Smith (eds.) The Global Coffee 
Economy in Africa, Asia, and Latin America, 1500-1989 Cambridge: Cambridge 
University Press. 

Tucker, Richard. 2000. "The Last Drop: The American Coffee Market and the Hill Regions of 
Latin America" in Tucker, Insatiable Appetite: The United States and the Ecological 
Degradation of the Third World Berkeley: University of California Press. 

Ukers, William. 1935.AllAboutCoffee. New York: The Tea and Coffee Trade Journal Company. 
Wild, Antony. 2004. Coffee: A Dark History New York: W. W. Norton. 
Wrigley, Gordon. 1988. "The History of Coffee," chapter I in Wrigley, Coffee Essex: Longman 

Scientific and Technical and New York: John Wiley.