The story we know: Emergence of the Atlantic beef trade

The milestones of the global beef trade are clearly marked by advances in technology. Because grazing animals can be driven, they have long been traded ‘on the hoof’, as part of established trade networks, or as herds were seasonally relocated. The trade in fresh meat is much more recent. During the early nineteenth century, rapidly growing European cities began seeking supplies of meat from abroad, not merely to feed hungry populations but also as a guard against potential decimation of national herds by disease or over-culling. Britain initially sourced European imports, especially pork products from Denmark and Germany. Across the ocean awaited even greater opportunities. In the middle of the century, London began importing meat from the United States. From the stockyards in Chicago, St Louis, and Kansas City, a new generation of entrepreneurs began shipping beef and pork, either tinned or fresh, preserved by the cold air of the North Atlantic.

At the same time, the race was on to develop the technology that would allow meat to cross the tropical barrier. Producers in cattle-rich countries like Argentina and Australia were prevented by cost and sanitary controls from shipping live animals, and enjoyed only a regional demand for dried or salted beef.Footnote ⁵ Sensing the possibilities of the European market, various government and scientific bodies offered generous prizes to the inventor who could produce a reliable method of preserving fresh meat. From these competitions came such ideas as sealing meat in melted paraffin, fumigating it with ammonia or nitric acid, injecting it with brine, and reducing it to a concentrated paste. Many of these solutions did retain the meat’s nutritive value, but only refrigeration produced a fresh product that was acceptable to a consumer market. In 1875, Thomas Eastman made the first transatlantic shipment of refrigerated beef from New York to Great Britain. Two years later, the steamer Frigorifique, outfitted with an early ammonia-compression refrigerating plant, arrived to great fanfare in Rouen carrying Argentine beef that had been shipped for 110 days.Footnote ⁶

Refrigerated shipping transformed global commerce in beef. By the close of the century, Argentina had replaced the United States as Europe’s main supplier. British investors poured huge sums into developing export facilities in Buenos Aires, driving expansion in specialization and scale. In addition to clashes of interest between breeders, farmers, grazers, and fatteners, the most intense competition was among the top beef exporters, especially after well-capitalized American meat interests, such as Armour and Chicago’s Swift (which purchased the British and South African-owned La Plata Cold Storage Company in 1907), began challenging the British firms. The ensuing trade war among the largest players expanded productive capacity, while driving down consumer prices, resulting eventually in internally policed percentage quotas between British and American interests.Footnote ⁷

One innovation to emerge from this era of competition was the Vestey Brothers’ Union Cold Storage Company. From their start shipping American beef to Britain, the Vesteys built a global network that included ranches in South America and Australia (which was forced by distance to ship frozen beef), and retail in Liverpool and London. This was the first successful model of vertical integration in the meat industry. Their cold storage network, by far the world’s largest (capacity in 1912 was estimated at millions of carcasses), gave the enterprise the unique ability to weather global fluctuations in price, one of many ways in which scale favoured the largest producers.Footnote ⁸

Although the United States had by this time become a net importer of beef, the First World War gave the American industry a temporary reprieve. War decimated European production, pushing up both demand and prices. The market again crashed after the armistice, as European nations drove down strategic reserves and reinstated quotas or duties on imports. By the end of the 1920s, much of the international meat trade had been corralled into complex networks of bilateral trade agreements.Footnote ⁹

Much about this eye-catching story epitomizes the forces that drive the globalization of commodity chains. Technological advances in refrigeration and shipping allowed ever cheaper transportation over ever greater distances. Capital flows, coming especially from Britain and the United States, developed the industry in major export centres in Argentina and Australia, as well as in smaller ones like Brazil, Uruguay, Mexico, Venezuela, and Southern Rhodesia (Zimbabwe).Footnote ¹⁰ As productive efficiencies and competition drove down costs, consumers developed new tastes for beef, which became a regular feature of a working-class Londoner’s diet.Footnote ¹¹

But there are also good reasons to look beyond it. Focusing on the point-to-point relationships between the largest exporters and the largest and wealthiest importers shows well the cutting edge of the trade: the advantage of scale, introduction of new technologies, new sources of investment, new types of productive organization. But it overlooks the networks of smaller producers and markets. Taken together, this lateral trade was also of high value, and was subject to very different iterations of the political and economic forces that smoothed the highway of trade between Buenos Aires and London. Such lateral networks often developed in parallel to, or remained aloof from, what would otherwise be mistaken for the entirety of a global trade system. As China’s live animal trade shows, they often built on the advantages and connections of pre-existing trade networks.

Sheep and security in Hulunbuir

Having practised animal husbandry since prehistoric times, China already possessed a long history of distance trade in live animals, consisting particularly of seasonal drives of grazing animals from northern pastures into the agrarian heartland. Because it traversed what later dynasties would call the ‘fanbu’, a ring of ethnically distinct and semi-sovereign areas such as Mongolia and Tibet, the live animal trade was politically sensitive. Successive regimes kept it under tight control. The Ming dynasty (1368–1644) bottlenecked the trade through a series of official ‘horse markets’ (ma shi) that regulated interaction between the agrarian and pastoral worlds, appeasing Mongol sovereigns who wanted access to Chinese goods, while ensuring a regular supply of animals such as military horses. The subsequent Qing (1644–1911) abolished the markets, and restricted trade instead to approved Chinese firms, generally those from the province of Shanxi, which also were subject to a variety of regulations. Traders had to carry a pass designating their status, and were forbidden to remain on the grasslands for more than one year. In return for permission to conduct trade with, and later beyond, the fanbu, merchant houses provided logistic support for the dynasty, transporting grain and goods during the military campaigns of the eighteenth century, provisioning garrison cities such as Guihua (Höhhot), and conveying tax and tribute from Mongol princes.Footnote ¹²

In places such as the northern region of Hulunbuir, the animal trade provided income and strategic support for pastoral defences. Unlike neighbouring Outer Mongolia, which was ruled by Mongol princes, this region was part of the military district of Heilongjiang, and its defence was directly coordinated by the dynasty. The Qing first instituted pastoral garrisons in this area in 1732, five years after the Treaty of Kiakhta affirmed the Ergune river as the border with Russia. In an effort to establish the vast grassland as a permanent defensive perimeter, the Heilongjiang Governor Jorhai resettled the region with 7,000 Barag Mongols, who were relocated from near present-day Qiqihar and Outer Mongolia, and organized into military units called banners. Solon Ewenki were resettled along the river into border defences known as kharuul (Chinese kalun).Footnote ¹³ Each of the banners was allotted specific areas of pasture, from which they produced vast livestock wealth. The Barag brought with them 15,494 horses, 9,494 head of cattle, and 93,540 sheep, and the Qing gifted the Solon another 100,000 animals.Footnote ¹⁴ Two centuries later, a local schoolmaster was able to boast that ‘Rich households count their sheep in the tens of thousands, and small ones in the hundreds … [Even] the poorest households have a cow and a few sheep.’Footnote ¹⁵ Necessities, such as brick tea, cloth, and boots, which could not be acquired locally, were to be supplied by eight Shanxi trade firms.Footnote ¹⁶

The pastoralists paid for these purchases through an annual cycle of long-distance animal trade. Around the year 1800, the merchant houses established an annual market at Ganjuur Temple, 180 kilometres south-west of the walled town of Hulun (present-day Hailar), a site chosen for the mutual convenience of buyers and pastoralists.Footnote ¹⁷ The five-day-long market was set for the beginning of the eighth lunar month, to coincide with an annual pilgrimage and festival cycle, as well as the time when herds had been fattened by summer grazing. Merchants returned to China with carts of hides and wool, as well as live animals that were driven southward on a two-month overland journey (see Figure 1), passing through other commercial centres, such as the Buddhist temple city of Dolonnuur and the vital Great Wall pass at Zhangjiakou (also known as Kalgan, meaning ‘gate’), and thence on to urban markets for sale and slaughter. The annual exchange dominated economic life in Hulunbuir, supplanting the smaller trade routes plied by itinerant traders, and providing producers with the vast majority of their income for the entire year.Footnote ¹⁸

Figure 1 Hulunbuir caravan trade. Map by Thomas DuBois.

Mongolian cattle, Chinese beef

What can the content of this highly structured exchange tell us about the consumption of grassland animals as food? To answer, we might begin by looking not at cattle but rather at the much larger sheep trade. Japan began collecting extensive economic intelligence on this trade during the early twentieth century. A consular report from 1922 estimated the total sheep sales at the Ganjuur market at 20,000 animals. Similar sources from a few years earlier estimate 200,000 annual live sheep exports from all of eastern Inner Mongolia (which by this point included Hulunbuir). While the sheepskins and wool sold at Ganjuur were easily shipped by caravan, the primary reason for driving live sheep was their value as food. Although Zhangjiakou was not an especially large city, its residents consumed large amounts of fresh meat, which, as Table 1 shows, resulted in a significant slaughter of sheep.

Table 1 Slaughterhouse statistics from four Chinese cities, 1934

^a Achieved by multiplying the January 1934 price paid per animal by the number of animals killed, and dividing by the amount of meat produced. These figures would differ from the actual marketplace prices, but nevertheless are a reasonable indicator of relative value.

Source: AS, 17-27-195-04, ‘Shengchu tuzai diaocha (Survey of animal slaughtering) 1933–1936’.

Cattle were traded in smaller numbers, but their food value was proportionally higher. The Japanese sources above estimate the Ganjuur live cattle trade at 3,000 head, and the total exports of eastern Inner Mongolia at 10,000 head, yielding sheep to cattle ratios of approximately 7 : 1 and 20 : 1, respectively. However, these numbers alone are misleading. Cattle were significantly more expensive than sheep at the point of purchase: one head at Ganjuur sold for 70 yuan, compared to 5 yuan for sheep, a 14 : 1 difference that correlates roughly to the amount of meat per animal (400 jin for cattle versus 30 jin per sheep).Footnote ¹⁹ Despite the large number of sheep that were moved through Zhangjiakou, that city’s slaughterhouses produced significantly more beef (770,000 jin) than mutton (464,000 jin) between 1933 and 1934.Footnote ²⁰

Clearly, there were substantial differences in how these two animal exports were being used as food. One is that live sheep were valued primarily as meat, but cattle were also used as farm and transport labour, a difference that Kuo Chung-hao has traced back through classical husbandry treatises such as the sixth-century ‘Techniques for the welfare of the people’ (qimin yaoshu).Footnote ²¹ Vincent Goossaert and others have discussed the ways in which both custom (Buddhist proscriptions against taking life, the Confucian moral command to repay with kindness the ox’s lifetime of labour) and law (which aimed to protect animal stocks and to keep order among the stockbreeding population) protected cattle from wanton slaughter. The Qing statute ‘On the killing of horses and cattle’ (zaisha maniu), which forbade the slaughter of work horses and cattle, was copied nearly verbatim from the previous dynasty’s legal code, and its spirit had appeared in previous codes dating back two millennia.Footnote ²²

Who in China actually consumed cattle as food? Writing in 1882, John Nevius, an American missionary who spent much of his adult life in China, described in fairly absolute terms a revulsion for beef: ‘Beef is never exposed for sale in the Chinese markets. The meat of the few cattle which are used for ploughing is, when they are killed, disposed of privately, almost clandestinely. There is a strong and almost universal prejudice against eating beef, and the practice of doing so is declaimed against in some of the moral tracts.’Footnote ²³ Yet, there clearly was a market for beef. The late eighteenth^-century gastro-medical compendium, Gastronomy of the Sui garden (Suiyuan shi dan), included instructions for preparing all manner of meat, fish, and game. On the topic of cooking beef, it advised the buyer to ‘braise the meat until extremely soft in a mixture of two parts water and three parts wine, adding soy sauce to the remaining liquid. Since beef has a distinct taste, no other ingredients should be added.’Footnote ²⁴ Several decades before the publication of Nevius’ book, the British diplomat Henry Charles Sirr had already written glowingly of the quality of the beef he had eaten in Canton, which was ‘nearly as good and well fatted as it is in England’.Footnote ²⁵

Even areas that restricted the slaughter of cattle recognized beef’s nutritional value, particularly for the elderly. The fact that the trade had to be controlled legally, with the constant promulgation of new proscriptions against private cattle slaughter, is itself a testament to the prevalence of beef consumption. But again, such restrictions were often enacted in response to local disasters; they were neither indiscriminate nor universal.Footnote ²⁶ Some areas, especially those with significant Muslim populations, consumed large amounts of cattle or water buffalo meat, and traded in speciality beef products, such as ganba, a dried and spiced beef produced in Yunnan.Footnote ²⁷ Cattle were sold and slaughtered freely in certain markets, most notably urban areas that could quickly absorb large amounts of fresh meat. An important terminus of the pastoral animal trade, the Muslim quarter of Beijing, now known colloquially as Cow Street (niu jie), even had its own slaughtering grounds. A nineteenth-century source described not only the scale of the trade but also the high degree of specialization in the processing of cattle and sheep:

The high price of beef in Zhangjiakou derived from a particular combination of scarcity and taste. The commercial houses that dominated the grassland trade were meticulous observers of consumer demand. They knew local tastes well enough to source different sorts of tea to suit the palates of different Mongol buyers, and were no less knowledgeable about the Chinese market for animal imports. Cattle driven in from the grasslands represented a large investment, not only because of the high purchase price but also because cattle are more difficult than sheep to drive over dry or rocky ground. While many of the cattle driven as far as Zhangjiakou were slaughtered in the city, it made more sense to continue on to larger markets like Beijing, where the price of beef was higher. As Table 1 shows, beef was less expensive in cities such as Ji’nan, where cattle were raised specifically for food, than it was in urban slaughtering centres such as Guangzhou and Qingdao, which were also centres of beef export, and was cheapest and most plentiful in cattle-skinning centres like the northern towns of Qiqihar and Heihe.Footnote ²⁹ But as we shall see, market price was affected by many factors, of which proximity to supply was only one.

For more than two centuries, the caravan trade was both politically necessary and commercially viable. Cattle from the grassland were regularly driven into northern cities like Beijing, where a significant part of their value was as food. Despite the many logistic, legal, and cultural obstacles to the slaughter of cattle, beef was a part of the Chinese diet, at least regionally, and was in some cases particularly prized. But beyond political restrictions, this method of trade was constrained by logistic complexities. Driving cattle was slow and seasonal, and required a mechanism for transferring value in the other direction. Just as in the American West, developing the Chinese beef trade would require a better way of getting animals to market.Footnote ³⁰

Russia and the emergence of Harbin

It was not long before foreigners began eyeing Hulunbuir’s large supply of beef cattle. Russian traders had been active in both Outer and Inner Mongolia even before the 1860 Peking Treaty legalized flow of people and livestock. By the beginning of the twentieth century, the Trans-Siberian Railway had connected European Russia to the Pacific, while new settlements in Amur and Maritime provinces along the Chinese border were emerging as important markets in their own right. This population grew as the First World War and the Russian Revolution drove capital and people into the Russian Far East.

Russian demand for beef cattle was served by a live animal trade, which was driven overland at various points along the long and sparsely populated land border with Mongolia and China. A concerned Qing official wrote in 1886 about kharuul in Hulunbuir trading cattle and horses across the Ergune river.Footnote ³¹ A generation later, Japanese consular reports described a trade in live cattle by Mongol herdsmen to Russians all along the border. Further east, Russian and Chinese middlemen in the Chinese border town of Hunchun plied a small but brisk trade, driving 30–40 head per day from Korean villages on to Vladivostok.Footnote ³² Legal restrictions on the trade were extremely hard to enforce. Authorities monitored sales at places such as Zhangjiakou, both to ensure tax payment and because Russian merchants were known to purchase stolen livestock.Footnote ³³ Animals entering Russia were supposed to be held at the border but, in reality, traders could easily avoid the expense of quarantine simply by travelling off the main road.Footnote ³⁴

Initially, the slaughtering was done in Russia rather than China. Russian merchants in western Mongolia drove livestock to meat processing plants in Irkutsk.Footnote ³⁵ Cattle from Hulunbuir had a shorter journey, to the border town of Zabaikalsk, which was dotted with slaughtering and processing facilities.Footnote ³⁶ Meat processed in Russian slaughterhouses was sent on by rail to a developed internal market, one that seems to have continued for years after the Russian Revolution, when many of the animal buyers at the Ganjuur cattle market in Hulunbuir still worked for Zarim, a Kazan (Tatarstan) concern that sold meat to buyers in both European and Asiatic Russia. Yet, in the absence of domestic processing capacity, meat exports from China were limited to items such as intestines, which could be preserved in salt for use as sausage casings. According to Japanese sources, Hailar exports of sheep and cow intestines reached markets in Berlin, Riga, and London.Footnote ³⁷

The export of fresh beef began with the construction of two new Russian railways: the Chinese Eastern Railway (CER), which traversed northern Manchuria, and the South Manchuria Railway (SMR), which connected to the CER and ran south to the port of Dalny (Dalian) (see Figure 2). Opened to passenger traffic in 1903, the CER initiated a new wave of settlement and trade within the cattle-producing regions of Hulunbuir and northern Manchuria. Its five-year construction brought in 200,000 labourers to lay track, fell trees, and dig tunnels, as well as thousands more who provided food, housing, and entertainment for the workers. New towns, such as the border crossing at Manzhouli (across from Zabaikalsk), the lumber town of Yakeshi, and the farming centre of Zhalantun, were founded by or grew with these railway pioneers. The line also changed the ethnic composition of the region. Most of the workers were Han Chinese, but the line also employed Russian surveyors and engineers, and attracted Russian entrepreneurs and adventurers; by 1922, the growing town of Hailar was about one-third Russian.Footnote ³⁸ These new immigrants brought with them new tastes and new animal breeds that helped transform the livestock market.

Figure 2 Railways and the Harbin beef trade, c. 1921. Map by Thomas DuBois.

The CER developed the meat trade in two ways. It moved people and goods across the densely forested Hingan mountains, which had previously had been an impediment to travel, particularly in winter. The line lowered the costs of carrying goods such as coal into the growing city of Hailar, but most of the trade flowed out: east through Harbin and on to Vladivostok, where goods were loaded onto oceanic steamers operated by the railway. This combination of railway and maritime networks made it profitable for the first time to export bulky items to a global market, prompting the development of timber, mining, and farming industries across the length of the line.Footnote ³⁹ The railway itself was a magnet for investment and industry. In addition to moving heavy goods, the rail line also encouraged the development of banking in towns like Qiqihar and Hailar, easing the flow of productive investment, and monetizing the export economy, much to the advantage of producers such as pastoralists. Even after Japan gained control of most of the SMR (along with the city of Dalian) in 1905, the two lines remained the focus of capital-intensive industries such as mining and manufacturing. The commercial arm of the CER itself invested along the length of the line, building hotels and clinics, and especially productive improvements, such as lumber yards and wool-washing stations, to encourage industrial development. Because the railways and adjacent lands were leasehold territories under foreign sovereignty, they were partially insulated from Chinese restrictions on cattle slaughter. The CER actively encouraged the meat industry by discounting shipping, and providing free use of icehouses and storage facilities in Hailar and Manzhouli.Footnote ⁴⁰

The two lines converged on the city of Harbin, which quickly developed into a major centre for beef production and processing. Both rail lines ran through cattle producing areas. The CER crossed the grasslands of Hulunbuir, which held approximately half of the estimated 300,000 head of cattle in Heilongjiang province, while the SMR ran through the grain belt of Manchuria. The widespread use of oxcarts in Manchuria promised a steady supply of farm and draft animals; a 1917 survey of Heilongjiang counted nearly half as many cattle as pigs.Footnote ⁴¹ The railway also gave both pastoralists and farmers more freedom to sell animals, at any time of the year, at Hailar or any number of smaller markets along the line. The needs of the growing dairy industry, also centred in Harbin, ensured a steady supply of calves, which were raised to mature weight and sold.Footnote ⁴² Ample space, plentiful grain and bean husks from the region’s numerous oil mills, and six months of cold weather (when most of the slaughtering took place) all favoured the growing industry in Harbin.

Beyond its access to cattle-producing regions, Harbin was uniquely suited to its new role as a meat-packing centre. Early twentieth-century Harbin was a commercial boomtown. A hub of soybean and grain export industries, its population had swelled from virtually nothing to one-third of a million within a single generation.Footnote ⁴³ The city was not only a wealthy and productive centre of finance and industry; it was also a locus of epidemic research.Footnote ⁴⁴ In 1911, Harbin became home to the Manchurian Plague Prevention Service, which remained on hand to oversee veterinary and hygiene concerns in the new slaughterhouses and markets.Footnote ⁴⁵ The novelty of hygienic oversight cannot be overstated. Writing in 1898, the Surgeon General of the United States Navy described the traditional slaughtering ground outside Shanghai as a fetid hellscape of flies and foul odours, where sick and dying animals were slaughtered for meat, and pools of dung and blood mixed and rotted under the hot sun. Even with the city under foreign administration, it required a twenty-year struggle to erect a modern abattoir and effect regulation of meat storage and transport.Footnote ⁴⁶ The new city of Harbin had no such legacy to contend with; its industry would be born on a modern footing.

War also facilitated the rise of the industry in Harbin. Responding to demand created by the Russo-Japanese War, the city’s slaughterhouses produced 2,792 tons of meat, primarily pork, in 1905. A survey from two years later listed slaughtering and meat processing among the city’s earliest specialized industries. Pig slaughtering spiked again during the First World War: in 1910, the railway carried 16,706 pigs into the city; by 1918, that figure had reached 85,500. In tandem with a fall in the price of pork, the number of pigs slaughtered crashed to 33,823 the following year. Partially as a result of disruption to rail traffic after the October Revolution, only 2,810 animals were imported into the city in 1919. Imports rebounded to an average of 10,000 per year between 1920 and 1924, and then fell off entirely after 1925.Footnote ⁴⁷

Harbin’s beef industry followed a slightly different trajectory, but relied on the same advantages. As with pork, the shipping of live cattle by rail grew in importance in the first decades of the twentieth century. In 1903, only 682 head of cattle were shipped to Harbin by rail. Three years later, that number had grown to 11,538, just under half of the 28,000 sold in the city. During these years, live cattle were still being exported to Siberia and Europe, 15,000 head in 1908 and 18,499 in 1912, but these live exports were no longer for lack of domestic slaughtering capacity.Footnote ⁴⁸ By 1908, three industrial slaughterhouses along the CER: two in Harbin (in Fujiadian, and Old Market Street), along with a third in Hailar, processed 68,510 animals, just under half of which were cattle, mostly slaughtered in the larger Fujiadian slaughterhouse. Slaughtering cattle represented a high addition of value, compared to other animals, and the growth of slaughtering capacity in the region was one reason behind the decline in the live cattle trade to Russia, from 64,097 head in 1913 to 9,061 head in 1919. (Nationally, the cross-border trade in pigs and chickens continued more or less unabated during the same period.)Footnote ⁴⁹ Pastoral Hailar, in contrast, slaughtered mostly sheep, fewer cattle, and no pigs.Footnote ⁵⁰ In 1924, the abattoirs of Harbin accounted for 24,196 head of cattle, just over half the 45,863 processed by slaughterhouses along the entire length of the CER. Table 2 confirms the central role of Harbin, the 1911 export figure of 5.4 million kg of beef corresponding to roughly 20,000 animals.Footnote ⁵¹

Table 2 Meat shipments in Manchuria, 1911

Source: JACAR, A06033517400, Manmō keizai yōran (Overview of economy in Manchuria and Mongolia) [1916], p. 556.

The emerging beef industry in Harbin attracted global investors. Foreign buyers were active in the Harbin cattle market (da niu fangzi) as early as 1903, and the growing Russian population (estimated in 1905 at nearly one-quarter of the city’s 100,000 inhabitants) became a dominant power in the purchase of livestock from across the region. Much of the early investment in the industrial meat industry was Russian, either private investors in Harbin, exporters such as Zarim, or the CER itself.Footnote ⁵²

Russian investors were joined by the Union Cold Storage Company’s William and Edmund Vestey, who by this time had already proven themselves in the cutthroat markets of Chicago, Buenos Aires, and Northern Australia. The Vesteys saw great potential in China. The two brothers had first visited Nanjing in 1912, and the following year founded the International Export Company Chiangsu (also known by the Chinese name Heji yanghang), investing £250,000 in a facility meant initially to process ducks and egg albumin. In 1916, at the height of the First World War, that facility was able to process 200 tons of eggs per day, and had slaughtering and storage capacity for 20,000 ducks or chickens, as well as 3,000 pigs and 400 cattle. In that year, with the global market in meat booming, Union Cold Storage established the British Produce Export Company in Harbin. With the brothers themselves in possession of 99.7% of company shares, Produce Export invested £100,000 to build stockyards, slaughterhouses, cold storage (including the city’s first mechanical refrigeration plant), and shipping facilities in Harbin, as well as a purchasing office in Hailar.Footnote ⁵³

Although Produce Export’s registration documents spelled out the company’s intention to export beef, lamb, and pork to Britain, it seems that the London market for Mongolian beef did not materialize. In 1917, Produce Export purchased 10,000 hogs for export to Europe, and a report from 1922 shows the company contributing to the growth of imports of live cattle from Hulunbuir to Harbin.Footnote ⁵⁴ Beef from the Nanjing works did briefly reach Britain, but there is no evidence that beef from Harbin ever arrived at the Vesteys’ own retail outlets in London, which were still supplied primarily by enterprises in Argentina, Australia, and elsewhere. In the end, the Vestey operation was important to Harbin, but the reverse was less true; the Harbin works would remain only a small part of the sprawling Vestey empire. From the European perspective, they were probably most useful as a hedge against disaster elsewhere in the notoriously unpredictable global production chain.

Instead, beef processed in Harbin went on to a variety of markets in China and overseas. The wartime market for live cattle in Harbin peaked in 1916, with 37,000 head sold to Chinese, Russian, and British buyers. As the war drew to a close, the Harbin market was energized by speculation on the possibility of new export markets in Europe and elsewhere. Soon after the war, Manchurian beef ‘of very excellent quality’ was reportedly sent under French flag to ports in Egypt.Footnote ⁵⁵ But the real market for Harbin beef still lay to the east. Table 3 shows that between 1913 and 1916, the majority of beef was exported through the border at Suifenhe, and on to Vladivostok. Strategic but remote, Vladivostok had always been a challenge to provision. In 1898, the Russian government began negotiating with Australian producers to supply the city with seaborne shipments of frozen beef, which in the aftermath of the conflict with Japan briefly spiked at 9,000 tons.Footnote ⁵⁶ Ten years later, as the city overflowed with refugees and troops, it came to rely on more local suppliers. During the 1920s, the CER continued exporting beef and mutton to Vladivostok and Maritime Province, edging out earlier imports from Canada and Australia, and even preferred suppliers in Korea and Qingdao. A CER report attributed this change to the high quality of the region’s animals, and by all accounts the Vladivostok market did indeed prefer the taste of Manchurian to Australian beef, although the unpredictability of oceanic shipping capacity was at least as important.Footnote ⁵⁷ Russian purchases of Harbin beef began to weaken after the war, accelerated by the reopening of sea-lanes, an outbreak of bovine disease in 1921, and continued insecurity in eastern Siberia. Yet even after Sino-Soviet clashes closed the Chinese border to rail traffic in 1929, meat from Harbin continued to reach Russian markets.Footnote ⁵⁸

Table 3 Destination of meat shipments in Manchuria, 1913–16 (thousand kg)

^a Total shipments includes both imports and exports.

Source: JACAR, A06033525800, Tōshin tetsudō shōgyōbu shigyō seiseki gikyō (Operations of the CER department of industry), 2, 1915, pp. 161–2.

Against the backdrop of the growing Harbin industry, Russian importers continued to expand their presence, placing Russian and Mongolian buyers at the Ganjuur fair and locally throughout Hulunbuir. At the 1922 Ganjuur market, a concern known as the Morovich brothers bought 400 head of cattle, Zarim bought 1,200 head, and Barukhtonski 5,000 head, while Produce Export made an advance order for only 1,000. In the early 1920s, Russian investors began to consolidate their operations, establishing a Mongolian Beef Export Company and investing in refrigerated ships to export to Amur and Maritime provinces.Footnote ⁵⁹ Over time, Produce Export came to command an ever-larger share of the trade, purchasing 6,400 live cattle and 141,000 sheep in 1924 alone.Footnote ⁶⁰ But conflict was not inevitable, in part because Produce Export cooperated with Zarim and other experienced buyers at the point of purchase, and because the Harbin outfit seemed to be moving towards sheep, which could be sold frozen, and were thus more suitable for long-distance shipping.Footnote ⁶¹

The rise of beef export production in Harbin shows the decisive role of a number of factors. Most obviously, the railway developed finance and industry along its length, and allowed the production and export of fresh meat on a scale that had never been possible. The region’s turbulent history proved to be an unexpected asset. War provisioning created instant demand for meat, as did the flow of migrants, many of whom brought wealth, experience, and business contacts. The new industry uniquely suited this sort of investor. Harbin’s export firms understood international hygienic standards of veterinary inspection, slaughtering, and storage, and brought capital and connections that carried beef to new markets in Amur, Maritime Province, and Siberia. While the trade was not for the light of pocket, it appears that Harbin’s slaughtering and export industries did not necessarily favour size. The well-capitalized Vestey operation did not seem any better placed than even the smaller Russian firms, who could rely on the infrastructure of icehouses and shipping facilities built by the railway. Within this world, Produce Export seems to have acted uncharacteristically as a regional player, working with local firms, and serving local markets, rather than its global retail hub in London.

Qingdao turns to Japan

Even with the development of the Harbin industry, large numbers of cattle were being exported from agrarian regions along the coast. In 1903, a worried Qing official criticized the loss of cattle from the southern province of Guangxi: traders in the city of Wuzhou were selling cattle, 15,000–20,000 head each year, to the British garrison in Hong Kong. Further north, local officials in the port of Tianjin seized a shipment of live cattle destined for export. Although these officials had been acting in the spirit of the ban against slaughter, British and American consuls in Guangdong accused them of overstepping their authority and forcing honest merchants to renege on existing contracts. The dynasty’s overall policy remained unclear, or at least it appeared so to the American investor who inquired through the Beijing embassy about opening a slaughterhouse near Tianjin. Noting the likely investment of 150,000 silver taels and the creation of job opportunities, the embassy asked if the dynasty’s position on beef exports had changed. Citing the concerns of the Qing law, the response indicated that it had not. Seeking to exert some control over the trade along the coast, central authorities formally closed Cheefoo (Yantai) and Vladivostok to cattle exports, restricting the northern trade to the port of Qingdao.Footnote ⁶²

Located on the coast of Shandong peninsula, the city of Qingdao resembled Harbin in a number of ways. It was founded in 1898, and became the terminus of a foreign railway, in this case the German-financed line connecting to Ji’nan. Over subsequent decades, it remained an outpost of foreign capital and influence. As in Harbin, war turned the tide against the founding power. Japan seized Qingdao late in 1914, and thereafter maintained an aggressive presence, using the city as a major manufacturing and logistics base for north China. In subsequent years, Japan came to dominate trade through the port, in 1918 accounting for 86% of the ships and 78% of the shipping tonnage in and out of the harbour. Between 1916 and 1922, the value of Japanese trade with the port grew by a factor of five.Footnote ⁶³

Equally instructive are the differences between the two cities. Like Harbin, Qingdao was located near a large cattle-producing catchment, but, rather than pastoral Hulunbuir or the grain belt of Manchuria, Qingdao drew primarily from the densely populated agrarian regions of Shandong province, precisely where conflict over Chinese government restrictions on slaughtering was most pronounced.Footnote ⁶⁴ Although Qing authorities had opened Qingdao to the cattle export trade, repeated agrarian disasters during the 1920s prompted officials of the subsequent Chinese Republic (founded 1912) to control or stop altogether the sale of farm cattle to slaughtering concerns. In addition, the decade was marked by growing political tensions, and a pronounced antipathy to Japan, created in no small part by Japan’s refusal to cede its control of German interests.

During these years, Japan was itself developing a taste for beef. For much of the previous Tokugawa period (1600–1868), a Buddhist-inspired proscription against animal slaughter had shaped the Japanese diet. Just as in China, this ban was only partial: it permitted consumption of meat by the elderly, the hereditary warrior class, and foreigners, who kept their own slaughtering ground near the port of Yokohama. The 1868 Meiji Restoration ushered in a new era of beef consumption, if not as food then as aspiration. The Tsukiji district of Tokyo developed into a thriving slaughtering and beef-selling centre that attracted both Japanese and foreign merchants. Although actual per capita consumption of meat was vanishingly small (1921 figures show slaughter of 277,456 domestic and 40,000 imported cattle for a population of 56.8 million, suggesting an average of just over 1 kg of beef per person per year), the taste for beef became emblematic of the new era, particularly after the widely revered Meiji emperor adopted the diet publicly in the 1870s.Footnote ⁶⁵ The Japanese military was provisioned with beef during the decade of continental wars (the Sino-Japanese War, the Boxer suppression, and the Russo-Japanese War) between 1894 and 1905. Having exhausted local options, the Japanese quartermaster was forced to cast around for other supplies, including dried beef from Argentina, before deciding on the option of importing salted beef from Canada.Footnote ⁶⁶ Over subsequent decades, Japanese authorities closely monitored the supplies and price of beef on the civilian market.Footnote ⁶⁷

Cutting-edge beef export industries in Qingdao dated from the period of German control. German authorities had introduced regulations for hygienic slaughter and storage in 1899, and completed the first modern slaughterhouse in 1906, soon after the city was designated the sole northern port for legal cattle and beef exports. Built at a cost of 850,000 Marks, this slaughterhouse was fitted with refrigeration equipment imported from Dresden, and staffed with military veterinary inspectors, some of whom remained on for a time after the imposition of Japanese control in 1914. The plant processed 15,000 animals in its first year, initially serving the city’s 4,000 European and American residents, as well as a large catchment of Chinese consumers in nearby villages, who consumed the plant’s relatively steady production of pork. In contrast, the growing output of beef, shown in Table 4, was primarily for export.Footnote ⁶⁸

Table 4 Qingdao slaughterhouse tally by animal (head), 1908–18

Sources: COC, October 1918; COC, February 1919. 1914 figures differ between the two sources.

Like Hulunbuir and Manchuria, Shandong proved to be an excellent production base. The province was rich in cattle, 1.7 million head in 1932. Unlike stall-feeding, cattle raising in this dense agricultural region did not divert resources from human consumption (what Harro Maat has called an ‘anti-commodity’). Rather, animals were grazed while young, and later fattened on farm waste, such as the leaves and stalks of the region’s peanuts and sweet potatoes, prior to sale and slaughter.Footnote ⁶⁹ Known for their size and strength, Shandong cattle had long been sold as draft animals to inland provinces like Henan and Shanxi, although they do not seem to have reached the Beijing market, which was still primarily served by the overland trade of sheep and beef from Mongolia.Footnote ⁷⁰ But especially when mixed with imported German breeds, Shandong cattle were excellent beef animals, and, like the herds of the Hulunbuir grasslands, were known and prized for their taste. Live animals, raised by farming households, and purchased by Chinese agents in cattle markets in Ji’nan and Jining were carried by rail to Qingdao, whence fresh or preserved beef was shipped to destinations along the coast. As with Harbin, military provisioning during the First World War produced a sharp spike in demand, particularly as a replacement for Australian beef. In 1913 alone, 90,000 quarters (that is, 22,500 dressed carcasses) were shipped via Qingdao to Russian troops in Vladivostok.Footnote ⁷¹ A 1915 petition for harbour passage by an investor who had contracted an American refrigerated steamer to deliver 1,000 beef carcasses each month from Qingdao slaughterhouses to Tianjin, Dalian, Vladivostok, and Nikolaevsk confirms that an extensive, albeit irregular coastal trade continued to flourish during the war.Footnote ⁷²

Following the war, Qingdao beef exports found a new market provisioning frozen beef to American troops in Manila. Until the late nineteenth century, Manila had consumed locally raised cattle. During the 1880s, the city began introducing live animals from Indochina and China, carelessly sourced imports that brought with them waves of rinderpest. In his recent book, Daniel Doeppers discusses the great efforts that the Philippine Bureau of Agriculture exerted to insulate the native herd from this extremely destructive disease, including the introduction of mandatory veterinary quarantine, fencing off infected areas, and switching to safer Australian imports.Footnote ⁷³ For human consumption, the safest option was simply to import frozen meat. In 1898, the US garrison contracted with the Queensland Meat Export Company for an annual shipment of dressed meat that in 1910 totalled 250,000 pounds of beef, pork, and rabbits, plus 8,000 lamb and mutton carcasses.Footnote ⁷⁴ Even more than in Vladivostok, the disruption of shipping by disease and war wrought havoc on food imports, not only those from Australia but also regional supplies such as rice from Bangkok and Saigon. Prompted first by sanitary restrictions, Manila’s ‘beef famine’ provided a golden opportunity for Chinese producers. In 1916, an American merchant named William Katz contracted Qingdao slaughterhouses to supply 600,000 pounds of beef, which were to be shipped to Manila on the Warren and the Merritt, two navy transport ships that the American government had recommissioned for use as merchant freighters. Beef shipments to Manila, always carried frozen on these two ships, would continue for the next few years.Footnote ⁷⁵

Rather than competing for the Manila market, the Vesteys’ southern venture briefly made profitable exports to Britain. During the war, the original Nanjing works moved increasingly into beef production, buying Shandong cattle, which were purchased and slaughtered in Ji’nan, before being transported in refrigerated rail cars to Nanjing and thence to Shanghai for export. At the height of the wartime boom, a portion of these exports did indeed reach the British market. But the amounts were negligible, peaking in 1918 at roughly 10,000 tons (just 3% of Britain’s frozen beef total imports), and falling off immediately after the war’s end.Footnote ⁷⁶ Moreover, such actions turned local authorities against the firm, which in 1915, was reprimanded for illegally purchasing 22,000 draught cattle across thirteen counties. The cost of such antagonism became apparent during the 1920s and 1930s, when a combination of boycotts, famines, and floods prompted Chinese authorities to ban ‘treacherous foreign merchants’ (wai lai jian shang) from shipping beef cattle or beef by rail. The same railways gave special discounts for shipping work cattle to afflicted areas.Footnote ⁷⁷ Nor was the Nanjing operation able to expand into the nearby Shanghai market, which continued to be served largely by traditional means: Chinese middlemen purchased live animals at monthly cattle markets in places such as Danyang, just over 200 kilometres to the north-west, and shipped them into Shanghai for slaughter and sale in the city’s wet markets. Even the superbly efficient Vestey operation would have been hard pressed to compete with this low-cost source. By the 1920s, Vestey enterprises in Nanjing and Hankou had returned to their original, and more profitable, business of egg and poultry processing.Footnote ⁷⁸

At that time, the Qingdao beef industry had not yet turned decidedly to Japan. The many Japanese companies that bought, slaughtered, and exported Shandong beef were not simply the foot soldiers of extractive imperialism; Japanese enterprises often incorporated significant amounts of Chinese merchant capital, and most on-the-ground operations were effectively Chinese-run. It took a combination of factors for these ventures to focus exclusively on Japan. The first was a shift in other export markets. Fresh meat exports to Vladivostok were high through the war, but with the combination of peace and collapse of the rouble, these fell by more than half between 1918 and 1919 (from roughly 9 million kg to 4.4 million kg). Australian imports to Manila resumed in 1921, and the last Qingdao shipment of beef seems to have been in June of that year. The last shipment to Vladivostok was just one month later. However, transhipments from Hong Kong and other ports maintained China as the largest source of meat imports to the Philippines for at least another decade.Footnote ⁷⁹

New opportunities in Japan were made possible by an increase in refrigerated shipping capacity. The Manila market had relied on American refrigerator ships, while Vladivostok either used American ships, shipped live cattle or fresh meat in the winter, or exported salted meat in the warmer months. Lacking refrigerator ships, Japanese exporters, most of which were small-scale concerns, had tried without success such expedients as preserving fresh meat with sulphurous acid, or trying to win Japanese consumers over to salted meat. Japanese ships began cold-season shipments of fresh beef late in 1918, producing a sustained (if seasonally inflected) rise in Japanese imports (see Table 5). Thereafter, a growing number of small Japanese-owned companies plied ships (seventeen sailings in December alone) with a carrying capacity of between 400 and 600 carcasses from Qingdao to Japanese ports in Osaka, Kobe, and Shimonoseki, a trip that could be repeated once every three days. In that month, Japan imported just over 10,000 beef quarters, marking the first time that more than half of all Qingdao exports went to Japan. By 1923, four Japanese refrigerator ships were making the passage, allowing for fresh meat exports all year round. From that point on, nearly all of Qingdao’s rapidly growing beef export went to Japan, with a mere fraction of a per cent reaching Dalian and Shanghai. In 1925, Japan received all but 66,420 kg of the 11.7 million kg of frozen meat exported by the port.Footnote ⁸⁰

Table 5 Destination of Qingdao beef exports (in quarters), 1916–20

Source: COC, April 1921.

Beef imported from Qingdao occupied a particular niche in the Japanese market. Japan had already spent decades aggressively expanding its national herd, and produced much more than it imported. In part to protect domestic industries, the Qingdao product was often portrayed as inferior to Japanese beef (a group of Japanese butchers was prosecuted in 1921 for passing off Qingdao beef as domestic), but still commanded nearly twice the price of beef imported from Vladivostok.Footnote ⁸¹ Imports helped to bring beef within the reach of urban households. In 1920 Tokyo, 1 kg of lower-quality beef cost ¥1.40, roughly the same as ten domestic chicken eggs. While hardly cheap, beef was for many an attainable luxury.Footnote ⁸² This price continued to fall after the Japanese government waived import duties on beef and eggs after the devastating Kantō Earthquake of 1923. In 1924 Hiroshima, the price of beef was 16% lower than it had been in 1921.Footnote ⁸³

Coinciding with a devastating famine in northern China, this expansion of Japanese exports was hardly welcome. From 1921 to 1924, Japanese imports of Qingdao beef had doubled, to 60,000 head. Concerned that insatiable foreign demand would decimate the population of draft animals, one Chinese observer proposed moving the production of beef cattle to the northern grasslands, or the mountainous western frontier. A national law enacted in August 1923 aimed to limit production by levying separate taxes on slaughtering and sale of meat. While Chinese production of beef (including for its own domestic market) was never halted entirely, caps were placed on the number of animals that each slaughterhouse could process per day.Footnote ⁸⁴

Partially in response to these pressures, Japanese producers began consolidating their efforts, just as the industry was coming under ever-greater political control. Beef exports out of Qingdao were parcelled out to thirty to forty small exporters, no one of which still controlled as much as 10% of the total market.Footnote ⁸⁵ In 1924, most of these small exporters banded together to form the Qingdao Animal Husbandry Company (Seito saisaku kōshi). The following year, the Kobe Chamber of Commerce and Japanese consulate in Qingdao stepped in to mediate a dispute between the beef exporters and the shipping companies that provided refrigerated transport.Footnote ⁸⁶ Political influence over the trade soon came to assume a more official quality, similar to the sort of private–state hybrid companies that dominated Japanese trade in tea, flour, and other key commodities. This level of influence allowed beef exports to continue to expand, even as tensions with China mounted. By the early 1930s, the time of the surveys shown in Table 1, Qingdao was producing 75% as much beef as Shanghai, a massive international city that was roughly seven times larger.

Trade transformed dramatically in September 1931, when the Japanese garrison in Dairen initiated a three-month assault to capture Manchuria. This act, and the subsequent establishment of the Manchukuo client state, made Japan an international pariah, but also solidified Manchuria as a dedicated and captive trade partner for numerous commodities, including beef imports. Earlier that year, as floods wrought havoc across five provinces, the Chinese government had continued to reiterate its ban on live cattle and beef exports. One such announcement, specifically directed at the north-eastern provinces of Heilongjiang and Jilin, was released on 16 September, just two days before the sudden commencement of hostilities.Footnote ⁸⁷ The beef ban was repeated five days later but, unsurprisingly, at that point there was little hope of compliance. The Qingdao slaughterhouses were made exempt from the ban, technically because they were joint investments, and a source of foreign income.Footnote ⁸⁸ After the ban, they continued to operate, and remained Japan’s primary source of beef imports. Japanese-controlled Korea and Manchukuo also emerged as major, and strictly dedicated, exporters of beef to the home islands. Measured in tax revenue, nearly all of the chilled meat exported from Manchukuo in 1932–33 was shipped to Japan. Of all of the new nation’s many resource exports, approximately 31% of which went to Japan, only the trade in meat was so concentrated.Footnote ⁸⁹

No one road: the many faces of commodity globalization

China was a minor but hardly insignificant player in the global transformation of beef. There are no reliable statistics for the country’s total domestic production, but at the height of the South American beef boom in 1919, the single city of Qingdao exported about one-thirtieth as much beef as the entire nation of Argentina. Tallies from urban slaughterhouses in the 1930s confirm that Chinese beef production was substantial. Extrapolating these figures out to rural areas, where the vast majority of Chinese people lived, suggests that the industry may have indeed been one of the world’s largest.Footnote ⁹⁰

But the significance is not in the numbers. From a global perspective, China’s beef industries were connected to the Atlantic revolution in meat production, but in many ways remained aloof from it. Beyond the amount of beef produced, sold, and consumed, production out of China shows the same forces that shaped this better-known trade in a very different setting. The three stories presented here span roughly the same period, and contain many of the same elements, as the Atlantic transformation, including the late nineteenth-century transition from live trade to industrial slaughter and refrigerated transport, the explosion of investment in the two decades before the First World War, and political adjustments in the years that followed. The three production chains, shown together in Figure 3, span three successive modes of transportation: droving, rail, and oceanic shipping. Together they introduce a familiar set of forces: new technologies, domestic and foreign capital, and a range of strategic and political considerations, including policies intended to protect herds, stabilize border populations, monopolize production and import under state auspices, and procure meat for military and civilian populations.

Figure 3 Major cattle and beef trade routes, 1900–32. Map by Thomas DuBois.

The differences are at least as important. While export-oriented producers like Argentina and Australia developed smooth point-to-point chains that connected interior pastures to finishing, slaughtering, refrigeration, and shipping industries on the coast, Chinese production was more like a web, composed of individuated production chains that moved both live and processed animals to different internal and external markets. The diverse nature of Chinese production, and quick turnover of new markets, brought competition of all sorts. The global players that fought to control exports in South America or Australia were unable, or perhaps uninterested in trying, to dominate Chinese exports. In both Harbin and Nanjing, Vestey enterprises remained only one among many players, and no other globally significant producer of the time made deep inroads into the coastal trade. Japan’s successful domination of the Qingdao trade was less a function of productive efficiency than the willingness to use political clout, and eventually the road of open military aggression, to secure its many interests on the continent. If the Atlantic beef trade is a lesson in globalization, the Chinese one is a lesson in the inflection, and ultimately the limits, of those forces.

From a commodity chain perspective, the story of Chinese beef confirms Timothy Sturgeon’s caution against over-generalizing the forces that shape intermediation within an industry.Footnote ⁹¹ Simply put, there is no single, axiomatic nature that governs the historical development of the beef industry, in China or elsewhere. Even these three production chains exhibited remarkable diversity in how their product was made and brought to market. In the caravan trade, the purchase of live cattle was entirely buyer-driven, the time and terms of trade dictated solely by the Shanxi merchants, who held a monopoly trade in vital goods. In contrast, the beef trade was less visible because it was less distinct: cattle brought to the interior were traded in a series of markets, where they were mixed in with other animals, and sold for labour and transport, hides, and meat. Competition from cross-border Russian merchants, and later by the Harbin meat-packing industry, greatly raised the bargaining power of the pastoralists, but advantage still lay with the modern producers, who had the capital and connections to bring a superior product to new consumers in Vladivostok and eastern Russia. The Japanese industry in Qingdao was different again. Initially characterized by the decisive influence of entrepreneurial intermediaries such as William Katz, Qingdao became a captive node of the politically dominated supply chain to Japan. The three production chains were not merely geographically distinct, but their fundamental differences also encapsulate important evolutionary phases in an industry that maintained a distinct regional presence, neither global nor a casualty of globalization.