CHAPTER ONE Where the Rubber Meets the Road

PERMANENT LIVELIHOODS

Rubber is a relative newcomer to montane Southeast Asia, and especially to northern Laos. In the early twentieth century, French-colonial planters introduced the crop to southern Laos as part of their development of large plantation complexes in Cambodia and southern Vietnam. But their rubber varieties, imported from the Brazilian Amazon via British Malaya and London’s Kew Gardens,⁴ were adapted to the humid tropics and thus limited to the southern part of the Indochinese peninsula. The cold-adapted varieties found today in the mountainous uplands of southern Yunnan, northern Laos, and northern Myanmar were developed much later by Chinese agronomists during the Korean War, as China faced a Western blockade that shut the country off from the rubber supplies of Malaya, Thailand, and Indonesia. The Chinese varieties’ lower rates of production made them uncompetitive for export, but they served important domestic purposes. Helping create self-sufficiency during the embargo and aiding Chinese officials in their efforts to “modernize” the uplands of southern Yunnan, rubber played a key role in China’s Cold War–era agrarian transformation of its upland borderlands.⁵

The Cold War also helped bring Chinese rubber varieties to northern Laos. During the upheavals of the 1960s, ’70s, and early ’80s, a number of Lao refugees had fled across the border to Yunnan, where they learned rubber tapping in the state plantation system created in the 1950s around the new cold-tolerant varieties. In the late 1980s and early 1990s, some of them returned, bringing both the technical skills and the social connections that allowed northwestern Laos a degree of informal incorporation into China’s rubber sector.⁶ In 1991 Luang Namtha’s provincial Party Congress got on board, officially declaring rubber “a key poverty alleviation strategy and an instrument to stabilize shifting cultivation.”⁷ This signaled state enthusiasm for the crop as a “modern,” market-based livelihood distinct from the “natural economy” of subsistence production, and provided both political cover and a limited amount of economic support for some farmers to experiment with its development.

For Lao officials, rubber fit the governmental ideal of “permanent livelihoods” (asiip thavon or asiip kong thi). This concept drew on both indigenous and imported notions of proper settlement and economy. Generalizing the ideal of flooded-paddy rice production long practiced in the lowland parts of mainland Southeast Asia, it exemplified the legible farmer: fixed in space, market-integrated, and taxable by the state.⁸ The concept of permanent livelihoods also echoed colonial-era anxieties about shifting cultivation, which at the time was by far the dominant form of upland farming. French-colonial foresters echoed their British and Dutch colleagues across the region in calling indigenous shifting cultivators mangeurs du bois (wood eaters), and subsequent generations of officials and bureaucrats from India to Indonesia inherited the tradition of trying to “stabilize” shifting cultivation in favor of more commercially oriented and spatially fixed forms of agrarian production.⁹

While Lao authorities embraced rubber nationwide in the late 1990s and early 2000s, the crop had a particular significance in the north, where opium production had long been part of at least some upland farming systems. As a licit and legible crop, rubber symbolized stability and market integration where opium and upland rice evoked mobility and independent subsistence.¹⁰ And while the material similarities of opium and rubber were not lost on state officials—a provincial official once explained to me that both were latexes extracted through skilled-labor tapping—this was seen as a good thing because it meant that farmers familiar with opium production would readily take to rubber. But in the boostering that surrounded rubber development in the early 2000s, the differences were what mattered. Rather than being harmful and subversive, rubber was seen as secure: economically, socially, politically. It was a permanent livelihood par excellence.

This was easier said than done, however. Throughout the 1990s, as Laos had transitioned out of the socialist-bloc isolation of the late Cold War and into regional and global markets, officials across the country had exhorted upland farmers to intensify and sedentarize their agricultural practices. Often this “encouragement” came in the form of a village-scale zoning program called Land and Forest Allocation (hereafter LFA), which was rolled out in thousands of upland villages in the late 1990s and early 2000s. But despite leaving behind colorful land-use maps and pages of rule-laden documents, LFA failed to provide the resources—in particular the financing—that would allow the intensification conjured by its maps. The program thus had the effect of criminalizing shifting cultivation further by drawing village-scale land-use maps that could only be followed if long-fallow farming was abandoned.¹¹

Rubber first emerged in contrast to this top-down approach to sedentarization, as a smallholder-led initiative in a handful of borderland villages in Luang Namtha, where residents had access to two ingredients missing from LFA: social networks that extended into the Chinese rubber sector, and credit.¹² This happened most famously in a Hmong village just outside the provincial capital, where a former vice-governor helped members of his home community (some of whom had recently returned from China) get a sizable loan from a state development bank to finance the establishment of their plantation.¹³ Planted in the mid-1990s, their rubber matured to tapping age in the early 2000s, just as Lao-China development cooperation took off in the wake of China’s recently announced “Going Out” policy. This village, Ban Hat Nyao, was not the only place where state banks provided financing to establish rubber plantations, or where smallholders successfully mobilized capital to establish their own plantations. But it became by far the most famous, exemplifying for some the possibilities of smallholder entrepreneurialism and for others the need for organized management of the five key ingredients for rubber production: financing, market access, extension, land, and labor.¹⁴ The latter lesson would later form the grammar of the “3 + 2” policy.

For many Lao officials, this first generation of Lao rubber provided a proof of concept, while also emphasizing the imperative for outside assistance. China’s rubber sector, widely seen as the only commercial outlet for rubber produced in northern Laos, remained protected; despite joining the World Trade Organization in 2001, Chinese leaders had successfully exempted the sector from liberalization by referencing the Western blockade of the 1950s as evidence of its “strategic” importance.¹⁵ Lao rubber producers thus faced significant hurdles in selling their crop to processors in Yunnan, and often relied in the early years on informal methods to get their crop across the border.¹⁶ If rubber production in northern Laos was going to scale up, this would have to change.

Moreover, the key role of the state bank loan in enabling Laos’s most widely celebrated smallholder rubber success story highlighted the significant financing that would be needed if rubber was going to power the agrarian transition to “permanent livelihoods” that Lao authorities hoped for. While the cost of establishing a new rubber plantation would become a matter of some debate, the capital inputs were significant. In Luang Namtha alone, provincial Agriculture and Forestry officials had, by 2002, classified almost 200,000 hectares as land that was “suitable for rubber or eucalyptus plantations.”¹⁷ With authorities across the region thinking big about rubber, the roughly $400 per hectare needed for seedlings and other capital inputs was widely seen as prohibitive for both farmers and the Lao state.¹⁸

These issues were taken up when Chinese president Jiang Zemin made a state visit to Laos in November 2000 and signed a joint declaration on development cooperation, along with economic and trade agreements that included agricultural production.¹⁹ This diplomatic push helped usher in the northern Lao rubber boom of the mid-2000s by addressing financing and export in ways elaborated below. Widely framed as a win-win-win for Lao farmers, Lao officials, and Chinese companies, bilateral rubber-cooperation rhetoric combined social, economic, and environmental objectives. As one proposal put it at the time: “The population will receive permanent livelihoods, will be able to alleviate their hardship via this stability, and will have an elevated standard of living compared to the past; the state will benefit from reforestation, protection of the environment, and increased taxes and fees; and the investor will benefit from rubber processing and trade.”²⁰

“UNBLOCKING” LAOS

Bilateral rubber cooperation between Laos and China blossomed at a time when governments from North America to Europe to East Asia were widely embracing regional connectivity in the name of economic prosperity.²¹ In the 1990s and early 2000s, the once “isolated” and “remote” borderlands of northwestern Laos thus became progressively enrolled into regional imaginaries like the “Golden Quadrangle,” the “Greater Mekong Subregion,” the “Northern Economic Corridor,” and (later) China’s “Belt and Road Initiative.” As a 1997 Asian Development Bank proposal for the Northern Economic Corridor put it, the goal of these efforts was to reanimate “natural economic areas” like the one spanning northern Thailand, northern Laos, northern Myanmar, and China’s Yunnan province, which had once “thrived” as part of the ancient silk route between Sichuan and Assam but had been “disrupted by colonialism and the Cold War.”²² Such regional boostering channeled state and private resources into new geographies of connection and played on a mix of similarity and difference. If northern Laos’s “location, weather, and land” were “suitable for rubber planting” just like in Yunnan, bilateral cooperation was also premised on an economic division of labor. Laos would be the producer, and China the financier, processor, and consumer.

Efforts to enhance regional connectivity were, of course, nothing new.²³ In the 1860s French-colonial explorers had taken a great interest in the connectivity of northern Laos, which they reluctantly termed a “region of rapids.”²⁴ France’s Mekong Exploration Commission had been hoping to discover a river route to southern China that would have allowed Phnom Penh and Saigon—farther down the Mekong and, at the time, newly under French possession—to compete with British Hong Kong for access to trade within the Chinese interior. When the Mekong turned out to be hopelessly unnavigable, the commission turned its attention to the territory itself. Issues of connectivity interested them highly, as evidenced in their report, written by the geographer and military officer Francis Garnier. Highly impressed with the northern Lao city of Luang Prabang—“the first time since our departure [from Phnom Penh] that we had found a market in the sense this term has in Europe”—Garnier linked the economic prosperity of what the French called “upper Laos” to its isolation: “The distance of Luang Prabang from the theater of the wars which tore Indo-China apart in the eighteenth century contributed greatly to assuring its prosperity, no doubt after having been one of the determining causes of its foundation.… Today, the kingdom of Luang Prabang is the most important Laotian center in all Indo-China, the place of refuge and the natural focus of support for all the peoples from the interior who want to escape from the despotism of the Siamese.”²⁵

French interest in Laos’s connectivity with its neighbors was, of course, highly geopolitical. Historically, “upper Laos” had been linked by overland and river-based trade routes that ran from southwest to northeast, connecting regional centers like Luang Prabang and Muang Sing (in present-day northern Laos) to places like Nan and Chiang Rai (today in northern Thailand) and Jinghong and Kunming (in present-day Yunnan) and beyond.²⁶ In the late 1800s these connections posed a real challenge to the French, whose recently formalized protectorate over Cambodia was just the latest step in assembling what would become French Indochina. During the commission’s stay in Luang Prabang in 1867, Garnier waxed poetic about offering a similar “protection” arrangement to the Lao king there, hoping the French might replace the historical role of the Chinese empire in “exercising a domination benevolent and wise, which stimulated production … and increased the welfare and vital energies of the subject populations.”²⁷ In return for protection, he hoped for infrastructural changes (both political and material) that would undo the inconvenient truth that upper Laos was far more closely linked to Siam than it was to Cambodia or Vietnam.²⁸ The French, Garnier wrote, “would only ask him [the Lao king] to favor the development of commerce toward the southern part of the [Indochinese] peninsula, to help us do away with fiscal hindrances, and to improve the roads in this direction.”²⁹ This idea, it turned out, was premature—but not by much. Luang Prabang sought and received French “protection” in 1887 amid a deadly mix of refugee flight, social banditry, and local uprisings linked to China’s suppression of the Taiping and various Muslim rebellions in the 1870s, and by Siam’s efforts to consolidate its periphery against rising European interest in the chaos that followed.³⁰ By the early twentieth century, when French concerns were focused not on acquiring Lao territory but on governing it, this same concern with connectivity took the form of what colonial administrators called “unblocking.”³¹

“Unblocking” was a strategic description par excellence: it was not that Laos was “blocked” per se but that it was connected in the wrong directions. French administrators treated Laos until the 1930s not as a distinct national space but as a resource-rich hinterland for Vietnamese industry and excess population.³² In the early twentieth century, colonial administrators thus sought to minimize Laos’s connections to Siam and China while building roads to Vietnam over the Annamite Mountains and, to a lesser extent, constructing railways around the unnavigable sections of the lower Mekong. These infrastructure plans were consistently undercut, however, by a reluctance to spend scarce state resources. Colonial officials thus left the task largely to a mix of corvée labor (“tax” paid by the local population in the form of work, often on roads) and private-sector investment, often from Europe. Both spectacularly failed to deliver at scale; the former was beset by local resistance while the latter crashed hard in the Great Depression.³³ Laos thus entered the period after the Second World War as a regionally “remote” and “isolated” country, a leading example of what would come to be called Asia’s vast infrastructure deficit.³⁴

While the Cold War period is largely remembered as one of regional fracturing—as in the Asian Development Bank (ADB) proposal quoted above, and as elaborated in the next chapter—the 1950s and 1960s also saw the birth of a new imaginary of connectivity based on the Mekong River. Like French efforts to “unblock” Laos, this imaginary was more often aspirational than actual, but it nonetheless inspired the idea of a Greater Mekong Subregion (GMS), a vision of regional connectivity that embraced both old forms of French-era connectivity (between Laos, Cambodia, and Vietnam) and earlier historical forms (between Laos, China, and Thailand) that the French had sought to minimize or interrupt. Articulated in the early 1990s as Southeast Asia’s regional leaders sought to “turn battlefields into marketplaces” at the end of the Cold War, the GMS, like other regional imaginaries of the neoliberal era (and like the BRI, into which the Northern Economic Corridor has subsequently become enrolled), posited shared prosperity through enhanced connectivity and comparative advantage.³⁵ While exemplifying the shades of economic imperialism that critics called Thailand’s “resource diplomacy”—strengthened political ties with Laos, Cambodia, and Myanmar seemingly aimed at helping Thai businesses access those countries’ timber, mineral, and hydropower resources—the GMS was a distinctly multilateral initiative.³⁶ Yunnan was also part of the GMS from its initial formalization in 1992, and made up the northern end of multiple regional “economic corridors” that gave the GMS its internal structure (map 1.1). Lao leaders, having long viewed the country’s landlocked status as a hindrance to development, widely embraced the “corridor” approach in hopes of becoming instead a “land-linked crossroads” of the region.³⁷

Laos’s Northern Economic Corridor (hereafter NEC), a major road upgrade built between 2002 and 2007, thus played a key role in connecting wider regionalization efforts to the geography of Lao-China rubber cooperation. Linking southern Yunnan with northern Thailand through northwestern Laos, the NEC had been originally envisioned back in the late 1980s as the Lao portion of the so-called Golden Quadrangle, a ring road that boosters hoped would one day connect the region’s two biggest economies via both northwestern Laos and the eastern part of Shan state—essentially a bid to have Laos and Myanmar compete with each other for lower transport costs.³⁸ In 1994, with the Shan portion hampered by ongoing political instability, ASEAN representatives declared Laos’s NEC a “high priority” project, and after a delay due to the 1997 Asian economic crisis and an unsuccessful effort to get the road built by a private concessionaire, the project went ahead in the early 2000s. Under a joint agreement facilitated in 2001 by the ADB, Laos agreed to finance the road using loans—each covering roughly a third of the project—from Thailand, the ADB, and China.³⁹

ADB planners enthusiastically described the NEC as an “initializing project” that would “serve as a ‘locomotive’ for subregional economic development … along the north-south axis” of the GMS.⁴⁰ This description proved to be apt, although the development that emerged would be highly uneven. In 2004, as construction was beginning, the bank’s consultants worried that while the project would “automatically benefit” entrepreneurs in China and Thailand, it would not do the same for “subsistence-oriented ethnic minority shifting cultivators inhabiting the road impact zones”—in other words, uplanders like the residents of Khet Nam Fa. “Hence, there is a strong risk that the economic opportunities and benefits of road upgrading will bypass these rural communities, which will still be exposed to the associated potential negative consequences.”⁴¹ These fears proved more than justified.

“CERTAIN CONTENTIOUS ISSUES”

In early 2001, as the NEC agreement was being finalized, and in the wake of Jiang Zemin’s state visit to Laos, provincial officials in Luang Namtha outlined a plan to their subordinates. It focused on developing 10,000 hectares of new rubber plantations in cooperation with the Sino-Lao Rubber Company, a joint venture between Chinese rubber and import-export companies with whom they had been in discussions since the previous year.⁴² Writing to provincial- and district-level agricultural extension agents, they outlined the respective roles of what they called the initiative’s two “sides”:

The province of Luang Namtha agrees to facilitate rubber planting and factory construction by convincing upland farmers to give up shifting cultivation-based livelihoods and plant rubber; to assign government staff from the relevant offices in order to coordinate with the Chinese investor side; and to implement rules, laws, etc. in order to assist the Chinese investor side in working in accordance with the policies of the exporting country [i.e., Laos]. The Chinese investor side will be solely responsible for investing in the building of the rubber-processing factory, contributing the relevant technical inputs, and providing the market for Luang Namtha’s upland farmers by buying their rubber and processing it for export.⁴³

Carrying echoes of the 1991 Party decision on rubber, this passage summarizes a development cooperation model based on contract farming, with the Chinese investor “side” in the role of the contractor (providing inputs, technical specifications, and a guaranteed market) and the province in the role of recruiting Lao farmers to join the scheme. Rubber is framed in the language of permanent livelihoods, and the organizational division of labor evokes at once the sort of comparative advantage envisioned by regional economic boosters and the mutual assistance typical of South-South cooperation since the 1950s.⁴⁴

Contract farming, however, was not the vision of cooperation that Sino-Lao Rubber Company representatives had in mind, as evidenced by the conflict that followed. For the next five years, Luang Namtha provincial officials and Sino-Lao representatives remained at odds over the question of whether to grant the company a large land concession, an arrangement that would have let Sino-Lao develop its own plantations using hired labor. The half decade from 2000 to 2005 followed the launch of China’s “Going Out” policy, and came both in the immediate wake of Jiang Zemin’s state visit to Laos and at the beginning of what has since become arguably Laos’s most important bilateral relationship.⁴⁵ Yet it saw a flagship project of bilateral rubber development—Sino-Lao’s investment in Luang Namtha, a province often called Laos’s “gateway” to China—stall because of what well-placed observers, in early 2005, called the persistence of “certain contentious issues.”⁴⁶ This conflict concerned the details of transnational land access, and it played out broadly across northwestern Laos in the early and mid-2000s between Chinese rubber companies and various levels of the Lao government. Sino-Lao figured especially prominently in it because the company was negotiating with provincial authorities in not just Luang Namtha but also neighboring Oudomxai and Bokeo provinces (located to the southeast and southwest, respectively). This seems to have helped drive the three-province decision in late 2005 that originally drew me to the northwest. But the struggle over the concession issue was hardly limited to a single company.⁴⁷

A pair of competing proposals from this period are worth looking at for a few reasons. On the one hand, they illustrate the extent to which regulatory politics at the time focused on the question of Lao state assistance for Chinese companies. At the time, this was being negotiated separately by a handful of different companies across various locations and jurisdictions, so the proposals provide a view into the mechanics of an already emerging geography of uneven enclosure. In the proposal language, the question of state-managed enclosure manifested as a debate about which business model—company-controlled plantations or smallholder contract farming—would form the core of the 10,000-hectare Sino-Lao partnership being planned in Luang Namtha. (The 30,000 hectares in the passage quoted below refers to the three provinces of Luang Namtha, Oudomxai, and Bokeo collectively.) On the other hand, the two proposals also show the significant room to maneuver that existed in policy language, and that was exploited by companies like Sino-Lao, Bolisat Ltd., and a few dozen others in their efforts to secure land access during the boom years of 2003–8. In practice, these ventures reflected aspects of both proposals, which helped maintain the discourse of smallholder-centric development while also creating quasi-concessions like the ones in Khet Nam Fa.

Titled “Draft Plan for Cooperation in Rubber Planting between Lao PDR and the PR China, 2005–2007,” the first proposal was circulated by the Lao Ministry of Agriculture and Forestry shortly after a visit by Sino-Lao representatives in early 2005 aimed at resolving the “issues” mentioned above.⁴⁸ The document was addressed to the ministry’s provincial-level offices in Luang Namtha, Oudomxai, and Bokeo, and outlined a concrete plan for rapid, yet still smallholder-centric, rubber cooperation. Covering topics from physical geography to economics, it converged on a proposal that echoed the contract-farming language of provincial officials, quoted above, from a half decade earlier. This began by dangling the carrot of 30,000 hectares of land and corresponding labor availability:

Laos has an area of 236,000 square kilometers … [and] three target regions for rubber [in the north, center, and south].… These three regions comprise sixty percent of the country, or 140,000 square kilometers, within which twenty percent is capable of production, or roughly … 2,800,000 hectares. Therefore, planting 30,000 hectares of rubber [in the northwest] will not be a problem.… Based on national statistics, Luang Namtha contains roughly 150,000 people; 23,000 families and 50,000 labor units [i.e., adult laborers]; Oudomxai roughly … 80,000 labor units; and Bokeo … 45,000 labor units, providing sufficient labor capacity to plant an additional 30,000 hectares of rubber.⁴⁹

After conjuring a landscape ripe for rubber development, the proposal then explained that four-fifths of this area was intended for contract-based smallholder production, with the remaining 20 percent under direct company control. This plan was offered under the heading “Company + People” (Bolisat + Pasason), specifying 10,000 hectares per province, each to be divided into 2,000 hectares of company plantations and 8,000 hectares for local farmers. It also laid out a timeline for getting these planted rapidly, by the end of 2007.⁵⁰ Crucial to the ministry’s strategy was the balance of give-and-take: it was offering to actively “supply and provide land” for the project, “organize the population to participate in a united and disciplined manner, and coordinate the relevant state organizations.” On the other hand, the land promised directly to the company was comparatively small: the 2,000 hectares offered in each province was four times less than the land planned for contracted smallholders.

Three months later, Sino-Lao came back with a very different counteroffer. It began on the same terrain as the ministry’s plan, noting Luang Namtha’s proximity to Yunnan, its favorable mix of land and labor availability, and the mutual benefit that each side would gain by working together.⁵¹ But on the details that mattered most, Sino-Lao’s plan was far less inclined toward working with smallholders. Against the ministry’s proposal of 80 percent smallholder land, the company proposed that “regarding the gardens produced by cooperative investment with the population, these will be divided 50–50 by total area via a division [between the company and local farmers] that will occur one year after planting, and then each side will take care of its own.”⁵²

On one level, this can be read as a modest, if nonetheless significant, counterproposal. While a 50–50 rather than 80–20 split would hardly have been a trivial alteration, it could still have been interpreted as a compromise, proposing an equal division between the company and smallholders rather than a model dominated by either side. But a second crucial issue concerned the timing: Sino-Lao proposed that the plantation be divided “one year after planting.” This implied a planting process that was intensively planned and managed, and would have needed to include an active reconfiguration of land-tenure arrangements. Rather than distributing seedlings to farmers who would then plant them on their own land and with their own labor, Sino-Lao’s proposal relied on the up-front development of company plantations with wage labor; the 50–50 division would come subsequently. Such a plan was closer to the model of state farms or resettlement areas (nikhom in Lao and Thai), where rubber tapping might be farmed out to individual households or labor groups, but the initial planting was company-managed and based on alienated land and labor.

A second indicator of the distance between the two “sides” was how they valued their own relative contributions (table 1.1). Input values are crucial for negotiating cooperative business arrangements because the distribution of the final product should, in theory, reflect each side’s contribution. Using a per-hectare basis, the ministry’s “Company + People” proposal estimated the total cost of establishing a plantation and maintaining it for one year at almost $1,500; of this, it estimated that almost 70 percent came from smallholder labor inputs like land clearing, terracing, fertilizing, planting, weeding, and spraying. (The balance went mostly to seedlings, but also included things like pesticides and agricultural extension.) Sino-Lao’s proposal was accompanied by a set of figures that roughly mirrored the ministry’s categories. But the company’s counteroffer came in much lower, largely owing to how it valued smallholder labor. Against the ministry’s figure of almost $1,500, Sino-Lao estimated overall plantation establishment costs at just $534, and within this much lower estimate, it valued the share of smallholder labor at just over a third of the total, against the ministry’s estimate of over two-thirds. Each side’s valuation supported its own proposal, and each differed most in its valuation of the scheme’s central ingredient: the upland smallholder. Thus, as late as mid-2005, despite their agreement in principle—on cooperating to stabilize shifting cultivation, alleviate upland poverty, and create mutual economic benefits—the two “sides” remained miles apart when it came to how to actually work together.

AREAS OF INFLUENCE

My own research in the northwest began during this period of nascent but uncertain bilateral rubber cooperation on the one hand, and rapid construction of the Northern Economic Corridor on the other. While often discussed as separate sectors or projects, rubber and roadbuilding were in fact closely linked, both being parts of the wider effort to connect the hinterland of northwestern Laos with surrounding areas.⁵³ More specifically, road-building drove rubber planting. Suitability for rubber, after all, is not just a function of biophysical characteristics like soil quality, temperature, topography, and water availability. It is also intimately social, depending in particular on accessibility, availability of land and labor, and complementarity with other local sources of livelihood.⁵⁴ The building of the NEC rearranged all of these social “variables” significantly.

The risk of various forms of land grabbing was identified during the NEC planning process; road corridors are a relatively mature technology, and their effects on newly opened hinterlands have been recognized for centuries.⁵⁵ While critics of the GMS and similar geographical imaginaries often interpret enhanced regional connectivity in terms of states’ and businesses’ ongoing search for cheap resources—what economic geographers call capitalism’s “spatial fix”⁵⁶—infrastructure practitioners tend to deal with the downsides of economic expansion in the language of risks, acceptable trade-offs, and mitigation plans. At the scale of specific projects like the NEC, a key question developers faced was thus how far outward in space the corridor’s “area of influence” stretched. This had been flagged as an empirical question back in the late 1990s when the ADB took over the project, and it was still being addressed in 2006 when I began my fieldwork.

“Project area of influence” is a term of art that entered the development lexicon as a result of the struggles over accountability in the 1980s and 1990s.⁵⁷ On the one hand, it allows lenders of public money like the World Bank and ADB to evaluate, measure, acknowledge, and discuss the range of impacts that their projects create, often at significant physical distances from projects themselves. “Areas of influence” are thus part of a wider discourse of spatially nuanced impact analysis that includes terms like direct impacts, indirect impacts, induced impacts, cumulative impacts, regional impacts, and in-combination impacts.⁵⁸ These terms operate inevitably within what Michel Foucault called power-knowledge, wherein the meanings of words and the statements they produce are linked to high-stakes questions like, in the case of the NEC, how far from the road the protective mitigation activities required by the ADB would extend. Operating within the norms shared by multiple development banks, the mitigation of “indirect” impacts gave the NEC’s developers significantly more latitude than “direct” ones.⁵⁹ Specifically, the notion of a “project area of influence” allowed planners to acknowledge that the NEC would likely have significant negative impacts away from the immediate roadside, while nonetheless limiting their mitigation activities to the roadside itself. This had important consequences for the rubber development efforts that were getting started in the early 2000s, just as the NEC’s mitigation studies were getting off the ground.

The ADB hired the American firm Nathan Associates to do the project’s social and environmental impact assessment; this began in 2002 and was completed early the following year, right in the middle of the Sino-Lao rubber impasse described above. The Nathan study made it clear that the NEC’s anticipated “area of impact” would extend well beyond the ninety-seven villages through which the road would pass directly. In addition to the roadside villages, “other villages, which are not seen from the road but are affected by its commerce, further expand the ethnic and cultural diversity found in the Northern Economic Corridor. Most of the inhabitants of these villages are engaged in subsistence agriculture … [and have] some of the highest rates of poverty in Laos.… Appropriate development measures taken in conjunction with the improvement of the Project Road will be needed to help to bring these inhabitants to above poverty level standards.”⁶⁰

This call for “appropriate development measures” was elaborated in the study’s section on social impacts, which, along with the ADB’s final recommendation for financing to the bank’s board of directors, highlighted a pair of issues that have proven prescient. The first was the mismatch between Lao property law, which recognized “intensive” land uses like lowland rice paddies and tree plantations—permanent livelihoods, in other words—and the “extensive” and often illegible land uses like shifting cultivation, livestock grazing, and forest harvesting that predominated in the NEC’s area of influence.⁶¹ Tightly echoing the ADB’s own recently published Participatory Poverty Analysis for Lao PDR,⁶² the Nathan study noted the “severe hardships” caused by LFA (the government’s village-scale mapping program mentioned above) throughout Laos’s northern uplands via its largely unsuccessful effort to create “ ‘permanent’ livelihood substitutes.” These included exacerbated land-tenure insecurity (“issuing … land certificates to upland farms only if they meet the ‘permanent’ criteria”) and increasing soil exhaustion as a result of the shorter upland fallow rotations induced by the zoning process.⁶³

Second, and somewhat ominously, the chapter on social impacts noted that a huge amount of land—some 196,615 hectares in Luang Namtha alone, comprising 82 percent of the province’s officially designated agricultural production land—had been classified by provincial authorities as “suitable for rubber or eucalyptus plantations.”⁶⁴ This pairing of policy-induced tenure and livelihood insecurity with the conjuring of large-scale conversion to industrial plantations gave substance to the report’s call for appropriate measures to protect upland villages. It also echoed requests from local officials documented in the NEC’s Social Action Plan for “assistance to protect the needs and rights of the existing communities” from the already-apparent commercial pressure on the corridor’s land base.⁶⁵

The NEC’s planners ultimately settled on a pair of compromises that, proponents argued, would help prevent land grabbing within the corridor.⁶⁶ The first was to issue land-tax certificates to residents whose land parcels fell within fifty meters of the NEC’s centerline. This distance was judged to be that of “direct” impact and led to the documenting of over seven thousand parcels along the length of the road.⁶⁷ While these documents were widely referred to as “titles” by foreign consultants I met in the field as well as by subsequent official reports, they were in fact land-tax documents that provided a limited form of recognition—that of existing use and thus the “right” to pay tax—rather than a future right to permanent use. This point was emphasized by the Nathan-study authors who, writing just a few months after ADB planners had speculated that “all households in the project villages [might] be issued with land titles on a priority basis,”⁶⁸ noted that “the government places higher priority … on extending the land tax document to all homeowners as a means of increasing tax revenue and broadening administrative documentation.”⁶⁹ The NEC’s so-called titling component was thus widely perceived as a twofold compromise, given its spatial narrowness and the limited form of security it conveyed.⁷⁰ Various people I spoke with in 2006 acknowledged these shortcomings but expressed hope that any form of land documentation was better than none, given the rising pressure on land that was already becoming apparent from, among other things, rubber projects.⁷¹

The second piece of the compromise addressed the land outside the fifty-meter buffer but still within the NEC’s official area of influence, which was defined as the ninety-seven villages through which the road passed directly as well as twenty-five “other villages” mentioned above that sat away from the road but were nonetheless “affected by its commerce.”⁷² (This was a conservative accounting; as noted elsewhere in the Nathan study, the NEC was “more realistically” defined in terms of the full three districts through which the road passed.⁷³) Reflecting the room to maneuver on so-called indirect impacts, NEC planners removed responsibility for areas outside the fifty-meter buffer to a different project entirely: a grant-based initiative from the Japanese government. This project sought to pilot what the Nathan study called “scientifically designed, participatory, and well-implemented land-use planning and allocation in conjunction with integrated rural development and capacity building.”⁷⁴

This spin-off project targeted entire village territories rather than just the immediate roadside strip. But it came at a cost: the nine-village pilot was budgeted at $1.5 million, which would have translated to roughly $20 million if applied to all 122 villages in the NEC’s official area of influence. As a development professional affiliated with the NEC told me in 2006, even $10 million would have been “way too much” for the project’s mitigation budget; the loan needed to stay simple and lean, he explained, rather than trying to “do everything.”⁷⁵ Moving the burden of dealing with the land-grab threat to a different project thus allowed the ADB and the three national governments to have their proverbial cake and eat it too. As they proceeded with construction in 2004, they had a loan with an impressive 20.9 percent estimated internal rate of return.⁷⁶ They also had at least a claim—for themselves as well as for any potential critics—that the land-grab threat was being taken seriously.

AVAILABLE LAND

Events on the ground rapidly proved otherwise. Throughout the mid-2000s, numerous land deals involving “investors” of various kinds took place within the NEC, especially in the areas beyond the project’s fifty-meter mitigation buffer. Some of these were just a few minutes’ walk from the main road, like the large rubber plantation (thirty hectares or so) established in one of Vieng Phoukha’s roadside villages by a businessman from the district capital. Established in 2005, this deal was especially embarrassing for the organizers of the spin-off project discussed above because it sat in the middle of their main pilot village’s agricultural land.⁷⁷ While it exemplified the “delay in implementation” that the ADB would later use to rate the project less than satisfactory,⁷⁸ in fairness it simply illustrated the difficulty of stopping a land rush that was already underway. It was, moreover, entirely typical: entrepreneurial, ad hoc land-finding was taking place across the northwest, and indeed more generally across rural Laos.⁷⁹ Typically, it followed the roads.

During my research I heard multiple descriptions of how various entrepreneurs—foreign and domestic, private and state-affiliated—gained access to land; these ranged from accounts of “informal” or personal access to highly bureaucratic processes of approaching this office, then that, then another, and producing documents X, Y, and Z at each stage along the way.⁸⁰ While the differences often testified to internal jurisdictional struggles within the state (see ch. 5), their agreement on the need for actual, on-the-ground surveying and negotiation highlighted a common disregard for existing land-use maps. And while this sometimes frustrated technocratically inclined participants and observers both in and out of government, there were good reasons for going to the field to survey. One was the dynamism of the moment—the confluence of the rubber-planting boom and the new hinterland access created by the NEC. A second related to a longer-standing issue with how village-scale maps were produced. As noted above, LFA maps often depicted land use in aspirational terms, and the impossibility of farmers following them meant that across the country, both poor farmers and wealthy entrepreneurs had been violating them for years. Occasionally, LFA’s restrictions on land use came up in my and my colleagues’ discussions with farmers, but often things tended in the other direction.⁸¹

Let me give an example. In late 2007, my research assistant and I had a chance to speak with a man who was in the process of developing a modest (roughly two-hectare) rubber plantation. We had been traveling on one of the NEC’s feeder roads, about ten minutes from the main highway, when we spotted the telltale contours of a new rubber garden. The man was working with two laborers, and he chatted with us while the three took a break from hacking the new terraces out of the hillside. He was from the district center, and explained that he had recently purchased the land from a relative who lived in the village where the plot was located. Recently cut rice stalks were still standing, showing that the field had been used for shifting cultivation the previous growing season. When we pushed for the details of the sale, he told us readily that it had been approved by the local village head, and that he planned to take the sale documents to the district tax office to register it as soon as he finished planting. Over the weeks that followed, the plantation took full form. The terraces were completed within a week of our conversation, and within two weeks they were completely planted with young rubber seedlings.⁸²

The man’s account echoed what we had been hearing from local officials. According to local tax officials, land sales were typically approved at the village level and came under their (the tax department’s) purview relatively late in the process—often not until the new owner brought the previous owner’s land-tax receipt (transferred during the sale) to pay the next year’s land tax. The sale was also divorced from the zoning map, which was kept in the local Agriculture and Forestry office and should, in theory, have been consulted before the sale was approved. This did not necessarily happen, however, and this particular sale had almost certainly been illegal because it sat inside the village’s “local use forest” (pa somxai)—the area where residents were permitted to gather building materials and nontimber forest products.⁸³ As evidenced from the previous year’s upland rice stalks, the parcel had actually already been out of conformance with official zoning, even before the sale. This too was unsurprising. Local Agriculture and Forestry officials had told us a few times that LFA was unpopular with farmers in the area because of its efforts to limit them to three agricultural plots. Many had thus expanded their fields into new areas—often including those zoned as “forest”—for various reasons. In one of our meetings, a government staffer who knew the local situation especially well confided that these expansions were not just due to demographic causes; they also reflected the increasing encroachment of rubber on land used for food production.⁸⁴

Rubber companies like Sino-Lao and Bolisat Ltd. fit into this dynamic landscape in multiple and often complex ways. On the one hand, despite being part of the wider land rush, they were also subject to a regulatory check in ways that smaller land deals by local elites were not. In Vieng Phoukha, for instance, it was clear that Bolisat Ltd. was in competition with people like the land buyer described above; both were pursuing land along the NEC’s feeder roads in the same general vicinity. One of my informants told me that local elites were in fact strongly opposed to Bolisat Ltd.’s operations because the company was beating them to land that they were hoping to develop for themselves. The district governor, my informant explained, had embraced working with Bolisat Ltd. for precisely this reason. Even if it meant giving up land to the company’s plantations, the logic was apparently that doing so would be more manageable than having poor villages’ land base eroded piecemeal by numerous and essentially unregulated land sales.⁸⁵

This protective dimension was also apparent in the creation of the “3 + 2 policy” in late 2005, which was put forward as a putative solution to the impasse described above in the Sino-Lao case. The agreement was described by a team from Laos’s National Agriculture and Forestry Research Institute in language that emphasizes its departure from the concession model:

A meeting was held on 10 October 2005, in Luang Namtha, where representatives from … Bokeo, Luang Namtha and Oudomxay gathered to discuss foreign investment in rubber. Representatives from the three provinces agreed that providing land concessions to investors to manage rubber plantations will not resolve rural poverty, as farmers lose access to agricultural land and are merely hired by investors as laborers. On the other hand, representatives of the three provinces agreed that foreign investors should promote smallholder rubber plantations with a general profit-sharing arrangement of 70% for farmers and 30% for companies. They also agreed to support foreign investors that are willing to provide inputs on credit, and purchase latex from farmers.⁸⁶

The “3 + 2” agreement was also sometimes called the songserm policy, using the Lao term for “promotion” (as in the paragraph above: “investors should promote …”). It was also formally spelled out in the meeting minutes authorized by Luang Namtha’s provincial governor: “The cooperative investment mode is hereby agreed to be the 3 + 2 model, namely: Investors are responsible for three aspects: (1) capital, (2) technique, and (3) marketing. Villagers are responsible for two aspects: (1) labor and (2) land, in accordance with state land management.”⁸⁷ While both the final caveat on “state land management” and the concept of “promotion/extension” (songserm) left important room to maneuver, the expectation that bilateral rubber cooperation in the north would operate under a smallholder- rather than a concession-centric business model was widely shared.⁸⁸ Indeed, when problems with Vietnamese rubber concessions in southern Laos blew up as a national issue in 2007, Laos’s prime minister held up the “3 + 2” model as the “strongly promoted” alternative when he announced a national moratorium on concessions.⁸⁹

On the other hand, the land-finding efforts of at least some Chinese rubber companies proved quite successful. In part, this was no doubt due to their employing the same sort of opportunistic resourcefulness as local Lao entrepreneurs. In one of my interviews at a Chinese rubber company office in Luang Namtha, for instance, the bulk of the talking was done by an older Lao man who was a former provincial Industry and Commerce official and had gone to work for the company after retiring from his government job.⁹⁰ As I elaborate in chapter 4, this sort of reliance on local officials to navigate the terrain of land access was highly successful in areas like Khet Nam Fa.

But another part of rubber companies’ success involved the logic of wage work, which fit the economic realities of upland livelihoods much better than the “long pay” of contract farming with a slow-growing tree crop. Rubber may have fit the ideal of the permanent livelihood, but successful smallholder rubber typically involves at least some degree of state support during the six to ten years between initial investment and harvest.⁹¹ This support is missing from all accounts of “3 + 2” that I am aware of,⁹² and is reflected in a common answer to the question of how farmers were supposed to survive during the transition period: that it is possible to intercrop upland rice “for a few years.” A flimsy excuse, this nonetheless acknowledged the difficulty of the transition for farmers who depend on annual production for their food and livelihood security.

It was thus unsurprising that even the allegedly smallholder-friendly “3 + 2” schemes proved widely unattractive to most upland farmers. This was already apparent when I was doing fieldwork in 2007 and 2008. The lack of farmer interest was often explained to me in terms of the splits being offered: as I heard from numerous village heads, even the 70–30 split outlined in the official version of “3 + 2” was widely seen as unattractive since farmers wanted to own their plantations outright; many actual “3 + 2” deals were even worse, offering 60–40 or even 50–50.⁹³ But while this gave a reason, it also avoided a larger and more uncomfortable issue, given the widespread official boostering for rubber as a livelihood-improvement mechanism for the upland poor. Time and again, the “smallholder” rubber producer fit the profile of the man I met along the road, or the businessman from Luang Namtha: urban people with means who might own farms but were not typically identified as “farmers.” And even the so-called poorer farmers who participated in “3 + 2”—people who could not afford to finance inputs on their own—were still relatively well-off compared to their even poorer neighbors. Both groups tended to have multiple existing livelihood options. For the northern uplands’ majority, on the other hand, smallholder rubber, even with contract farming, was a leap too far.

Wage work, on the other hand, was not. This was where the Sino-Lao counteroffer examined above carried a logic that traveled widely: that of combining wage work with a division of plantation land rather than a sharecropping arrangement based on the division of the rubber (latex) crop. Between 2005 and 2007, a variant on “3 + 2”—sometimes called “4 + 1”—was developed in various pockets around the northwest. In some places its genealogy was obvious. Even though Sino-Lao’s proposal was ultimately rejected in Luang Namtha,⁹⁴ the company took the same plan to Oudomxai province, where it applied the wage-work-with-land-division model to a 5,000-hectare “3 + 2” agreement it negotiated with local authorities. As in Luang Namtha, provincial Agriculture and Forestry officials had objected to the scheme, but the company made an arrangement with the provincial governor’s office and used district-level technical staff to circumvent the objection.⁹⁵ Back in Luang Namtha, the same basic template of wage labor plus land partition was applied by other companies, including Bolisat Ltd., whose “4 + 1” arrangement in Khet Nam Fa bettered Sino-Lao’s Oudomxai deal by a full 20 percent: Bolisat Ltd. was promised a 70–30 land split in their favor, and in the end no partition took place at all (see ch. 4). Other variations on “4 + 1” occurred across the northwest (map 1.2).⁹⁶ Targeting the “lower uplands” under 800 meters where the risk of frost was low, these deals homed in on the region’s abundant supply of former shifting-cultivation lands (pa lao), frequently accessing high-quality land close to roadsides.

In 2008, as I was finishing my main research period, it was already clear that this concession-like variant had become “the predominant contracting mode,” as political scientist Weiyi Shi noted in the first major study of northern Laos’s rubber boom.⁹⁷ At the time there was already much debate about this “other” sort of contract farming, both in and out of government, since it used the heavy hand of “state land management” to secure company plantation lands. This meant using the basic tools of concession-making to create available land, either by invoking state claims of direct ownership over forests or invoking the state’s exclusive right to “manage” (khoumkhong) or plan agricultural land use through the zoning (chatsan) process. Moreover, the very thing that made these deals more compatible with upland livelihoods—their use of wage labor—also made them more concession-like. Echoing Shi’s worry that “it is not enough to ban the concession only to have its problems disguised under a new face called ‘contract farming,’ ” a provincial official complained to me in mid-2007: “There is a problem with contract farming. [It’s] not really like contract farming—it’s like a concession to a company: a big area. The company says it’s contract farming but it’s not actually contract farming.”⁹⁸

These quasi-concessions were not properly inventoried at the time, in part because national-level land management authorities deemed them outside the purview of their already limited concession-inventorying efforts.⁹⁹ Even a decade later, provincial-level technical staff in Luang Namtha and Oudomxai admitted that they were still unsure how much land had ended up in each “contract farming” category.¹⁰⁰ In 2013, Luang Namtha’s Agriculture and Forestry office published a “concept note” on rubber production in the province that is nonetheless illuminating in its nonchalant reference to rubber as needing huge amounts of alienated wage labor rather than relying on the household labor of smallholders. While acknowledging the crop’s origins in state efforts “to generate income and alleviate poverty, stop slash-and-burn and poppy cultivation, and create more permanent jobs for people of the province,” the document makes clear the shift since those hopeful early days.¹⁰¹ Calculating a labor shortfall of forty-nine thousand people between the plantation owners and the working bodies necessary for tapping their rapidly maturing holdings, the “concept note” reflects the slippage from the type of smallholder conjured by “permanent livelihoods” rhetoric—the stabilized shifting cultivator turned household-scale rubber grower—to the larger plantation owners described above: Chinese companies and various levels of Lao elites.¹⁰² And as we have already seen in Khet Nam Fa, even this “opportunity” for local people to “get some income by selling their labor power” did not always pan out.

While scholars have rightly pointed out the extensive conjuring of empty land that often underlay Laos’s representation as “a business-friendly resource frontier,”¹⁰³ this chapter tells a subtly different story. Some Chinese rubber companies did indeed buy into the myth of empty land, reflecting Laos’s frequent portrayal in China as underpopulated and resource-rich.¹⁰⁴ But a better reflection of investment politics appears in the materials examined above, drawn not from the public transcript of boostering and the media but from the more guarded arena of regulatory debates and specific investment projects.¹⁰⁵ Contra land being empty or unused, this material shows that land was instead what we might call “socially” available: accessible to certain preferred uses like rubber even if it was already being used in other ways, and even if the channels for availability—Chinese rubber “promotion” projects, Lao elites’ plantation schemes—remained themselves subject to ongoing deliberation, debate, and state intervention. Land’s social availability also appears in the statistical conjuring exercise quoted above, where Lao officials asserted that developing 30,000 hectares of new rubber plantations would “not be a problem.” Ministerial officials may not have known precisely where those 30,000 hectares would go, but that was beside the point. Their proposal was about recombining land and labor—the “hectares” in the three northwestern provinces and the local population’s “labor units”—in new ways, precisely because their existing configuration was seen as undesirable.

This recombination could have gone a number of ways, in part because of the possibilities created by the Chinese central government’s “opium poppy replacement” program. Established in late 2004, the program used a mix of tax credits, import allowances, and direct cash subsidies to support Chinese companies in developing rubber and other agricultural-commodity “promotion” schemes across northern Laos and northern Myanmar.¹⁰⁶ On the one hand, given Beijing’s apparent seriousness about using alternative-livelihoods development to help stem the flow of Southeast Asian opium into China’s heroin market, the program might have used its significant resources to help finance the smallholder rubber transition, much like state enterprises did decades earlier in Malaysia and Thailand.¹⁰⁷ This would have been consistent with the program’s purpose of facilitating Chinese investment by lowering the costs of working with upland farmers in new and economically risky contexts.¹⁰⁸ Moreover, it would have followed Lao officials’ request for smallholder-centric rubber cooperation, mirroring both China’s long-standing commitment to “noninterventionist” international cooperation and more recent rhetoric, such as that accompanying the BRI, about using state resources to do international development better than under the prevailing neoliberal model.¹⁰⁹

On the other hand, the devolution of the program’s oversight to the provincial level—and in particular, the decision to give Yunnan’s provincial commerce department administrative power over the program’s operations—articulated with a narrower pursuit of business interests.¹¹⁰ The poppy-replacement program thus exemplified Yunnan authorities’ wider strategy of tapping into China’s national policy aims and associated revenue streams by becoming “a grand passageway to Southeast Asia” via official channels like the “Going Out” policy and the GMS economic corridors.¹¹¹ To the extent that it more narrowly supported the interests of the companies involved, the program risked being simply financial fuel for an upland land rush, wrapped in the legitimating guise of alternative development.

The evidence points toward the latter, but also highlights the persistence of significant opacity when it comes to what happened where, why, and how. This illegibility obscures crucial details. Of all of the places where rubber cooperation was “promoted” by Chinese companies and local Lao authorities, only some communities lost land. The socially uneven distribution of enclosure and dispossession has been noted throughout Laos’s concession boom, as well as more widely throughout the global land rush.¹¹² In Laos, it jibes with a long-standing recognition that land policy is subject to differing interpretations by local authorities, often due to how it connects with various “local interests and power struggles.”¹¹³ While this can have positive implications in some cases, it also has clearly negative ones in circumstances like those discussed here. In contexts like the NEC’s upland villages, where property formalization was still in flux, and where the “promotion” of rubber was often paired with a strong dose of coercion to accept any development assistance on offer, negotiating with “investors” and government officials demands both political acumen and active citizenship. The lack of spatial detail about where and how Chinese rubber schemes operated leaves unanswered the questions of how land’s social availability was negotiated on the ground and how this differed from place to place.

We need to look carefully at the historical terrain on which citizenship is created and negotiated if we are to make sense of the uneven enclosures exemplified by Khet Nam Fa and the slippage to “4 + 1” more broadly. While mainland Southeast Asia’s economic integration has been famously called “turning battlefields into marketplaces,” this phrase is usually interpreted both regionally and metaphorically by critics and proponents alike.¹¹⁴ Instead, we need to take this phrase more literally by examining Southeast Asia not just as a Cold War landscape in a general sense but also as an interlinked network of actual, local landscapes where Cold War conflict took place on the ground, and where place-specific struggles over land access continue today. It is in these groundings that we see how the slippages from protective efforts like “3 + 2,” a regulatory pushback against large-scale land concessions to Chinese rubber companies, came about, and how the quasi-concessions of “4 + 1” became targeted into particular communities and landscapes. Examining the “global” land rush in the landscapes and communities of northwestern Laos thus illustrates not just the mechanics of land grabbing but also shows the ways in which legacies of geopolitical conflict can linger on the ground, animating the micropolitics of land access far longer than they have any right to.