Shifting cultivation stability and change: Contrasting pathways of land use and livelihood change in Laos

Applied Geography 46 (2014) 1e10

Contents lists available at ScienceDirect

Applied Geography journal homepage: www.elsevier.com/locate/apgeog

Shifting cultivation stability and change: Contrasting pathways of land use and livelihood change in Laos Thoumthone Vongvisouk a, b, *, Ole Mertz a, Sithong Thongmanivong b, Andreas Heinimann c, d, Khamla Phanvilay b a Geography Section, Department of Geosciences and Natural Resource Management, University of Copenhagen, Øster Voldgade 10, 1350 Copenhagen K, Denmark b Faculty of Forestry, National University of Laos, Vientiane, Laos c Centre for Development and Environment (CDE), University of Bern, Hallerstrasse 10, 3012 Bern, Switzerland d NCCR North-South, Ministry of Environment and Natural Resources, Prime Minister’s Office Compound, P. O. Box 1438, Vientiane, Laos

a b s t r a c t Keywords: Cash crops Land use change Land policies Lao PDR Livelihood Swidden

Rural areas in Laos are experiencing a rapid transformation from traditional rice-based shifting cultivation systems to more permanent and diversified market-oriented cultivation systems. The consequences of these changes for local livelihoods are not well known. This study analyzes the impact of shifting cultivation change on the livelihood of rural people in six villages in three districts of northern and central Laos. Focus group discussions and household interview questionnaires were employed for data collection. The study reveals that the shifting cultivation of rice is still important in these communities, but it is being intensified as cash crops are introduced. Changes in shifting cultivation during the past ten years vary greatly between the communities studied. In the northern study sites, it is decreasing in areas with rubber expansion and increasing in areas with maize expansion, while it is stable in the central site, where sugarcane is an important cash crop. The impacts of land use change on livelihoods are also diverse. Cash crop producers hold more agricultural land than non-cash crop producers, and rubber and sugarcane producers have fewer rice shortages than non-producers. In the future, livelihood improvements in the central study site may be replicated in the northern sites, but this depends to a large extent on the economic and agricultural settings into which cash crops and other development opportunities are introduced. Moreover, the expansion of cash crops appears to counteract Lao policies aimed at replacing shifting cultivation areas with forests. Ó 2013 Elsevier Ltd. All rights reserved.

Introduction

recognition that it represents a form of forest degradation, as a shifting cultivation landscape is composed of areas of fallow vegetation with different ages and compositions (Herold & Skutsch, 2011; Mertz et al., 2012). Accordingly, there has been pressure to conserve forests for Reducing Emissions from Deforestation and forest Degradation (REDDþ). In addition, other factors such as land policy reforms, the increasing commoditization of land, and cash crop introductions are also causing the gradual transformation and/ or complete elimination of shifting cultivation systems, which have both positive and negative implications for the environment and local livelihoods (van Vliet, Mertz, Birch-Thomsen, & Schmook, 2013). This transformation is occurring particularly rapidly in Southeast Asia (Fox et al., 2009; Padoch et al., 2007), and a global review of case studies has shown that Southeast Asia has seen the fastest decline in shifting cultivation during the last 10e15 years (van Vliet et al., 2012). However, it is also recognized that strong data are still limited, both in terms of the areas occupied by shifting

Shifting cultivation remains an important agricultural system in many developing countries, where it dominates in some areas and co-exists with more permanent agricultural systems in others (van Vliet et al., 2012). While shifting cultivation has been on many research and political agendas for decades, it has recently gained renewed interest because of the global interest in reducing greenhouse gas emissions and ensuring that carbon stocks in tropical forests are not further reduced due to deforestation and forest degradation. In this context, shifting cultivation has mainly been seen as an agent of deforestation, although there is increasing * Corresponding author. E-mail addresses: [email protected], [email protected]. dk (T. Vongvisouk), [email protected] (O. Mertz), [email protected] (S. Thongmanivong), [email protected] (A. Heinimann), [email protected] (K. Phanvilay). 0143-6228/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apgeog.2013.10.006

2

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

cultivation and the number of people who are mainly dependent on it as a part of their livelihood (Mertz et al., 2009; Schmidt-Vogt et al., 2009). One country that has been studied rather intensively is Laos. A first land coverage and population data-based attempt at mapping shifting cultivation for an entire country was done in Laos, demonstrating that this agricultural system is still widespread in many parts of the country (Messerli, Heinimann, & Epprecht, 2009). Although earlier land coverage data were generated using different methodologies and are therefore difficult to compare to the study by Messerli et al. (2009), it does appear that shifting cultivation increased from approximately 2e2.5 million hectares in the 1990s (Cohen, 2009; Hansen, 1998; Kallabinski & Lundgreen, 2005; Linquist et al., 2005; World Bank, 2008) to approximately 6.5 million hectares, which is equal to 28.2% of the total country area, in the mid-2000s (Messerli et al., 2009). However, another study showed that the shifting cultivation land area in northern Laos declined during the 2000s (Hurni, Hett, Heinimann, Messerli, & Wiesmann, 2013). More than 550,000 people live in shifting cultivation areas in northern Laos, and these areas have a poverty rate of 46.5%, which is much higher than in the rest of the country (Heinimann et al., 2013). Shifting cultivation change in Laos results from multiple drivers, as seen in other parts of the world. The Government of Laos (GoL) considers traditional shifting cultivation as one of the causes of deforestation and forest degradation in Laos. Hence, a reduction in shifting cultivation and the promotion of permanent agriculture are aims of agricultural development policies in Laos (Castella et al., 2013; Lestrelin & Giordano, 2007). One of these policies is the Land and Forest Land Allocation (LFA), which aims to halt shifting cultivation in upland areas and promotes the involvement of local people in permanent agricultural production. Moreover, increased trade and market integration with neighboring countries (mainly China, Vietnam, and Thailand) and the partial commoditization of land has spurred many local people to shift from subsistencebased to market-based production. Commodity crops are typically watermelon, maize, sugarcane, rubber, and vegetables (Thongmanivong & Fujita, 2006; World Bank, 2008). Changes from subsistence-based to market-based production have not only had an impact on land use change, but they have also had an impact on the livelihoods of local people. Case studies from northern Laos show that local people, who have capital investments in their land, gain advantages from the change, while poor people with limited financial capital investments in their land are unable to benefit from the change. Therefore, poor people are likely to either become laborers for cash crop farmers and rubber planters (Thongmanivong & Vongvisouk, 2006) or get involved in off-farm activities by necessity, rather than by choice (Bouahom, Douangsavanh, & Rigg, 2004). Generally, a reduction in social equity in local communities occurs when land use policies such as the LFA are implemented and a market economy is introduced into rural areas. For example, in the Luang Namtha Province, wealthy Akha households have better access to agricultural land in the areas close to roads, where they have established rubber plantations. However, immigrants, who were moved from more remote areas because of the eradication of opium and to be located closer to roads and the district center, became dependent on working in the rubber plantations owned by the wealthy households (Cohen, 2009). In this article, and in line with the recommendations of Heinimann et al. (2013), we aim to gain a better understanding of the causal relationships between land use and livelihood change during the 2000s in northern and central Laos. While Heinimann et al. (2013) combined statistical information with aggregated village census data and remote sensing approaches for spatial analysis, we use household data to understand why poverty

remains a crucial issue in the region, despite government efforts to alleviate poverty through land allocation policies. Our specific aim is to look at how a change (or no change) in the practice of shifting cultivation has affected local people’s livelihoods and whether the new land uses have produced the intended positive outcomes. We also assess the implications of these land use changes for potential future REDDþ projects. Study site, definition of terms and methods The study was carried out in six villages in northern and central Laos: Namha and Sopsim in the Namtha District of the Luang Namtha Province; Keo Xik and Houay Moon in the Hua Meuang District of the Huaphan Province; and Phonthong and Taotharn in the Hin Heub District of the Vientiane Province (Fig. 1). Although the villages of Phonthong and Taotharn share one administration (one village headman), they are still distinguished as two villages. The population in the four villages in the Namtha and Hua Meuang Districts is mainly comprised of ethnic Khmu individuals, but the villages differ in facilities such as road access (Table 1). In both districts, two of the selected villages are located along the main road (one in each district), while the others are located along local roads with more difficult access, especially in the rainy season. The two districts have different conditions for investment. The Namtha District is the main district of the Luang Namtha Province and is located along highway R3 connecting Xishuangbanna (China) and Chiang Rai (Thailand), which has become an important Southeast Asian trade route with considerable economic activity (Fox, Vogler, Sen, Giambelluca, & Ziegler, 2012). The Hua Meuang District is located close to the LaoeVietnamese border and thus is also on an international trade route with considerable economic activity; however, roads are considerably smaller and travel is slow. Finally, the Phonthong and Taotharn villages in the Hin Heub District were selected to include a district where shifting cultivation was likely to be transformed due to the proximity of Vientiane Capital. The characteristics of the studied villages are summarized in Table 1. All villages practice a combination of traditional shifting cultivation of upland rice and vegetable and cash crop cultivation (e.g., rubber plantations and hybrid maize). In this study, shifting cultivation refers to the cultivation of swiddens with upland rice and other crops followed by a number of years of fallow. Some cash crops, such as hybrid maize, are also grown in a shifting cultivation type of system with shorter fallow than upland rice, but farmers consider this a very distinct farming practice and do not label it shifting cultivation. Moreover, when we use the term ‘shifting cultivation’, it covers the entire system of swidden fields and fallow vegetation; when we refer to ‘upland rice fields’, the term excludes fallow areas. Data were collected between July and September 2010 from all study villages, and two main types of information were gathered: focus group discussions, including a mapping of the village land and individual household interviews using a structured questionnaire with a number of open questions to allow for individual opinions. The focus group discussions included representatives of various village organizations after collecting basic information on each village, such as main land uses, land use changes, demographical information, and livelihood conditions. A total of 222 households participated in the questionnaire survey, and the exact number per village is listed in Table 1. We randomly selected households for interviews based on a list of households provided by the village during the focus group discussions. We interviewed representatives of the selected households, mainly the head of household. The questionnaire contained quantitative questions relating to current land use and livelihoods followed by perception-based questions on changes in land use,

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

3

Fig. 1. Study sites in Laos.

livelihoods, household incomes (ranked by important sources), and the causes of these changes during the previous ten years. For instance, questions focused on recent household income by source, income change compared to ten years ago, and the factors affecting changes in household income. We subsequently coded the answers into aggregated categories that were used in the analysis. The acquired data on land use and livelihood change were complemented by comparison with previous studies carried out in the three regions when such studies were available and could provide data suitable for comparison.

Results Current land use Shifting cultivation is still relatively important in all districts. According to household interviews, communities in the Namtha District reported upland rice fields accounting for 30% of the total agricultural area. The equivalent figure in the Hua Meuang District is 40%. In both districts, the communities had more than two ha of upland rice fields per household. In the studied villages in the Hin

Table 1 Summary of characteristics of the studied villages. District

Village

Resettlement Distance from town

Distance from main road

Total population Main land uses Total no. of No. of in village households surveyed households in village

Namtha

Namha

1976

Along the main road

132

36

614

Sopsim

1984

12 Km from main road

72

30

352

Hua Keo Xik Meuang

1999

5 Km from main road

29

24

191

Swidden, paddy, rubber, cardamom Swidden, paddy, rubber, jatropha Swidden, paddy, hybrid maize

Along the main road

70

52

393

Swidden, paddy, hybrid maize

Along the main road

36

45

196

Swidden, paddy, rubber, jatropha, teak, acacia, sugarcane

Along the main road

97

35

612

Swidden, paddy, rubber, jatropha, teak, acacia

436

222

2358

Houay Moon 1984

Hin Heub

Total

Phonthong

1981

Taotharn

1981

29 km from Namtha District center 17 km from Namtha District center 58 km from Hua Meuang District center 52 km from Hua Meuang District center 22 km from Hin Heub District center 21 km from Hin Heub District center

4

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

Fig. 2. Percentages of total agricultural land holdings of the studied communities in Namtha (Grazing land: n ¼ 4; Upland rice field: n ¼ 55; Paddy land: n ¼ 27; Fallow <5 yrs: n ¼ 20; Fallow 5e10 yrs: n ¼ 9; Plantation: n ¼ 58; Cash crop land: n ¼ 3; Others: n ¼ 26), Hua Meuang (Grazing land: n ¼ 24; Upland rice field: n ¼ 75; Paddy land: n ¼ 5; Fallow <5 yrs: n ¼ 3; Fallow 5e10 yrs: n ¼ 3; Plantation: n ¼ 4; Cash crop land: n ¼ 71; Others: n ¼ 4) and Hin Heub (Grazing land: n ¼ 19; Upland rice field: n ¼ 51; Paddy land: n ¼ 57; Fallow <5 yrs: n ¼ 16; Fallow 5e10 yrs: n ¼ 3; Plantation: n ¼ 34; Cash crop land: n ¼ 33; Others: n ¼ 19). “n” is the number of respondents mentioning each land use type.

Heub District, 26% of their agricultural land is occupied by upland rice fields, but there is only approximately 1 ha per household (Figs. 2 and 3). This is also reflected in the amount of fallow land available to the households, which is lowest in the communities in the Hin Heub District. The communities in the Hua Meuang District have the largest cash-crop cultivation area compared to the communities in the other districts. Hybrid maize production is of specific importance in Hua Meuang, and Lao traders organize the export of hybrid maize to markets in Vietnam. Plantation agriculture is less important in Hua Meuang than in the communities in Namtha and Hin Heub, with rubber in particular having gained importance in recent years in the Namtha District due to the Chinese market. Change in land use Data on land use change based on household interviews are shown in Table 2. More than half of the respondents in the Namtha communities reported that their upland rice field areas had decreased, which was also confirmed in the focus group discussion. However, in the Hua Meuang communities, almost two-thirds of the respondents reported increasing or stable upland rice field areas. More than half of the respondents in Hin Heub also reported an increase or stability in areas of upland rice fields. The availability

of fallow land is mostly considered stable by households in all communities, but relatively few households responded to the question concerning fallow change, especially regarding older fallow forest categories. The decrease in upland rice fields in Namtha has clearly been brought about by an increase in plantations (mostly rubber). Twothirds of respondents reported plantations to be stable, and it appears that rubber cultivation is also reaching a level of stability in Namtha. This is due to an earlier ‘boom’, as reported from the Sing District, where farmers located in areas close to roads converted their shifting cultivation areas to rubber plantations (Thongmanivong & Vongvisouk, 2006). The conversion from shifting cultivation areas to rubber plantations is driven by the government policy that aims to reduce shifting cultivation (Alton, Bluhm, & Sannikone, 2005; Shi, 2008) and increase private investment. Moreover, local people observe that other villages in the same district are successful with rubber plantations. It is mainly households with larger land areas that convert swidden to rubber. Households with less land tend to maintain their swidden fields with upland rice, which is done to maintain some level of self-sufficiency in rice and is a result of the limited areas available for paddy (wet) rice cultivation in the studied villages in Namtha (see Fig. 2). The increase in upland rice fields in the Hua Meuang communities has been accompanied by an increase in hybrid maize

Fig. 3. Average agricultural land holdings per household in the studied communities in Namtha (Grazing land: n ¼ 4; Upland rice field: n ¼ 55; Paddy land: n ¼ 27; Fallow <5 yrs: n ¼ 20; Fallow 5e10 yrs: n ¼ 9; Plantation: n ¼ 58; Cash crop land: n ¼ 3; Others: n ¼ 26), Hua Meuang (Grazing land: n ¼ 24; Upland rice field: n ¼ 75; Paddy land: n ¼ 5; Fallow <5 yrs: n ¼ 3; Fallow 5e10 yrs: n ¼ 3; Plantation: n ¼ 4; Cash crop land: n ¼ 71; Others: n ¼ 4) and Hin Heub (Grazing land: n ¼ 19; Upland rice field: n ¼ 51; Paddy land: n ¼ 57; Fallow <5 yrs: n ¼ 16; Fallow 5e10 yrs: n ¼ 3; Plantation: n ¼ 34; Cash crop land: n ¼ 33; Others: n ¼ 19). “n” is the number of respondents mentioning each land use type.

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

5

Table 2 Changes in land use during the past ten years in the six studied villages as perceived by households. For each type of land use, the listed values indicate the percentage of households that reported one of the four change (or no change) categories. Percentages are based on the total number of answers in each category and are therefore not percentages of the total number of households surveyed in each district. Land use change

Grazing land

Upland rice fields

Namha and Sopsim, Namtha district Increase (%) 50 11 Decrease (%) 25 58 Stable (%) 25 22 Variable (%) e 9 No. of respondents 4 55 Keo Xik and Houay Moon, Hua Meuang District Increase (%) 29 44 Decrease (%) e 16 Stable (%) 63 28 Variable (%) 8 12 No. of respondents 24 75 Phonthong and Taotharn, Hin Heub District Increase (%) 21 18 Decrease (%) e 39 Stable (%) 79 39 Variable (%) e 4 No. of respondents 19 51

Paddy land

Fallow<5 yrs

Fallow 5e10 yrs

Plantation (rubber, etc.)

Cash crop land (maize, etc.)

Others

33 22 41 4 27

15 30 55 e 20

e 44 56 e 9

24 9 67 0 58

e e 100 e 3

27 12 62 e 26

20 e 60 20 5

e 33 67 e 3

e e 100 e 3

e 25 75 e 4

56 9 35 e 68

e e 100 e 4

35 e 65 e 57

e e 100 e 16

33 e 67 e 3

6 e 94 e 34

12 e 88 e 33

e e 100 e 19

Note: (e) means no respondents reported (or equal to “0”).

cultivation close to the road for easy marketing, while the areas far away from the road are kept for rice cultivation. Although only a few households responded to the change in fallow land, those who did reported a stable availability of fallow land. This finding could indicate a reversal of the trend observed in a previous study of the Hua Meuang District (Pieng Xai and Keo Xik villages), in which 61% of interviewees reported that their fallow length had decreased compared to two decades earlier (Seidenberg, Mertz, & Kias, 2003). The stability or increase of upland rice fields and short fallow land in the Hin Heub communities has occurred simultaneously with the consolidation of paddy rice, plantation and cash crop areas. Current livelihoods As indicators of livelihood, we mainly use a combination of household income, expenditures, rice shortages, and, to some extent, measures of equity. While we recognize that there are other important aspects of livelihood and welfare, we find these to be the main indicators of livelihood change in poor rural households that, until recently, were mainly subsistence-based production units but now are seeing a commodification of their land use systems. Households in the Hin Heub District have the highest income, as they earn an annual income of LAK 8.6 million from on-farm

activities and LAK 6.1 million from off-farm activities per household (Fig. 4) (Exchange rate June 2012: USD 1 ¼ LAK 8000). This is mainly due to the proximity to the Vientiane Capital and hence better access to markets and jobs, e.g., construction. Income from on-farm activities in the Hua Meuang District is also relatively high compared to that of farmers in the Namtha District mainly due to the boom in hybrid maize cultivation and marketing in the Huaphan Province in the late 2000s. Non-timber forest products (NTFP) are an important source of income for local people in Namtha and Hua Meuang. NTFPs are very limited in Hin Heub (Fig. 4), where the diversification of household income, more intensive and diverse farming practices, and less access to forests limit dependence on NTFPs. In Namtha, households with rubber are still waiting for income from rubber plantations, as they began to establish plantations only four years prior to the survey. Generally, rubber trees grown in northern Laos will provide latex when they reach eight years of age, and they will provide an annual net income when they reach ten years of age (Manivong & Cramb, 2008). However, according to villagers in the Namha village, they can tap for rubber latex when the tree reaches six years if they manage the plantation well. The average annual expenditure per household surveyed in the three districts is LAK 4.0 million in Namtha, LAK 2.7 million

Fig. 4. Average household income by source in the studied communities in Namtha (n ¼ 66), Hua Meuang (n ¼ 76), and Hin Heub (n ¼ 80).

6

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

in Hua Meuang, and LAK 8.2 million in Hin Heub. The higher expenditure in Hin Heub can be seen to correspond to its higher income and a lower reliance on subsistence production. The distribution of expenditures, however, demonstrates the different development stages of the surveyed communities (Fig. 5). In Hin Heub, most is spent on consumable goods, healthcare, clothes, transportation and education, whereas in Namtha, a comparatively high proportion is spent on agricultural inputs related to rubber cultivation. Expenditure for rice is highest in the two districts in the north and corresponds to a lower level of agricultural intensification in upland rice production as well as the recent focus on rubber and maize on land previously reserved for upland rice cultivation. Change in livelihoods The perceptions of households in the studied communities indicate that 50% of the respondents in the Hua Meuang District and only 11 and 19% of the respondents in the Namtha and Hin Heub Districts, respectively, perceived that they have better livelihoods (or reduced poverty) compared to ten years ago. One of the indicators that households used for well-being and acceptable livelihoods is the ability to produce sufficient rice for their own consumption, which can also be referred to as rice sufficiency. This is important to most households as a security measure even if they have sufficient income to buy rice. A total of 56% of households in Hin Heub declared that they are recently experiencing increased rice sufficiency compared to ten years ago. On the other hand, 27% of interviewees in the Hua Meuang District perceived that they produce an insufficient amount of rice for their own consumption, and 12% of the respondents indicated that they work harder compared to ten years ago but are still unable to produce sufficient amounts of rice for their household needs. In the Namtha District, even if they work harder (27% of the respondents), 30% of the respondents perceived that they face more frequent rice insufficiency now compared to ten years ago (Fig. 6). Household incomes have declined during the past ten years by 62% in Namtha and 46% in the Hua Meuang Districts. Conversely, in the Hin Heub District, 85% of respondents reported that their household income had either increased, remained stable, or varied (Fig. 7). In all communities, households reported an increase in annual expenditure compared to ten years ago by at least 66%.

Impacts of land use change on livelihoods Determining a direct causality between land use change and livelihood indicators is complex, as livelihoods are influenced by many different factors. However, in the two northern districts, where the livelihoods of the studied communities rely primarily on agriculture, we can assume that land use change also affects income levels, expenditure and assets. In the Hin Heub District, off-farm income sources are gaining importance and to what extent land use change has affected livelihoods is less clear. In each village and in the combined data set, no significant relationship was found between household income levels and the amount of area planted with rubber (regression r2 ¼ 0.05); furthermore, there was no relationship observed between income levels and amount of upland rice area. As mentioned above, this could be because rubber is not yet generating e and thus not reflected in e household income. On the other hand, it could also be hypothesized that households with higher income would invest more, thereby increasing inequity in the communities; however, this does not seem to be the case. When comparing total land holdings and rice shortages with the income levels of different groups, it appears that rubber planters in Namtha have larger land holdings and fewer rice shortages than non-rubber planters (Fig. 8). This is counter-intuitive, as it would be expected that rubber planters produce less rice; however, land holdings seem to be sufficiently large for rubber planters to maintain high rice production levels. In Hua Meuang, hybrid maize producers also have larger land holdings, but they have more frequent rice shortages, most likely because maize cultivation competes directly with rice cultivation in shifting cultivation fields and labor concentration. Additionally, it should be noted that only three households are in the non-maize producer category. Finally, sugarcane producers in Hin Heub have larger land holdings and fewer rice shortages than non-sugarcane producers. This lends to the hypothesis that land investments are indeed made mainly by the wealthiest in the community, at least in terms of land holdings and access to rice. To further disaggregate how changing the shifting cultivation of upland rice to other forms of agriculture affected livelihoods, we analyzed the income levels of different groups of households that had e or had not e engaged in rubber, maize or sugarcane cultivation and which provided information about changes (or no change) in shifting cultivation (Fig. 9). In Namtha, it appears that rubber planters who also decreased their shifting cultivation

Fig. 5. Average annual household expenditure in the studied communities in Namtha (n ¼ 66), Hua Meuang (n ¼ 76), and Hin Heub (n ¼ 80).

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

7

Fig. 6. Perceptions of livelihood change during the past ten years in the studied communities in Namtha (More poverty: n ¼ 8; Reduce Poverty: n ¼ 19; Rice deficiency: n ¼ 52; Rice sufficiency: n ¼ 13; Work harder: n ¼ 46; land scarcity: n ¼ 16), Hua Meuang (More poverty: n ¼ 4; Reduce Poverty: n ¼ 54; Rice deficiency: n ¼ 29; Rice sufficiency: n ¼ 6; Work harder: n ¼ 13; land scarcity: n ¼ 0), and Hin Heub (More poverty: n ¼ 1; Reduce Poverty: n ¼ 9; Rice deficiency: n ¼ 2; Rice sufficiency: n ¼ 27; Work harder: n ¼ 1; land scarcity: n ¼ 6). “n” is number of respondents answering each category.

(Rubber-SC down) have higher incomes than rubber planters who increased or kept shifting cultivation stable (rubber-SC stable_up). This is because rubber planters, who practice limited shifting cultivation, have a larger portfolio of income-generating activities, which is most likely because they spend less labor on shifting cultivation. Households with rubber but no shifting cultivation (Rubber-no SC) and those without either of the two (No rubber-no SC), appear to have higher incomes than all of the rest. Both categories, however, contain relatively few households, and the average income levels are driven up by one household in each category that earns a high-income working as a driver or raising buffaloes. Removing those two households brings the income levels in those groups down to levels that are comparable to other categories. The three households in the Hua Meuang District that do not have maize cultivation are not shown in Fig. 9. There is no difference in income levels between the households that cultivate maize and decrease the shifting cultivation area (Maize-SC down) and households that cultivate maize and increase or keep their shifting cultivation areas stable (Maize-SC stable_up). Finally, households in the Hin Heub District that planted sugarcane and increased or kept shifting cultivation stable (Sugarcane-SC stable_up) have the

highest incomes. Several households in this category have a high agricultural income and mentioned that they intensified upland rice cultivation as well as other agricultural practices. Households that produce sugarcane without shifting cultivation (Sugarcane-no SC) also had high incomes, which in several cases can be attributed to off-farm labor. Overall, the results from the three districts indicate a very mixed picture with respect to the effects of land use change on livelihood. In the Namtha District, the expansion of rubber plantations has not yet contributed to household income, as the plantations are taking up land and labor that were previously used for rice production but are not yet producing; this explains the increase in rice insufficiency and decline in income. However, rubber planters seem to be doing better than other households in terms of income, rice sufficiency, and land holdings. In the Hua Meuang District, only 18% of respondents reported increased income, even though hybrid maize boomed in the late 2000s. This finding signifies that maize may not be providing the expected net income to farmers either because maize prices are too low, input costs too high, or both. However, as all households follow more or less the same strategies, there seems to be little difference in terms of equity. As the agricultural inputs obtained for the maize cultivation are governed under a contract

Fig. 7. Changes of household income during the past ten years in the studied communities in Namtha (n ¼ 66), Hua Meuang (n ¼ 76), and Hin Heub (n ¼ 80).

8

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

Fig. 8. Average land use areas and periods of rice shortages by different producer groups in the studied communities in Namtha (Rubber planter: n ¼ 41; Non-rubber planter: n ¼ 25), Hua Meuang (Maize producer: n ¼ 73; Non-maize producer: n ¼ 3), and Hin Heub (Sugarcane producer: n ¼ 37; No-sugarcane producer: n ¼ 43).

farming system, there is, so far, little flexibility in using these for other agricultural activities. In Hin Heub, entrepreneurial households are benefitting whether they continue to employ shifting cultivation or not. Here, the opportunities for purchasing agricultural inputs, such as seedlings and fertilizers, make the intensification of short fallow shifting cultivation more feasible than in other districts. Combining seasonal off-farm and agricultural labor makes some households considerably wealthier than others. Consequently, inequity in the Hin Heub communities is likely to be higher, but as the income levels of even the poorest households are higher than in the other communities, development here seems to be generally beneficial. Discussion and conclusion The changes occurring in the northern parts of Laos are part of an overall trend in land use change and a decline in shifting cultivation in Southeast Asia (Mertz et al., 2009; Padoch et al., 2007). Southern Yunnan saw some of the earliest government- and private investor-driven conversions of shifting cultivation to rubber plantations, and the rubber area tripled during the early 1990s (Li, Ma, Aide, & Liu, 2008). Thailand and Vietnam have also experienced a rapid decline in shifting cultivation areas, which is due to the existence of government policies that reduce shifting cultivation and incentivize the production of cash crops for more sedentary agriculture (Fox et al., 2009; Padoch et al., 2007). In insular Southeast Asia, the oil palm expansion in Malaysia and Indonesia is also causing a rapid decline of shifting cultivation in many areas (Mertz, Egay, Bruun, & Colding, 2013).

The impacts of these changes on livelihoods are generally not well understood. Although the transformation from shifting cultivation to cash crops can help improve household livelihoods, it may also lead to social inequity, as the transformation leaves some groups marginalized and worse off (Cramb et al., 2009). This effect was confirmed in Sarawak, Malaysia, where income and asset levels increased sharply as former shifting cultivators took up smallholder oil palm cultivation only for those who had the necessary land and capital to invest. Households with fewer assets were struggling to join the oil palm adventure, though it was clearly the path that most households saw as being important in the future (Mertz et al., 2013). In parts of Yunnan, income levels have also increased significantly as shifting cultivation has been abandoned and rubber has taken over (Fox & Castella, 2013; Sturgeon, 2010). In many of these locations, a new class of poor people is emerging, as immigrants are working in the plantations owned by the households who formerly focused on shifting cultivation (Cramb & Curry, 2012). The results of the present study in Laos confirm some of these trends but also show that the successful replacement of shifting cultivation with commercial crops is highly dependent on the economic and agricultural settings into which they are introduced. Rubber cultivation has yet to yield income improvements for the communities in Namtha District, but these may become substantial once the plantations are established and producing. Success will depend on the ability of communities to maintain their rights to the land and obtain fair market prices for their rubber. Otherwise, the lack of access to land and labor for cultivating upland rice and other crops in shifting cultivation, as well as for collecting forest products,

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

9

Fig. 9. Average income of households that have different combinations of cash-crops and changes in shifting cultivation (SC) in the studied communities in Namtha (Rubber-SC down: n ¼ 19; Rubber-SC stable_up: n ¼ 14; Rubber-No SC: n ¼ 6; No rubber-SC down: n ¼ 13; No rubber-SC stable_up: n ¼ 6; No rubber-No SC: n ¼ 6), Hua Meuang (Maize-SC down: n ¼ 11; Maize-SC stable_up: n ¼ 60), and Hin Heub (Sugarcane-SC down: n ¼ 13; Sugarcane-SC stable_up: n ¼ 11; Sugarcane-No SC: n ¼ 13; No sugarcane-SC down: n ¼ 8; No sugarcane-SC stable_up: n ¼ 16; No sugarcane-no SC: n ¼ 16). In Hua Meuang all but three households cultivated maize, hence the “no maize” combinations were excluded. SC down means that households reported decreasing SC during the past ten years, SC stable_up means that households reported stable, increasing or variable shifting cultivation. Note that the Y-axis scale differs because of large income differences in the studied districts.

may further exacerbate the current trends of declining income and rice shortages in some households. In Hua Meuang, maize is grown in a shifting cultivation-like system with fallow, but success in terms of improving livelihoods seems to be absent thus far. This finding could be due to the very recent introduction of maize, which may not yet have translated into livelihood improvements. Alternatively, it could be that the contract farming deals are unfavorable for the farmers, which is confirmed by the rather mixed experiences with contract farming in Laos. A case study from the Kenethao District in the Xayaboury Province shows that contract farming can indeed be unfavorable for farmers. Here, verbal agreements between investors and maize growers are not formalized; if farmers cannot repay debts incurred from necessities such as road construction to maize fields and tilling, the investor can assume ownership of the land and rent it to the farmer (Manorom et al., 2011). On the other hand, Willi (2011) demonstrated that early pioneers of contract farming of maize in the Xiengkhor District in the Huaphan Province were fairly successful. Here, Vietnamese traders introduced maize in the early 2000s; because of the location along the border between Laos and Vietnam, local people developed good relations and trade connections with Vietnam. The more diverse farming and livelihood strategies in Hin Heub have provided livelihood improvements in the communities studied in this district. Shifting cultivation has produced positive effects for households that both expanded and intensified the production of upland crops in the studied communities, as seen elsewhere in

Southeast Asia (Cramb et al., 2009). On the other hand, we also found that many households abandoned shifting cultivation and focused on other farm or non-farm livelihood strategies. The Land and Forest Land Allocation (LFA) program in Laos has played an important role in these land use change processes, especially in the two northern districts that are located close to national protected areas. However, the opportunity to earn more income from cash crops appears to be the main driver of land use change in all three districts. In Hin Heub, grocery shops and salaries from jobs in the private sector and governmental institutions are also not related to the LFA, nor is paddy rice cultivation, which was classified as private property and practiced before the LFA was implemented elsewhere in Laos (Sandewall, Ohlsson, Sandewall, & Viet, 2010). Moreover, while the LFA has been fairly effective in reducing shifting cultivation, the ‘spared’ land has not been converted to forest but rather to land for other agricultural purposes, such as the production of maize and rubber. Overall, the livelihood impacts of land use and shifting cultivation change are very diverse and dependent upon the local context; as these developments are still fairly recent in many parts of Laos, conclusions regarding long-term effects remain preliminary. Cash crop producers tend to have larger land areas than non-cash crop producers, indicating that there is a risk of increasing inequity in villages that embark on large-scale cash crop production. However, if the communities in Hin Heub and other areas where land use has already been intensified can be considered a model for future development in the northern districts of Laos, there are good

10

T. Vongvisouk et al. / Applied Geography 46 (2014) 1e10

prospects for poverty alleviation and increasing incomes in these areas. Short fallow shifting cultivation may still play a role in this development, as farmers in Hin Heub have managed to both expand and intensify shifting cultivation practices. The production of rubber and maize in the northern districts is helping these communities experience similar development. Therefore, intensification processes and the conversion of shifting cultivation to more permanent types of agriculture are no guarantee that forests will be spared from the cultivation and expansion of cash crops. This theory has important implications for the future success of the LFA as well as the implementation of REDDþ, as both those with local interests and large-scale commercial operators have more to gain from cash crop plantations than from payments for carbon storage. It is therefore difficult to see REDDþ being successful unless it is coupled with other incentives to promote forest protection. Acknowledgments Funding for this study was provided by three research projects: ‘Transition of shifting cultivation systems at the agriculture/forest frontiers e sustainability or demise’ and ‘Property and Citizenship in Developing Societies’ (ProCit) funded by the Danish Social Science Research Council, FSE, and ‘Impacts of Reducing Emissions from Deforestation and Forest Degradation and Enhancing Carbon Stocks (I-REDDþ)’ funded by the European Community’s Seventh Framework Research Programme. We are also appreciative of the assistance from a Master’s degree student, Mr. Visien Singhaphanh and the following staff from Faculty of Forestry of the National University of Laos: Mrs. Phansamai Phommexai, Mr. Duangta Buaphavong, Mr. Ammalin Phanvilay, Mr. Metmany Soukhavong, and Mrs. Somsanook Arkavong, who all helped with field data collection. References Alton, C., Bluhm, D., & Sannikone, S. (2005). Para rubber study Hevea brasiliensis La P.D.R. Lao-German Program Rural Development in Mountainous Areas of Northen Lao PDR, GTZ, Vientiane, Lao PDR. Bouahom, B., Douangsavanh, L., & Rigg, J. (2004). Building sustainable livelihoods in Laos: untangling farm from non-farm, progress from distress. Geoforum, 35, 607e619. Castella, J. C., Lestrelin, G., Hett, C., Bourgoin, J., Fitriana, Y. R., Heinimann, A., et al. (2013). Effects of landscape segregation on livelihood vulnerability: moving from extensive shifting cultivation to rotational agriculture and natural forests in Northern Laos. Human Ecology, 41, 63e76. Cohen, P. T. (2009). The post-opium scenario and rubber in northern Laos: alternative Western and Chinese models of development. International Journal of Drug Policy, 20, 424e430. Cramb, R., Colfer, C. J., Dressler, W., Laungaramsri, P., Le, Q. T., Mulyoutami, E., et al. (2009). Swidden transformations and rural livelihoods in Southeast Asia. Human Ecology, 37, 323e346. Cramb, R., & Curry, G. N. (2012). Oil palm and rural livelihoods in the Asia-Pacific region: an overview. Asia Pacific Viewpoint, 53, 223e239. Fox, J., & Castella, J. C. (2013). Expansion of rubber (Hevea brasiliensis) in Mainland Southeast Asia: what are the prospects for smallholders? Journal of Peasant Studies, 40, 155e170. Fox, J., Fujita, Y., Ngidang, D., Peluso, N. L., Potter, L., Sakuntaladewi, N., et al. (2009). Policies, political-economy, and swidden in Southeast Asia. Human Ecology, 37, 305e322. Fox, J., Vogler, J. B., Sen, O. L., Giambelluca, T. W., & Ziegler, A. D. (2012). Simulating land-cover change in Montane Mainland Southeast Asia. Environmental Management, 49, 968e979. Hansen, P. K. (1998). Shifting cultivation development in northern Laos. In E. C. Chapman, B. Bouahom, & P. K. Hansen (Eds.), Upland farming systems in Lao P.D.R. e Problems and opportunities for livestock (pp. 34e42). Canberra: Australian Centre for International Agricultural Research. Heinimann, A., Hett, C., Hurni, K., Messerli, P., Epprecht, M., Jorgensen, L., et al. (2013). Socio-economic perspectives on shifting cultivation landscapes in Northern Laos. Human Ecology, 41, 51e62. Herold, M., & Skutsch, M. (2011). Monitoring, reporting and verification for national REDD plus programmes: two proposals. Environmental Research Letters, 6.

Hurni, K., Hett, C., Heinimann, A., Messerli, P., & Wiesmann, U. (2013). Dynamics of shifting cultivation landscapes in Northern Lao PDR between 2000 and 2009 based on an analysis of MODIS time series and Landsat images. Human Ecology, 41, 21e36. Kallabinski, J., & Lundgreen, D. (2005). Land use planning: an approach to poverty reduction and stabilisation of shifting cultivation in the Lao uplands to improve upland livelihoods. In B. Bouahom, A. Glendinning, S. Nilsson, & M. Victor (Eds.), Shifting cultivation and poverty eradication in the uplands of the Lao PDR: Technologies, approaches and methods for improving upland livelihoods e Proceedings of a workshop held in Luang Prabang, Lao PDR, January 27e30, 2004 (pp. 161e 168). Vientiane, Lao PDR: National Agriculture and Forestry Research Institute. Lestrelin, G., & Giordano, M. (2007). Upland development policy, livelihood change and land degradation: interactions from a Laotian village. Land Degradation & Development, 18, 55e76. Li, H. M., Ma, Y. X., Aide, T. M., & Liu, W. J. (2008). Past, present and future land-use in Xishuangbanna, China and the implications for carbon dynamics. Forest Ecology and Management, 255, 16e24. Linquist, B., Saito, K., Keoboulapha, B., Phengchanh, S., Songyikhangsutho, K., Phanthaboon, K., et al. (2005). Improving rice based upland cropping systems for the Laos PDR. In B. Bouahom, A. Glendinning, S. Nilsson, & M. Victor (Eds.), Shifting cultivation and poverty eradication in the uplands of the Lao PDR: Technologies, approaches and methods for improving upland livelihoods e Proceedings of a workshop held in Luang Prabang, Lao PDR, January 27e30, 2004 (pp. 299e 313). Vientiane, Lao PDR: National Agriculture and Forestry Research Institute. Manivong, V., & Cramb, R. A. (2008). Economics of smallholder rubber expansion in Northern Laos. Agroforestry Systems, 74, 113e125. Manorom, K., Hall, D., Lu, X., Katima, S., Medialdia, M. T., Siharath, S., et al. (2011). Cross-border contract farming arrangement: Variations and implications in the Lao People’s Democratic Republic. In Manila, Philippines: Greater Mekong SubregionPhnom Penh plan for development management research report series (Vol. 1. Issue No. 2). Asian Development Bank. Mertz, O., Egay, K., Bruun, T. B., & Colding, T. S. (2013). The last swiddens of Sarawak, Malaysia. Human Ecology, 41, 109e118. Mertz, O., Leisz, S., Heinimann, A., Rerkasem, K., Thiha, Dressler, W., et al. (2009). Who counts? Demography of swidden cultivators in Southeast Asia. Human Ecology, 37, 281e289. Mertz, O., Müller, D., Sikor, T., Hett, C., Heinimann, A., Castella, J. C., et al. (2012). The forgotten D: challenges of addressing forest degradation in complex mosaic landscapes under REDDþ. Geografisk Tidsskrift-Danish Journal of Geography, 112, 63e76. Messerli, P., Heinimann, A., & Epprecht, M. (2009). Finding homogeneity in heterogeneity e a new approach to quantifying landscape mosaics developed for the Lao PDR. Human Ecology, 37, 291e304. Padoch, C., Coffey, K., Mertz, O., Leisz, S., Fox, J., & Wadley, R. L. (2007). The demise of swidden in Southeast Asia? Local realities and regional ambiguities. Geografisk Tidsskrift-Danish Journal of Geography, 107, 29e41. Sandewall, M., Ohlsson, B., Sandewall, R. K., & Viet, L. S. (2010). The expansion of farm-based plantation forestry in Vietnam. Ambio, 39, 567e579. Schmidt-Vogt, D., Leisz, S., Mertz, O., Heinimann, A., Thiha, T., Messerli, P., et al. (2009). An assessment of trends in the extent of swidden in Southeast Asia. Human Ecology, 37, 269e280. Seidenberg, C., Mertz, O., & Kias, M. B. (2003). Fallow, labour and livelihood in shifting cultivation: implications for deforestation in northern Lao PDR. Geografisk Tidsskrift-Danish Journal of Geography, 103, 71e80. Shi, W. (2008). Rubber boom in Luang Namtha: A transnational perspective. LaoGerman Program Rural Development in Mountainous Areas of Northern Lao PDR, GTZ, Vientiane, Lao PDR. Sturgeon, J. C. (2010). Governing minorities and development in Xishuangbanna, China: Akha and Dai rubber farmers as entrepreneurs. Geoforum, 41, 318e328. Thongmanivong, S., & Fujita, Y. (2006). Recent land use and livelihood transitions in northern Laos. Mountain Research and Development, 26, 237e244. Thongmanivong, S., & Vongvisouk, T. (2006). Impacts of cash crops on rural livelihoods: a case study from Muang sing, Luang Namtha province, Northern Lao PDR. In S. Mahanty, J. Fox, M. Nurse, P. Stephen, & L. Mclees (Eds.), Hanging in the balance: Equity in community-based natural resource management in Asia (pp. 106e121). Bangkok and Honolulu: Regional Community Forestry Training Center for Asia and the Pacific (RECOFTC) and East-West Center. van Vliet, N., Mertz, O., Birch-Thomsen, T., & Schmook, B. (2013). Is there a continuing rationale for swidden cultivation in the 21st century? Human Ecology, 41, 1e5. van Vliet, N., Mertz, O., Heinimann, A., Langanke, T., Pascual, U., Schmook, B., et al. (2012). Trends, drivers and impacts of changes in swidden cultivation in tropical forest-agriculture frontiers: a global assessment. Global Environmental Change, 22, 418e429. Willi, Y. (2011). Of maize and men. Contract farming in the Houaphanh Province of the Lao PDR: Scale, scope and impacts on peasant livelihoods. Bern, Switzerland: Master’s Thesis. CDE Publications, University of Bern. World Bank. (2008). Lao People’s Democratic Republic: Policy, market and agriculture transition in the northern uplands. Washington, DC: World Bank.