Using Forests to Enhance Resilience to Climate

Using Forests to Enhance Resilience to Climate Change: Case Study of the role of forests in Adaptation strategies of Agricultural Households living in Savannakhet Province in Lao PDR

ForCC Project Technical Report

April 2013

(Note: This is a draft document. The final version will be published online at: www.profor.info/)

TFE““D 1

Acknowledgement This case study is part of CIFO‘ s Using Forests to Enhance Resilience to Climate Change (ForCC) project, administered by the World Bank Program for Forests (PROFOR). Funding is provided through a TFESSD-grant (099814). For the Lao PDR case study, the project involved a collaboration between CIFOR and the following national partners:  

Forestry Research Center (FRC) within the National Agricultural and Forestry Research Institute (NAFRI), and Department of Forestry (DoF).

The following scientists carried out the work for this case study: Aaron J.M. Russell 1 (team leader), Joost Foppes 2 , Sounthone Ketphanh 3 , Somphachanh Vongphasouvanh 3, Serge Rafanoharana 2, Bruno Locatelli 4 , Laykham Sihanat 55, Phayvone Phonephanom 5, Khonesavanh Louangsouvanh 5, Nellie Anyango Nakondiege 5 , Somcham Nanthavong 5 Contact email: [email protected]

1

Scientist, Center for International Forestry Research (CIFOR), Bogor, Indonesia Consultant, Center for International Forestry Research (CIFOR), Bogor, Indonesia 3 Deputy Director, Forest Research Center (FRC), National Agricultural and Forestry Research Institute (NAFRI), Vientiane, Lao PDR 4 Senior Scientist, Center for International Forestry Research (CIFOR), Bogor, Indonesia 5 Researcher, Forest Research Center (FRC), National Agricultural and Forestry Research Institute (NAFRI), Vientiane, Lao PDR 2

2

Executive Summary – English D la d Dipte o a p Fo ests DDF o e % of Lao PD‘ s total fo est estate, a d ep ese t a u i ue vegetative ecotype spanning parts of SE Asia. They provide direct ecosystem services such as timber and Non-Timber Forest Products (NTFPs) for human consumption and livestock grazing. They also support rice production through regulation of hydrological flows and limiting erosion. DDF land is targeted for conversion to commercial plantations, where large tracts of land are being allo ated to fo eig i est e t o pa ies ith o o pe satio fo lo al o u ities loss of benefits. National policy makers indicate a need for better data to inform decision-makers involved in balancing the needs of local rural livelihoods with national development priorities aimed at attraction of foreign investment. This study covers three villages in Savannakhet province: (a) a PSFM village , where no concessions occurred, and communal forest land covers 3.74 ha/capita; (b) a Eucalyptus village , where 45% of village forest land was converted to Eucalyptus plantations, and remaining communal forest covers 0.73 ha/capita; and (c) a Sugarcane village , where 75% of village forest land was converted to Sugarcane plantations, and where there is no remaining communal forest. This study assessed impacts of different land-uses on livelihoods and provides a quantification of ecosystem service values for each. The methodology combined secondary data collection with: national, provincial, district stakeholder consultations, village stakeholder PRAs, and an illustrative sampling of households were surveyed about household demographics, sources of income, expenditures, and food self-sufficiency. This study highlights the critical contributions from forests to provisioning and regulatory ecosystem services that reduce exposure and sensitivity of rural households to climate variability and increase adaptive capacity. In the PSFM village, forest resources are the primary source of cash income across wealth categories: NTFPs (55%), livestock sales (26%). The most important NTFPs were mushrooms (10%), frogs and snails (16%), insects (10%), bamboo and rattan (7%). DDF provisioning services were valued at $54 per ha per year. Across all three communities, forest resources constitute an average of 35% of annual total income. Communities with less forests relied to a greater degree on labor income (8 vs. 29-33%), on remittances from migration (1 vs. 5-13%), and had reduced livestock holdings. Due to limited demand for labor in sugarcane and eucalyptus concessions, the majority of labor income comes from outside the village. In the PSFM village, non-cash income represented 68% of total income compared to cash income. In the other communities, non-cash income represented 35-41% of total income. Total average HH incomes look similar ($1514-1736/year) however it should be underscored that this represents a significant decrease in food self-sufficiency as it costs more to replace gathered wild products by the purchase of cultivated products. The limited data available regarding the productivity of Eucalyptus and Sugarcane plantations on DDF soils suggests these to be high-risk investments. There are also numerous examples of conflicts between concessions and dispossessed communities. Such forest conversions must be recognized as relatively irreversible due to slow forest regeneration on shallow DDF soils. A CIFOR climate change analysis projects an increase in primary productivity for rice in Savannakhet province over the period 2030-2080, due to increased average temperatures with no big changes in rainfall. Very little data are available on the impact of climate change on livestock productivity. However, on the whole, stakeholders displayed low levels of concern with climate change. They

3

express much greater concern with the impacts on livelihoods of rapid, on-going allocations of concessions. The key findings on the impacts resulting from land use change can be summarized as: (a) Loss of forest land reduces livestock resources which are a key source of resilience; (b) Forests are not a poverty trap but a resilience strategy. DDF ecosystem services represent critical and sustainable resources that are used by all wealth groups, not only by the poor; (c) Conversion of communal forest land to concessions increases land conflicts; (d) The poor DDF soils limit the success of forest conversions to plantations ; (e) Fo est o e sio s a ep ese t aladaptatio s to li ate ha ge as the represent a high-risk investment, with limited local benefits and irreversible consequences. (f) Communities believe that DDFs regulate the supply of water to drinking wells and paddy fields and protect their land from erosion. More research is needed to assess and quantify these values. Livestock raised in DDFs are a key source of resilience for local farmers, compensating for the risks of rain-fed rice farming on drought-prone shallow soils. Their main sources of uncertainty are fear of, and conflicts associated with losing village forest land to concessions or other private investors. Their priority is gain to tenure security over communal forests. Lacking mechanisms to do so, there is significant pressure to subdivide the forest into individual plots (resulting in a net loss of ecosystem services to all). This study suggests several implications for policy-makers. Measurable values of forest ecosystem services should be incorporated in all decision making systems around conversion of forest land. More effort is needed to design drought-resilient livestock and agricultural systems. The Forest Investment Program (FIP) could address gaps in forest land tenure, integrate water management with forest management and engage communities in law enforcement, e.g. through payment for stewardship of ecosystem services.

4

້ ຍ

ຄັ ົ

ລາວ ່ າ ຄ

ວ

າຊ າ ວ ັ

ຽ

່ າ ຄ ່ັ

່ າວ ້

ຊະ

່ຄ ັ

່ າ ຄ ຍັ

່ າ ັ

້

່

າ

ະຸ ັ

້

່

ົ້ າ

າຍ າລັ

ວ ້ າ

ຍ ່ ້

ະຸ ັ າະ ັ

ະ ົ

ຍ

່ ຽວ ັ

່ຊ

າ

ະົ

ົ້ າ

ັ

້ າ ່ ່

້ າ

າັ

້ ັ

້ າ

ະ່

້ ັ

່ ຽ ່ າວ່

າ

າ້ າ

າລ ົ ຊວ ະ

້ າລັ

່ ຄວາ

າ ັ

າ ັ າ

ລະ ຍັ

າ

າ ຊ ້

ະ

ະ

າ ວ ່ ຽວ ັ

ະຊາ

ຽ ້ ວຍ ົ

າ ້ າ

່

າ ,

່ າ ຄ ່ຊ ້

ະ າ

່

່ າ ຄ ້ າ

ລະ

ະ ຍ າຍ ັ

ະ ລະ

ໆ ່

່

່ຊ ້

່

ຽ

າລາ

SUFORD) ້

%

າລັ ່ ວ ລວ

ຍ ້

ະົ

້ າ

.

່ ົ່ າ ັ

່ຊ

່ າ ້

່

ະ າຊ ົ

ັ

້ ວ

້

ຄ:

່ ່ າ ້ ລວ ້

ລະ

ຊາ

.7

/ຄ ົ ;

າລັ

າ ຊ ້ ່

າ

ລະ ຄ.

.7 %;

່ າ ້ ລວ ້ າ

7 %

ະຸ ັ

່

່ າ ັ ່

ລ າ

າ ລະ ົ

ວ

່ ັ

່

ວ ້ ລຍ.

ະົ າ

.

່ ່ າ ້ ລວ

ລະ

າ

ົ

້ ວຍ.

າ ລະ

ລະ

.

ຍ,

າ້

າັ

່ ຽວ ັ

ະ າ ັ

້ ຄ ່ າ

ັ

າ

ະ ວັ

່

າ ະ າ ້

ຊ ົ ຸ ຊ

່ ລະ ະ

ລວ

້ ວ

າົ

າ ວ

່ ລ ້ ຽ ັ ້

ະ ັ້

ະ າ

ະລ

າ ະ າ ່

້

ັ

າຄ

ຸ

ວ

າ

າ

ລະ ້

າ ະ າ .

າ

່ າ

ລັ

້

ະ ້

່

ະ ະ າ,

້ ຸ

້

າ ລ ະ ັ

່ າ ັ

ະ າຍ ່

່ າ ໆ

່ ່ າ ້ ລວ ້ າ ຍັ ້

ົ້ ັ

່

າ້ າ ້ າ ັ

່ ້

ະ

າລະ າ ັ

ັ

ັ

່ ້

ັ

ຍົ ັ

າ ່

າຍ ັ

່ າ

້ າ

້ າ ່ ັ

ົ້

້ າ ້ າ ຊ

າ າ ່

່ ້

ັ

າ ຊ ້ ້

າ ລ ົ

້ າ ່

ວ ວ ະຍາ

່

ະ

້

້

າ ັ

້

າ

ວ

ລະ

ັ

ລ ົ

າ ະ າ ່

ັ ່ ຽ

າ

ົ າ າ

້

້ າ ່ັ

ະ ້

ຄ

ລະ ົ

ົ າ

າ

່

າຍ ່

ວ ້

່ ຄ ົ

າ ລັ

ັ

່

້ າ ່ ັ

າ

ັ

່ າ ໆ າ ລະ ົ

ະ

່ ຽ ັ

້ ັ

ຄວາ

ລະ ັ ົ

.

າ ້ າ ຄວາ

າ້

າ

ະລ

່

ຊວ

ຸ ້

ລ າ ້

າ

່ າ ໆ

ຄວາ ົ

່

ຊາະ ່

ະ າ

້

່

າ

າ

່

້

ັ ັ້

້

າ

່ າ ໆ ົ

້ ຊ ່ ວຍ ັ

່ າ ຄ

ັ

ະ

້ .

່ ່ າ ້ ັ ້

%

່ ວ ່ ວ

PRA)

ລະ

ຄ

ຄ ົ ວ, ັ

າ

າ າ .

5

່ າ ຄ

່ ລະ ້ າ ,

າ າ ່ ວ

່ າ ໆ

່

າ

ະ ລາຍ ັ

ລະ

ລະ ັ

າຍ າ

າ ຄ

າ ,

ຊ ຸ່ ຄ ົ ວ,

່ຊ

າ ຄ ້ າ ົ ້ ຄວ

ວ

ລະ

ົ ວ່ າ ັ າ

າຄ

ຊ ້ ່ າຍ ່ າ ໆ

,

ຄ ົ ວ

ລະ

້ າ

້ າ ່

າລາ ້ າ ່

່ ຊະ

່ ່ຄ

ັ

່ າ ຄ າ

າ

ັ

ລະ

55%

ຍ 16%,

່ ັ

ລ າ

%),

າ

າ

າ

ວ ່ າຊັ

ຄວາ ້

ັ ້ າ

ລາຍ ັ

່ ່

່

ຄ ່ າ ະ ລ ່ ຍ ່ັ ່

ລາຍ ັ

າຄ ່ າ ່ັ

່ າ ້

ຍາ ່

າ

່

າຍ ັ ້

ລະ ວ ່ າ

ະ ັ

າ ັ

່ ່

ະ ັ

່

ຍ ະ ລ ່ ຍ 35%

າ

ຊ ້

ວ ່ າ

້ າ

ລະ

(8 vs

(1 vs 5່

າ ວ ່ າ ັ້

່ັ

້ ວ

່ ົ

ຽ

35-41%

ຽ

ຄຽ ຄ ່ ັ

ັ

ລາຍ ັ

, ັ ົ

່ າ ່

. າ

າ າ ່

ລ ົ

າ ຸ

່

ລາຍ ັ ັ

ຄ 1,514$-1,736$/ ,

ວ

ຸ້

ະລາ . ້ ວ

້ ຍັ

, ່ ຍ ້

ລະ

ວ ່ າ ະ າ

່ າ ຄ

ົ ວ ລ ້ ັ າ ຸ

າລາຍ

່ ້

ຍ ້

າ ົ ວ່ າ ັ

າ ວ ່ າ

່ າ ້

າ

າ

າ

ຄວາ ່ ຽ

້ ວ ່ າ ັ

້ າ .

ັ

້ າ ຄ າ ຊ ະລ

າ

ຍ

ົ

້ ັ

າ າ

ະ

/ .

າ ່ ່

້

ັ

ະ ະຄ ້ າຍຄ ັ

ລະ ຊາວ ້ າ . ່ າ ຄ

ຄ

້ ັ

່ ຽວ ັ

າ

າ ້ າ ລາຍ ັ

າ

ະລ

ະ

າ ລ ົ

າ ະ າ

ະ ັ

າ

າ

ລາຍ ັ

າ ະ າ

່ າວ ັ

າ

ຄ ່ຊ ັ : ັ

ັ

ລາຍ ັ

ລັ ົ

າ

າັ

່

ລາຍ ັ

ົ ວ ລ ່ັ

່

້ ້ ວ ່ າັ

າ

ັ້

ຄ ົ ວັ

່ າວ ້ ່ ຽ

່

າ ້ ້

ຄ

້ ະລ າ ຊ ັ

າ າ

າ ະ າ ່

ຍ ້ ້

ຄ ົ ວ ່ າ ໆຄ:

ົ້ ົ

ຍລ ົ

68%

າ

່ ວ

າ

້ ້

ັ

້ ລະ ຸ ວ ່ າ

ັ

ັ

ະ

້

າ

າ ້ າ

່ າວ ລາຍ ັ

2 ້ າ ່

ຄວ

າ ້

້ າ ່ັ າລາ

່

ັ

າ ່

່ າ

ວາຍ 7%.

SUFORD)

າ ະຄ

ຄ ່ າ ະ າ 54$/ ັ

່ ່ າ ້ ຍັ

າ

່ຄ

ລະ

່ າ ້ ່ າ ໆ ້

າ ລ ້ ຽ ັ

າັ

່

ຍຄ

ລະ ັ

ັ

້

ະ າຊ ົ ະົ ັ

ົ້

ລ ້ ຽ 26%. ຊະ

່ າ ໆ

້

ັ

ລະ

້ າ ລາຍ ັ ່

່ າ ໆ 10%, ່

%),

້

ັ

ະຍາ

າ

່ າ ະລ ັ

າຍ ັ

້ າ ່

າ ້

ັ

່ າວ

າ ້ າ

, ່ຊ ົ

າ

້ ຊະ

3້ າ , ັ ົ

ລາຍ ັ 9-

ົ

ະ ້

້

່ າ

ລະ

10%, ົ

່ ່ າ ້ ່ັ ້

າ ້

ຄ

ຍ ່ ່ າ ໆ

່ ຽ

່ າ ຄ

ລ ້ ວ, ້

ັ

ລະ

່ . າ ວ ຄາະ ະ າ

າ ່ ຽ

ະລ ັ

ໆ

ວ

່

ລະ

າ

າ ່ ຽ າ

່

ົ

ັ

ວ

ົ າ ົ້ າ ັ

າ ະ ວ ະົ

າ

່ າ ໆ່

້

່ ະ

່

້ ວ ຄາະ ລ

ຄວາ

ະຸ ັ າ

ຄ ້ າ່ າ ້ ົ ້ ຄວ

າ

ະລ

ົ ້ າ ່ຊ

ຸ

ະ

າ ະ

າ ່ ຽ

້

່ຊ . ຍ ່ ຽວ ັ

່ ວ ່ ່ ັ າ ່

່

າ ຊ ້ ່

ົ

້

ລຍະ

່ າ

າ

ະົ

.

າ ່ ຽ ່ ຊວ

ັ ລວ

2030-2080,

າຍ

ະ ັ້

່ ຽວ ັ

າ ລ ້ ຽ ັ

ຍ

ລວ

ລ ້ ວ

້ າ າ າ າ ັ

. າ າ

6

ັ

້ ວ ຄາະວ ່ າ

CIFOR ່

າ

າຍ

າ ້ າ

ັ ະຸ ັ

າ

ະ

າົ

ະ ລ ່ ຍ ້ ວຍຄ ່ າ່ ່ ່ ຽ

າ ່ ຽ

ຄວາ ົ

ະ

າ

່

ະ່ ົ

ຍ

່ າ ໆ ້

ະ ່

່

າ

້

າ າ ່ ວ ັ

ະ ວັ

ະລ າ ້ ້

່

້ າ າ າ

ຸ່ ້ : ້ ່ັ ລ

່

ະຊາຊ ົ

ລະ

່ ່ າ ້ ຸ ້

.

າ ່ ່ ຍ ະ າ ່ ຸ ຍ ່ າ ໆ າ ລະ ົ

າ

່

ະ ົ ຍ ຍ ົ ່ ຽ າ

.

່

່ ່ າ

ະ າ າ

ັ ່ ຄັ ່

ວ ້

າ ່

ລ ້ ຽ ່

້ ່

່

້ າ ຄ

ັ

ຍ ົ້ າ ະັ

ຊາວ ້ າ ່ ວ າ ລ ົ

່ າ ໆ າ

າ

້ ້ ຊ

່ ຊ ຄວ ່ ່

າ ້ າ

້

ລະ ົ

ະ

່ ຽວ ້ ະ ັ

ັ

່ ຽ ້

າ ັ

ລະ ຽ ົ

ັ

າ ັ

ລະ

່

ະ ຄວາ ່ ລ ່ຊ ັ

ະ ົ

່ຊ ່ າ ັ ວ

າ ັ

ລະ

ໆ ັ

່ າຍ່

າຄວາ

່ າ ້ ່

່ າ ັ

;

ົ ວ” (maladaptations) ້ າ າ າ ່ ່ ່ ່ ຍ າັ ້

່ຊ

ລ

້ າ

ຄັ

າຍ ່ າ ້ .

7

ະ ້ ຍ ົ

ະ

ັ

່ າ

້

າ

ວ

ົ ວ. ະ ຍ າຍຄ:

ັ

ວ ວ ະຍາ ່ າ ້ ັ

າ່ າ ໆ ັ

າ ລ ້ ຽ ັ ລະ ຊ ່ ່ ັ

້

,

່ ່ າ ້ ລວ ້

້ ຍັ

າ ລະ ົ ລະ

່

າ ະ າ ່

ັ ຽລະ ົ

າລະ ົ

,

ະຸ ັ

າ

າຸ ້

່

ຍ ັ

ັ

າ

່

າ

(FIP)

າ

າ ້

້

ຄົ

້ າ

້

ລ ້ . ້

ຊາວ ້ າ

ຄ ້ າ ະ ້ ົ ້ ຄວ

ັ

່ ຽ ັ

;

າລັ ຊາວ ະ

າ

ັ າ ລ ົ

້

າ ້ າ ຊັ ະຍາ ່ າ ຊາະ ້ ັ ້ ລະ ຄ ຄ ້ າ າ ່ າ ່ົ ລາ ້ . ົ ຄວ

ຍ ຍາວ າ

ຄ ່ າ່ າ ໆ

່ າ ້ . ັ

ລະ ລະ

ຍ ້

ັ

ະ

ະ

າ ່ າ ່

າ

ົ

າ ັ

ຊ ້

າ ່

່

ະຍາ

າ ້ ັ ້

້

າ

ົ້

ລ ່ ້

່ າ ຄ . ຄ ່ າ ຄ

່ າ ຄ າ

າຄັ

ັ

ັ

ຊັ ້ ຊ ົ ່

ັ

າ ະຊາ

ັ

າ ຄ ະ າ

າ ່

ັ້

ະ ້

າ າ

າ

ລ ະ

່ ່ າ ້

ຽວ

່ ;

ຍາ

ຄວາ

່ ່ າ

້

າ ລ ົ

່ ່ ່

ະຊ ົ . ້

າ້

າ

່

ຍັ ່

ັ້

່ າ ໆ

ຄວາ ່ ຽ

ຍ ່

່ຊ ້ າ າ ່ ່ ັ

ຄວາ

່ າ ຄ

ລ ົ ;

າ ລ ັ ົ

າ ່ ວ ່ າ “ຍາ

ຄວາ ່ ຽ ່ ້ ັ ່ າຄ ັ

າ ະ າ

າັ

ະົ

. ຊ ່ າ ັ້ ວ ົ ົ ່ ັ ່ າວ ່ ຊ ຸ ຊ ່ຊ ່ າ້ າ ້ າ , ່ ້ າ າລັ ັ ວ ຍັ ຄວາ ້ ່ ່ າ ັ

້ ່

າ

່

້

ັ

່

່ າ ຄ

ະ ້

ັ

ລະ ຍາ

່

່ ່ າ ້

່ ຽ ້ າ ່ ຽ ັ

າ ຸ ຄວາ ຸ ຍາ ່ ວ , ັ ົ ັ ່ ວ ວ ະຍາ ່ າ ຄ ້ ັ າ າ ະຄ ຄ ັ ົ ວ ຸ ລະ ັ າ ະ ຸ ຍາ ົ່ າ ັ້ .

່ ່ າ ້ ລວ

່ ຽ ້

ຊາວ ້ າ ຸ

າ

ຄ ົ ວ່

າຄ

່ ັ

ັ

ຄ ຸ

.

າ ລ ້ ຽ ັ

າ ່

. ່ າ ້

ຄ.

າ ້ ່

ລ ົ

້ ຊ ົ ຸ ຊ

ລະ

ວ ່ າ ່ າ ໆ ົ້ າ

ລະ

Table of Contents Acknowledgement .................................................................................................................................. 2 Executive Summary – English ................................................................................................................. 3 ົ

ຄັ

້ ............................................................................................................................................... 5 ຍ

Table of Contents .................................................................................................................................... 8 List of Figures .................................................................................................................................... 11 List of Tables ..................................................................................................................................... 12 List of Acronyms and non-English/Latin terms ................................................................................. 13 1

2

3

Introduction .................................................................................................................................. 14 1.1

Using forests to enhance resilience of agriculture to climate change.................................. 15

1.2

Scope of Study....................................................................................................................... 16

Research Context .......................................................................................................................... 16 2.1

Forests in Lao PDR................................................................................................................. 16

2.2

Ecosystem services from Forests in Lao PDR ........................................................................ 18

2.3

Savannakhet Province ........................................................................................................... 19

2.4

Forests in Savannakhet – Dry Dipterocarp Forests ............................................................... 19

2.5

Agriculture in Savannakhet ................................................................................................... 20

2.6

Land use change in Savannakhet .......................................................................................... 21

2.7

Demographic trends ............................................................................................................. 22

2.8

Trends in Migration............................................................................................................... 22

2.9

National Policy Context ......................................................................................................... 23

A

Role of Forests in Climate Change Adaptation policies in Lao PDR ...................................... 23

B

Policies related to Forests and Concessions ......................................................................... 23

Study Approach ............................................................................................................................. 25 3.1

Study sites ............................................................................................................................. 25

3.2

Methodology......................................................................................................................... 27

A

Data collection methods ....................................................................................................... 27

B

Stakeholder consultations and workshops: .......................................................................... 27

C

Village level data collection .................................................................................................. 27

D

Household Socio-Economic Survey ....................................................................................... 28

3.3

Data Analyses ........................................................................................................................ 28

A

Economic valuation for non-cash ecosystem services.......................................................... 28

B

Contributions of Livestock Assets to Resilience .................................................................... 28

8

4

5

6

C

Household food security and nutritional diversity ............................................................... 28

D

Caveats regarding intra-household differentiation in vulnerability ..................................... 29

E

Focus Group discussions ....................................................................................................... 29

F

Economic valuations of ecosystem services ......................................................................... 29

G

Climate change scenario modeling of changes in ecosystem services ................................. 30

Climate change in the region ........................................................................................................ 30 4.1

Historical and future projected climate conditions .............................................................. 30

4.2

Projected impacts of climate change on ecosystem services ............................................... 32

4.3

Projected impacts of land-use change on ecosystem services ............................................. 32

Livelihood vulnerability analysis ................................................................................................... 33 5.1

Primary sources of cash income ........................................................................................... 33

5.2

Household Non-Cash Income................................................................................................ 35

5.3

Contributions of NTFPs from Focus Groups.......................................................................... 36

5.4

Household Expenditures ....................................................................................................... 37

5.5

Balancing household cash income and expenditures ........................................................... 39

5.6

Contribution of assets to resilience: Land ............................................................................ 40

5.7

Contribution of assets to resilience: Livestock ..................................................................... 41

5.8

Household Food Security and Nutrition ............................................................................... 43

The values of Dry Dipterocarp Forests Ecosystem Services.......................................................... 45 6.1

Valuation of provisioning services ........................................................................................ 45

6.2

Valuation of regulating services............................................................................................ 48

6.3

Valuation of Co-benefits ....................................................................................................... 48

6.4

Total value of Dryland Dipterocarp Forest Ecosystem Services ........................................... 49

6.5

Substitution Values of Conversion to Plantations ................................................................ 50

6.6 Summary of impacts of concessions and PSFM programs on the provision of ecosystem services.............................................................................................................................................. 52 Dry Forest Contributions to Resilient Agricultural Livelihoods in Laos: A Policy Brief ......................... 54 7

Literature Cited ............................................................................................................................. 59

8

Annexes ......................................................................................................................................... 63 Annex 1: Case study 1: PSFM village ................................................................................................. 64 A

Land use history .................................................................................................................... 64

B

Analysis of household vulnerability ...................................................................................... 65

C

Household Sources of Cash Income ...................................................................................... 65

D

Total Household Income (Cash plus Non-Cash) .................................................................... 66 9

E

Household Cash Income vs. Expenditures ............................................................................ 66

F

P“FM illage – Detailed household income and expenditures ......................................... 68

Annex 2: Case study: Eucalyptus village ........................................................................................... 69 A

Land use history .................................................................................................................... 69

B

Analysis of household vulnerability ...................................................................................... 70

C

Household Sources of Cash Income ...................................................................................... 70

D

Total Household Income (Cash plus Non-Cash) .................................................................... 71

E

Household Cash Income vs. Expenditures ............................................................................ 71

F

Eu al ptus illage – Detailed household income and expenditures ................................. 73

Annex 3: Case study: Sugarcane village ............................................................................................ 74 A

Overall land use history ........................................................................................................ 74

B

Analysis of household vulnerability ...................................................................................... 74

C

Household Sources of Cash Income ...................................................................................... 74

D

Overall Household Income (Cash plus Non-Cash) ................................................................ 75

E

Household Cash Income vs. Expenditures ............................................................................ 75

F

“uga a e illage – Detailed household income and expenditures .................................. 77

Annex 4. Range of estimated cash and non-cash income values (Sources: survey and PRA exercises) .......................................................................................................................................................... 78 Annex 5. Market prices for primary NTFPs collected in case study communities, Savannakhet Province (April, 2012) ....................................................................................................................... 79 Annex 6: Key policies impacting forests and concessions ................................................................ 80

10

List of Figures Figure 1: How Ecosystem Services relate to Climate Change: a conceptual framework (Source: Locatelli ea, 2008) ......................................................................................................................................... 16 Figure 2. Changes in Forest and Land-use during 1982-2002 (from Vongsiharath 2006) .................... 17 Figure 3: Map of land and forest classifications in Savannakhet Province (Source: PAFO).................. 20 Figure 4: Migrating Population Pyramid (Source: ILO-IPEC 2003) ........................................................ 23 Figure 5: Current mean annual temperature (left) and precipitation (right) in Lao PDR (star: Savannakhet). ................................................................................................................................. 31 Figure 6: Current monthly mean temperature (line) and precipitation (bar) in Savannakhet (Rafanoharana et al 2012) .............................................................................................................. 31 Figure 7. Extreme scenarios of impacts on NPP for climate and land-use change projections (2030, 2080)............................................................................................................................................... 33 Figure 9: Overview of household cash income across three case studies ............................................ 34 Figure 10: Overview of household non-cash income across three case studies .................................. 35 Figure 12: Household expenditures by wealth group in three villages ................................................ 38 Figure 13: Household expenditures by village ...................................................................................... 39 Figure 14: Total land area available per household in three case study communities ........................ 41 Figure 15: "PSFM village" Swedish-Lao Forestry Project Land Use Map (1999) ................................... 64 Figure 16: Illustration of divergence between PSFM village data sources (Source: Household survey and PRA) ......................................................................................................................................... 67 Figure 18: Illustration of divergence between Eucalyptus village data sources (Source: Household survey and PRA).............................................................................................................................. 72 Figure 19: Illustration of divergence between Sugarcane village data sources (Source: Household survey and PRA).............................................................................................................................. 76

11

List of Tables Table 1 : Area and productivity of top 5 industrial crops in Savannakhet Province, 2009 (Source: IUCN and NERI 2011) ............................................................................................................................... 21 Table 2: Key Village Demographics ....................................................................................................... 26 Table 3: Estimates of labor emigration to Thailand (Source: 2005 Census) ......................................... 26 Table 4: Household Cash Income in three Dry Forest Villages, Savannakhet, 2012 ............................. 33 Table 5: NTFP contributions to total household cash income (Source: PSFM village PRA Discussions) ........................................................................................................................................................ 37 Table 6: NTFP contributions total household non-cash income (Source: PSFM village PRA Discussions) ........................................................................................................................................................ 37 Table 7: Cash incomes and expenditures in three villages and wealth groups .................................... 39 Table 8: Contribution of livestock assets to household resilience ....................................................... 42 Table 9: Food Security - Rice consumption across three communities ................................................ 43 Table 10: Food Security - NTFP consumption across three communities ............................................ 44 Table 11: Values of livestock per hectare for three land use types in the PSFM village ...................... 46 Table 12: Values of Annual Non-Timber Forest Products sold from three land-use types in the PSFM village.............................................................................................................................................. 46 Table 13: Value of timber resources in three land use types in the PSFM village ................................ 47 Table 14: Time needed to grow to harvestable size for three dominant species in dry dipterocarp forests in Savannakhet ................................................................................................................... 47 Table 15: Villager's estimate of annual consumption of timber for house construction in the PSFM village.............................................................................................................................................. 47 Table 16: Values of firewood per hectare from three land use types in the PSFM village .................. 48 Table 17: Watershed regulatory values of dry forests, calculated as 25% of total rice production value in the PSFM village ......................................................................................................................... 48 Table 18: Value of DD forests in terms of potential carbon credit ....................................................... 49 Table 19: Economic values of dry forest ecosystem services as annual returns to the PSFM village .. 49 Table 21: The value of Dry Forest Ecosystem services as livelihood assets in the PSFM village .......... 50 Table 22: Economic values of Eucalyptus plantations on former Dry Dipterocarp forest soils. ........... 51 Table 23: Economics of sugar cane farming on former Dry Dipterocarp Forest soils. ......................... 52 Table 20: Available land per Household (HH) in the PSFM village........................................................ 65 Table 25: Paddy and Forest converted to Eucalyptus........................................................................... 70

12

List of Acronyms and non-English/Latin terms Acronym/Term chap chong din DAFO DDF degC DoF FDI FIP FOMACOP FRC GOL ILO-IPEC JFM khao peuak khao san LUC MAF mai chik mai koung mai hang mai sat MOU NAFRI NAPA NGO NSCC NTFP PAFO PDR PFA PRA PROFOR PSFM TFESSD WRI

Definition/Translation Land claimed by an individual for future expansion of paddy District Agriculture and Forestry Office Dryland Diptercarp Forest Degrees Celsius Department of Forestry Foreign Direct Investment Forest Investment Program Forest Management and Conservation Program Forestry Research Center Government of the Lao PDR I te atio al La ou O ga izatio s I te atio al P og a e o the Elimination of Child Labour Lao-Swedish Joint Forest Management project Non-milled rice Milled rice Land Use Change Ministry of Agriculture and Forestry Shorea obtusa Dipterocarpus tuberculatus Shorea siamensis Dipterocarpus obtusifolius Memorandum of Understanding National Agriculture and Research Institutes National Adaptation Programmes of Action Non-Governmental Organization National Strategy on Climate Change Non-Timber Forest Product Province Agriculture and Forestry Office Lao People s De o ati ‘epu li Production Forest Area Participatory Rural/Rapid Appraisal Multi-Donor Trust Fund for Program on Forests Sustainable Forest Management through Rural Development Multi-Donor Trust Fund for Program for Environmentally and Socially Sustainable Development World Resources Institute

13

1 Introduction Lao PDR is a landlocked nation in south-east Asia with a total land area of 23.68 million ha. It shares borders with China and Myanmar in the north, Vietnam in the east, Thailand in the west and Cambodia in the south. Lao PDR has approximately 9.5 million ha of forest cover that constitutes 40% of the total land area. Approximately 70% of the country land area is classified as sloping hillsides and mountains. The population growth rate is 1.8% annum. By 2010, the total population reached 6.25 million people. The most concentrated areas are towns located along the Mekong River and main tributaries including the capital city of Vientiane. % of the atio s populatio lives in rural areas (NSC 2009). Lao PDR has experienced a steady economic growth of 6.5 percent per annum on average between 1990 and 2009 despite the regional and global financial crisis. This is largely due to the demand for resources from its neighbors -China, Thailand and Vietnam, and from the foreign direct investment (FDI) inflows over the past decade. Gross domestic product (GDP) per capita has risen from USD 818 in 2007-2008, to USD 906 in 2008-2009 and USD 1018 in 2009-2010. The Go e e t of Lao PD‘ s th (GoL) Seventh National Socio-Economic Development Plan (7 NSEDP) for 2011-2015, approved by its National Assembly in June 2011, spe ifies the Go e e t s i te tio to attract significant FDI and to increase forest cover from 40% in 2010 to 70% by 2020. These developments have had positive impacts on poverty reduction in Lao PDR - the share of poverty was reduced by 30 percent in one decade lifting one eighth of the total population out of poverty. Despite the positive achievements, the poverty rate in Lao PDR remains higher than in neighboring countries. Furthermore, risks of relying on unsustainable resource management practices include increased environmental degradation and inequality. Sustainable forest and agriculture resource management is important when considering the gross domestic product figures for 2008-2010, when the sectors together contributed 30%. They also provided 75% of total employment. Domestic consumption of wood and non-timber forest products (NTFPs) are valued at around USD 31.4 million, while exports are valued at around USD 74.4 million. In rural areas, it is estimated that NTFPs contribute between 30-70% of income for forest-dependent households (World Bank, 2012). However, evidence is building to show that much FDI is aimed at conversion of forest resources, and a recent study by IUCN/NERI (2011) of the impacts of concessions in Savannakhet indicated a number of potential adverse impacts on livelihoods and ecosystem se i es. This a u de i e the go e e t s o je ti es of i easi g fo est o e a d sustai a l poverty alleviation. Sustainable resource management is also important for adapting to climate change. Floods and droughts are the major natural disasters for Lao PDR (GoL NAPA, April 2009). Most of the flooding occurs during May to September, when Monsoon rains accumulate in the upper Mekong river basin and result in river basin flooding. Some of the key climate change trends for Lao PDR suggest that number of droughts and floods over the last three decades has increased. It is estimated that the south and central regions of the country, where about two thirds of the ou t s populatio li e, fa e on average 1.5 serious floods or droughts every year (GFDRR, YEAR). There has been a decrease in total rainfall between 1961 and 1998 and an average increase in temperature of 0.1 to 0.3°C per decade between 1951 and 2000. An increase in temperatures along with a decrease of rainfall during the dry season might lead to longer and severe droughts. Climate change and the increase in 14

frequency and in magnitude of these events are expected to make more people food insecure in particular in the rural areas. (World Bank, 2011, Adaptation Portal) Agriculture in Lao PDR is mostly dependent on rainfall. In some parts of the country, more than 80 percent of the households rely on the use of rain water for rice cultivation. Rice, in turn, is central for household food security. A low intensity of natural disasters, therefore, can increase the vulnerability of rural farmers because of the high degree of poverty in these areas. It is expected that climate change might threat and lead to a loss in the agriculture production (in particular rice) affecting the economy of the country and the food security (World Bank, 2011, Adaptation Portal)

1.1 Using forests to enhance resilience of agriculture to climate change Climate change is likely to impact different populations differently depending on their respective levels of vulnerability, characterized by McCarthy et al (2011) as: the degree to which a system is susceptible to, or unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate of climate variation to which a system is exposed, its sensitivity, and its adaptive capacity. Forests, by providing various ecosystem services, can reduce the vulnerability of populations. The ecosystem services Ecosystems services from forests can be classified broadly in four groups, as follows (MEA 2005):    

supporting services (e.g. water and nutrient cycling), provisioning services (e.g. forest foods, fuel wood), regulating services (e.g. water purification, erosion control) and cultural services (e.g. esthetic or spiritual values).

Societal decision-making with regard to each ecosystem service type has the potential to assist in the reduction of vulnerability in specific ways. The definition of vulnerability adopted by the IPCC (2007) in the context of climate change reads as follows, the degree to hi h a syste is sus epti le to, or u a le to ope ith ad erse effe ts of li ate ha ge , and builds on a conceptualization of vulnerability to Climate Change being composed of three components: exposure, sensitivity and adaptive capacity (MEA 2005). These have been defined by the IPCC (2007) as: 

 

Exposure is the ature a d degree to hi h a syste is exposed to significant climate ariatio s . Sensitivity is the degree to hi h a syste is affe ted, either ad ersely or e efi ially, y climate-related sti uli . Adaptive capacity is the ability of a system to adjust to climate change, to moderate damages, to take advantage of opportunities or to cope with the consequences .

The contributions that each of type of ecosystem service makes in shaping the exposure, sensitivity, and adaptive capacity of the socio-ecological system is illustrated in the figure below (Locatelli et al, 2008). The conceptual underpinnings of ecosystem-based adaptation and applications to analysis of the contributions made by forests to resilience made in this project can be reviewed in Russell et al (2011).

15

Figure 1: How Ecosystem Services relate to Climate Change: a conceptual framework (Source: Locatelli et al, 2008)

1.2 Scope of Study The scope of this study is to explore how, if at all, forests, through the provision of ecosystem services, can contribute to the adaptation of agriculture in the southern part of Lao PDR. The study focuses on adaptation of small holder agriculture in a Savannakhet province, where dry dipterocarp forests are the dominant forest type. Due to beneficial topography and climate, Savannakhet Province is also known as the breadbasket of Laos, representing over a quarter (619,950 ha) of the total wet season rice paddy area under cultivation in the country (619,950 ha) (LAC 2012). With the development of significant infrastructu e suppo ti g “a a akhet s i teg atio ith the e o o ies of Thaila d a d Vietnam, there is growing pressure on policy makers to permit forest conversion for agricultural and other plantation concessions. Concerns were raised by senior officials in the Department of Forests (DoF) regarding the conversion of forests in this region about the impact of the conversion of forests, including whether it has implications for how households adapt to climate change. This case study is part of a larger set of studies on using forests to enhance resilience to climate change in different sectors.

2 Research Context 2.1 Forests in Lao PDR Lao PDR had 9.5 million ha of forests in 2010. Between 1982 and 1989, forest cover declined from 49% of total land area to 47%. During the same period, the area of potential forest (are with less than 20% canopy cover and areas classified as degraded forest) increased from about 8.5 million ha to about 11 million hectares (FIP IP, 2011).

16

Table 1: Land use in Lao PDR in 2010 (Source: DOF presentation in Annual MAF Conference in January 2011 as cited in FIP IP, 2011)

Land use type

Area (million ha)

Current forest area

9.5 (40%)

Potential forest area (stocking 10,000ha.

While it is not unusual for a government to grant limited concession allocation powers to lower levels of government, the challenge in Laos is that there are not required to coordinate or communicate these through any central authority. In effect, these levels of government are competing with each other for sources of revenue to be collected from the concession-holders, and consequently have little incentive to communicate or show transparency regarding concessions to each other (Barney and Canby 2011; Schoenweger and Uellenberg 2009, Schumann et al 2006). In addition, Barney and Canby (2011) discuss the challenges to enforcing accountability on government institutions by the conflicting reporting and harvesting guidelines stipulated in the Lao Forestry Law (2007) and several Prime Ministerial decrees and orders. On a more local level, concessions have come under wide-spread criticism for coercive tactics, and many instances where concession-holders have cleared lands already in use by local communities, with little or no consultation or compensation (CIDSE 2009; Schoenweger and Uellenberg 2009; Kenney-Lazar 2010). Periodical nationwide outcries over a range of concession-related injustices, rapid national deforestation, and government complicity in both, have led to go e e t s i positio of numerous moratoriums on logging and concessions in general, none of which appear to have been very effective at altering the rate of land conversion and forest cover loss (Lang 2001; Barney 2011; Dwyer 2011; Kenney-Lazar 2010; Schoenweger and Uellenberg 2009; Vientiane Times): 

  

Aug.1991– temporary nationwide ban on logging, and ban on provincial concession allocations; May 2007 - moratorium on all concessions over 100ha in size, repealed in May 2009; July 2009 – moratorium on new land concessions over 1,000ha in size; June 2012 – moratorium on new mining and tree cropping concessions until 2015.

This most recent moratorium may be a potential window of opportunity to influence policies to fully consider ecosystem service values, dynamics of rural livelihoods and climate adaptation issues within the land and development planning and decision-making processes. In addition to forestry policies, forests rural stakeholder livelihoods and the concessions that are changing the landscape interact with a number of other national policies. Several of the most significant policies identified as impacting the case studies covered in this report are listed in the Annex 9).

24

3 Study Approach 3.1 Study sites The selection of the three study sites was done with the Provincial and District authorities. The selected villages were to provide particularly useful contrasting scenarios for analysis of the impacts of current agricultural and tree cropping concessions on local vulnerability and ecosystem services in DDF lands of Savannakhet Province. 

 

The first village, la eled the Participatory Sustainable Forest Management (PSFM) village , is situated in the northern part of Sonbouli District. Part of its forest land lies within the Dong Kapho National Production Forest Area, and this community is enrolled in a donor financed project on participatory sustainable forest management. The se o d village, la eled the Eu alyptus village , is situated in Palanxay District. Part of its land has been allocated to a concession of the Birla Company for the establishment of Eucalyptus plantations. The third village, la eled the “ugar a e village , is also situated in Palanxay District. Part of its forest land has been allocated to a concession of the Savan sugar company for the establishment of sugarcane plantations.

Compiling the 2005 census data available for our case study communities with demographic data from household surveys, we note that the Sugarcane village (at 1,139 individuals) has a population size around twice that of the Eucalyptus (529) or PSFMPSFM villages (689). In terms of household composition, however, the two concession villages indicate mean household sizes and mean numbers of adult labor to be roughly ¼ smaller than in the PSFMPSFM village. As a hypothesis to be tested, one potential partial explanation for these differences may relate to the increase rates of migration rates from concession communities that are located closer to the main highway.

25

Table 2: Key Village Demographics

Human Capital

PSFMPSFM village

Eucalyptus village

Sugarcane village

567 689 3.1% 85 8.1 50% 50%

339 529 4.7% 85 6.2 50% 50%

897 1,139 3.9% 179 6.4 54% 46%

12 7.5 4.5

12 5.8 2.9

12 5.9 2.8

A: whole village Population 2005* 2012** % Pop. Growth/yr. *,** No households** Avg. household size** % women* % men* B: interviewed sample No hh sampled Avg. household size Avg. adult labor/household *) National Census 2005 **) Village headman's registry

Interestingly, average annual population growth rates in the three villages appear to indicate an inverse relationship with household size. Whereas the PSFMPSFM village has the largest mean household size, population growth rates appear to be the lowest (3.1%/yr.), while the village with the s allest ea household size Eu al ptus appea s to ha e the highest g o th ate (4.7%/yr.). All of these growth rates appear to be well above both the national/provincial average growth rates, both 2.1% (IUCN and NERI 2011). While methodological differences between a e sus a d illage leade s ua tifi atio s a a tifi iall i ease the esti ates of u e t populations, local stakeholders do indicate that the Eucalyptus village, in particular, has received a significant amount of in-migration from surrounding communities (discussed further below). This does not seem to explain the relatively hi population growth rate in the Sugarcane village, however. Between the three case study villages, the 2005 census suggests that the largest proportion of adult labor migrants (relative to available labor) left the Sugarcane concession village (roughly 20% of the total labor force). This is the village in which almost all land not presently used for smallscale agriculture has been taken up by the concession, leaving little forest for collection of NTFPs and livestock grazing. These data are likely underestimates of the current situation as the impacts of forest conversions for concessions have become more significant over the last decade. Table 3: Estimates of labor emigration to Thailand (Source: 2005 Census)

Village No persons labor force SUFORD 567 383 Sugarcane 897 597 Eucalyptus 399 248

% migrants % of labor migrated 67% 30 8% 67% 120 20% 62% 10 4%

26

3.2 Methodology A

Data collection methods

This study employs an integrated socio-ecological systems approach to analyzing the impacts of different land and forest policy decisions on both local livelihood resilience and on the supply of ecosystem services that these rely upon. However, as dry dipterocarp forests (and the livelihoods that are derived from them) are among the least well understood or studied among forest types in Lao PDR, only a limited amount of secondary data was available to conduct an economic valuation of the costs and benefits of different land uses. To augment the available secondary data, some primary data were collected and stakeholder consultations were held.

B

Stakeholder consultations and workshops:

The project leader visited Lao PDR several times in 2011 i the p o ess of fo alizi g CIFO‘ s collaboration with the Department of Forestry (DoF) and the Forestry Research Center (FRC), during which time a series of consultations were held with key technical and policy experts. The project partnership held workshops with key national, provincial, and district stakeholders representing a range of institutions at the DoF headquarters in Vientiane (Mar.23rd, 2012), and the Provincial headquarters of the PAFO-Forestry section in Savannakhet (Mar.30th, 2012). During these workshops, the objectives of the research were shared and a range of valuable inputs provided made by stakeholders, including the PAFO selection of the key case study sites to be analyze by the ForCC project team. A detailed presentation of preliminary research conclusions was shared during to the FIP scoping mission (Sept.6th, 2012). Following a more complete analysis of the data, research conclusions were shared with district and province level stakeholders in a workshop held in Savannakhet (Nov.20th, 2012). This draft report is being shared for feedback from key stakeholders, and a national stakeholder feedback workshop will follow in early 2013.

C

Village level data collection

The respective District Agricultural and Forestry Office (DAFO) had notified the communities of the i te ded date a d ti e of the esea h tea s a i al, a d e uested that the asse le the illage s for a day-long meeting and data collection activity. The research teams were accompanied in the village by representatives of PAFO and DAFO. This data collection was conducted in April, when there a e elati el fe de a ds o stakeholde s ti e fo ag i ultu al activities; however the turnout differed significantly between villages. Whereas the PSFM village turned out with over 80 people split roughly between men and women, in Eucalyptus village roughly 26 women and 12 men attended, and in the Sugarcane village 12 each of men and women attended. In each village the total data collection took place over the course of 2 days. The primary modes of village-level data collection revolved around household socio-economic surveys and participatory rapid appraisal methods conducted in focus groups. With the assistance of the village leadership, all stakeholders present were allocated to one of three wealth categories (poorer, middle, and wealthier). It should be mentioned that village leaders did not always agree with the premise of there being three defined classes per se, but each village leadership eventually made the distinction based on a general perception of wellbeing. Each wealth group was subdivided into gendered groups, who then participated in the PRA focus group discussions. 27

D

Household Socio-Economic Survey

As the first groups completed their focus group activities, team members asked randomly selected individuals from each wealth grouping to participate in the household socio-economic survey. Where possi le, t o ea h f o the e s a d o e s g oups were invited to participate, but in the case of the Eucalyptus village, all surveys were completed by women. In total, four (4) household surveys were completed for each of three wealth groups in each of three villages. The household survey covered a number of thematic areas, including demographics, livelihood activities, sources of income, assets and food security.

3.3 Data Analyses A

Economic valuation for non-cash ecosystem services

All households were asked to detail their sources of income, both for cash and non-cash sources. The valuation of ecosystem provisioning services that are purchased or sold by households is straightforward. In addition, households consume large amounts of farm and forest products themselves, or may borrow and lend these to/from others (particularly in the case of rice). The values of farm and forest products were estimated by multiplying estimates of quantities produced with the local market price for the product (direct market pricing) , as assessed by the team locally (see summary provided in Annex 4). Income values per category are expressed in KIP, and converted to US dollars based on the overall mean conversion rate application during 2012 (1 USD = 8000 KIP).

B

Contributions of Livestock Assets to Resilience

One means of representing the relative contributions of livestock to the resilience of households to shocks is to compare the value of livestock owned to the total annual cash income of each household. We regard this measure as more relevant than the absolute cash value of the livestock themselves, as this ratio elates the a ket alue of attle di e tl to the household s relative cash vs. non-cash income requirement (and could therefore be regarded as a proxy for sensitivity to market pressures).

C

Household food security and nutritional diversity

As part of the household socio-economic survey, an attempt was made to estimate the adequacy of rice within the household. This included a quantification of rice grown, bought, sold, consumed and borrowed over the course of the year. The total per capita consumption within each household was compared with the government guideline indicating that each individual requires at least 350kg of rice per year (FAO/WFP 2011). Mention should be made that with respect to the rice, the purchase and sale prices differ: 

 

Rice is usually sold after harvest (Nov-Dec) in the non- illed state alled khao peuak at a rate of KIP2000/kg. Rice is usually purchased during the lean months (Apr-Jun) prior to the next harvest in milled form alled khao san at a rate of KIP5000/kg. In terms of nutritional value, 1 kg of khao san is equal roughly to 1.5 kg of khao peuak.

In addition to self-sufficiency in the staple (rice), we compare the relative amounts of NTFPs consumed and purchased by households – giving an indication of nutritional diversity.

28

D

Caveats regarding intra-household differentiation in vulnerability

This household survey was implemented in order to attempt to represent some of the differences in livelihoods, and reliance on different ecosystem services as well as other sources of income and expenditure, between different types of households and villages. The limited household sample within each village wealth category (N=4) and non-randomized sampling design means that the results cannot be regarded as statistically representative these groups, and should be interpreted in conjunction with other data sets available. Wherever the survey results appear to indicate the presence of significant trends and differences, these should be taken as hypotheses and points of departure for future, more detailed, and necessary research investments. Analyses of differences spanning villages or wealth groups similarly cannot be regarded as statistically representative of their class or whole village, however their larger sample size (N=12) suggests that where trends appear significant, that these results can be given more credence. Additionally, a certain amount of credibility regarding the larger trends is provided by methodological triangulation using stakeholder interviews, limited participant observation, participatory focus group discussions, and particularly through validation of major trends during village stakeholder feedback sessions.

E

Focus Group discussions

The focus group discussions followed an abbreviated version of the PROFOR PRA tool kit. Focus groups from each wealth and gender category were asked to estimate key sources of cash and non-cash household income by ranking, and to estimate how much of each resource came from the different land use types (paddy, DDF, mixed deciduous forest). These data highlight the variety of non-cash income sources, especially those derived from the dry forest. This exercise did result in a significant divergence from the results from the household surveys and the focus groups. A summary of these differences is discussed in the report. These differences cannot be explained adequately based on the data available to us, and in this report these results are primarily used to qualify or support the conclusions from the socio-economic survey. In addition, pa ti ipato o u it esou es appi g e e ises e e o du ted ith e s a d o e s stakeholde g oups p o idi g esea h tea e e s ith a ette o te tual understanding of how different sources of cash and non-cash income are distributed spatially across the landscape. These exercises also highlighted significant differences in the amount of different types of local environmental knowledge.

F

Economic valuations of ecosystem services

Most economic valuations and cost-benefit analyses of ecosystem services represent the values in terms of a Total Economic Value (TEV). This is not practical for the current case study for a number of reasons, most of which relate to the high level of uncertainty regarding the accuracy of current datasets and the range of land use change and climate change scenarios projected for the future. Additionally, TEV estimates make more sense when the different land use options represent different costs and benefits to be calculated at different time horizons. Based on the data to be presented, this does not appear to make much sense. Instead, all data has been presented in terms of the value of annual benefits accruing (or potentially accruing) to local stakeholders. Some methodological reservations are discussed further. One of the main methodological challenges identified in this study is that of collecting reliable consumption, use and sale data for NTFPs gathered by the communities. While an attempt has been 29

made to do so through the survey, the focus group data indicate that this data may significantly underestimate these values. The team attempted to get first-hand data on cost and production from the respective concession companies involved in each village, but neither company was willing to be interviewed or share information with the team. Therefore, all analysis regarding the ecosystem services comes from secondary data. Regardless, this data would not have been very representative of what these sectors could achieve over time as both concessions were only just established a few years ago. Therefore the o pa ati e a al sis of e os ste se i es et ee diffe e t la d use t pes should e ega ded as illustrative rather than definitive. Both the household and most of the secondary data sources are based on data collected in single snapshots of time or single locations. Where data is collected in a single location, the valuation may be skewed by any range of variables specific to that location but which may not be representative of the mean (i.e. poorer than average soils and drainage, stronger than average local leadership, worse than average company-village relationships, etc.). Where the data is collected over a single season, year or even growing cycle (i.e. 3yrs for sugarcane), the benefits reaped from the collection of NTFPs, or the returns on investment may be similarly be skewed by a range of short-term variables: unseasonably warm or cool weather, flash floods, delays in availability of fertilizer, high incidence of insect pests or fungal infections, etc.).

G

Climate change scenario modeling of changes in ecosystem services

This project also conducted a detailed analysis of past climate trends and a modeling of the likely impacts of future climate change scenarios on a range of ecosystem services at both near-term (2030) and longer-term (2080) time horizons. The specific methods used for analyzing climate change scenarios and the impacts on ecosystem services are described in detail in a separate report entitled A a alysis of li ate ha ge s e arios a d their i pa ts o e osyste ser i es i “a a akhet, Lao PDR (Rafanoharana et al 2012). In this report we limit ourselves to a discussion of the main findings from that analysis.

4 Climate change in the region This summary of impacts from climate change is drawn from a detailed analysis which has been produced as a separate report entitled, A a alysis of li ate ha ge s e arios a d their i pa ts o ecosystem services in Savannakhet, Lao PDR (Rafanoharana et al 2012).

4.1 Historical and future projected climate conditions Overall, the climate in Laos can be described as tropical monsoonal with a monthly mean temperature above 18 degrees Centigrade in every month of the year and distinct wet and dry seasons. The mean annual temperature distribution suggests a strong correlation with elevation gradients (see figure below). Consequently, the more mountainous Northern region and to a lesser degree, the Annamite Range7 alo g Laos s Easte a d “outhe o de s ep ese t a eas ith lo e ea a ual

7

Xai Phou Luang (

ລວ ) in Lao

30

temperatures. The Annamite Range and Central and Southern Laos in general receive the greatest amount of rainfall.

Figure 5: Current mean annual temperature (left) and precipitation (right) in Lao PDR (star: Savannakhet).

Historical rainfall patterns and temperature patterns in Savannakhet Province illustrate largely unimodal distributions (see figure below). The highest rainfall occurs during the Southwest monsoon rainy season (May to October), and reaches peak levels when in August. The average annual precipitation in Savannakhet is 2061.2 mm/year, with a monthly range between 21.0 mm (February) and 384.0 mm (August). The mean temperature in Savannakhet is 25.8 degC with a range between 21.5 degC (January) and 28.3 degC (May).

Figure 6: Current monthly mean temperature (line) and precipitation (bar) in Savannakhet (Rafanoharana et al 2012)

31

Based on this historical baseline of data, the team modeled 16 climate change scenarios (combining four Global Climate Models and four SRES emissions scenarios) to assess the likely impacts of climate change on Laos in general, and Savannakhet Province in particular over the 2030 and 2080 time horizons. Reflecting national trends, from a mean annual baseline temperature of 25.8 degC, by 2030, the mean annual temperature in Savannakhet is expected to increase with a range of 0.60 degC (PCM.B1) and 1.39 degC (HAD3.A1FI). Similarly, by 2080, the mean annual temperature increase is expected to range between 1.12 degC (PCM.B1) and 4.70 degC (HAD3.A1FI).

4.2 Projected impacts of climate change on ecosystem services Due to relatively stable precipitation and rising mean temperatures, all climate change scenarios project an increase in net primary productivity (vegetation growth) in Savannakhet Province. It should be noted that the proportional impact on growth of rice in paddies is likely to be more significant than that of DDF forests, however, as growth in the latter is primarily limited by soil depths. The expected impact on paddy field rice production is a net increase in primary productivity of 1.4-12.2% by 2030, or 14-53% by 2080. These findings are consistent with other studies (Lefroy et al 2010; Jintraet et al, 2012.). Looking at seasonal rainfall patterns in the longer term projections the climate change scenarios do indicate significant uncertainty regarding the amount of rainfall and runoff during the month of June. This is a critical period in the agricultural production calendar, when farmers transfer rice plants from nurseries to paddies, and when predictability of water levels is particularly important. Crop failure is a risk whenever there are major interruptions in rainfall, or should there be a sudden overabundance. Based on a modeling of Holdridge Ecological Life Zones8, the overall increase in temperatures is likely to contribute to a general qualitative shift in vegetative composition toward plant species more adapted to drier conditions. This should not, however be confused with a transition toward Dry Dipterocarp Forests, as DDF trees are more accurately characteristic of specific soil conditions than they are of temperature per se. Very little data are available on the impact of climate change on livestock productivity in DDF forests. Livestock graze on natural Arundinaria grasslands, which are resistant to drought but provide low productivity. Preliminary research on replacing the natural grasslands with improved, droughtresistant pasture species seems to be giving promising results as an adaptation strategy to climate change (Hacker et al, 1998). However, little such research has been conducted in Laos, and none in the dry forest contexts. Therefore, much remains to be understood regarding how such technologies can be implemented by district extension agents or communities.

4.3 Projected impacts of land-use change on ecosystem services O e all, usi g the Integrated Model to Assess the Global Environment (IMAGE 2.2) , this stud fou d that policy and economic drivers of Land Use Change (LUC) are likely to significantly outweigh any impacts from climate change on forest productivity. Indeed, when climate and land-use scenarios are combined, forest production is expected to decrease by 2030 under every scenario modeled (ranging 8

First developed by Holdridge in 1947, this is a simple bio-climatic model that describes suitability of vegetative classes based on precipitation and temperature. These “life zones” are presented as “humidity provinces”. Based on this model, any alterations in distributions of rainfall and temperature under future climate change scenarios can be expected to result in qualitative shifts from one life zone to another.

32

between 8-21%). This would correspond with the general perception among Lao stakeholders at all levels that land-use change is the key challenge to be addressed in addressing deforestation and degradation of DDFs.

Figure 7. Extreme scenarios of impacts on NPP for climate and land-use change projections (2030, 2080)9

Several (but not all) scenarios of the policy and economic drivers of land-use change project an inflection point around the middle of the next century, resulting in a gradual recovery in forest production levels but provide limited guidance due to the range of forest productivity projections in the long term (2080). When policy/economic and climatic drivers of forest productivity are combined the range of outcomes ranges between a net loss of productivity up to 51% and net increase of productivity of 19% (in relation to historical baseline levels). Overall, the message would seem to be that while climate change may improve rice production, policy makers cannot expect the same easy wins with regard to forest conservation as policy and economic drivers have to be addressed in order for any climatic drivers of growth to become felt.

5 Livelihood vulnerability analysis 5.1 Primary sources of cash income Overall, household cash incomes in the concession villages were more than double that of the nonconcession village, the PSFM village (see table below). Table 4: Household Cash Income in three Dry Forest Villages, Savannakhet, 2012

Household Cash Income SUFORD village (kip) Eucalyptus village (kip) Sugarcane village (kip) average (kip)

poor 1,041,750 3,842,500 4,302,000 3,062,083

middle 2,062,250 7,037,500 5,343,500 4,814,417

rich 8,430,000 13,476,500 10,157,500 10,688,000

average 3,873,833 8,118,833 8,077,000 6,689,889

Across all three villages, the three main sources of cash income as calculated from household surveys, are labor (45%), livestock sales (32%) and remittances (11%) from household members working elsewhere (see figure 6 below). Together, these top three cash income sources account for 90% of household cash income.

9

Change in NPP compared to the baseline (B) for two contrasting climate scenarios (CC1 - HAD3.B2A and CC2 - CSIRO2.B1A in 2030; CC1 - CGCM2.A2A and CC2 - CSIRO.A1A in 2080) and two contrasting land-use change scenarios (LUC1 - IMAGE.B2 and LUC2 - IMAGE.B1 in 2030) in 2030 (left) and (LUC1 - IMAGE.A2 and LUC2 - IMAGE.B1) in 2080 (right) in areas with natural vegetation

33

Houshold Cash Income in three villages 4,500,000 4,000,000

cash income (kip)

3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 500,000 Labor

PSFM

Eucalyptus

Sugarcane

917,500

24%

4,025,000

50%

4,206,250

52%

2,306,250

60%

2,861,667

35%

1,362,500

17%

Remittances

166,667

4%

752,500

9%

1,625,000

20%

Forest Products

291,750

8%

304,667

4%

469,083

6%

Agriculture

166,667

4%

75,000

1%

247,500

3%

Others

25,000

1%

100,000

1%

166,667

2%

Livestock

Figure 8: Overview of household cash income across three case studies

The greatest difference between the three villages in terms of cash income is that average incomes in the two concession villages were remarkably close (between $1015-1029), whereas average household cash incomes in the PSFM village were below half that amount ($484) (see table below). While cash incomes in the concession villages may well be higher than the PSFM village due to proximity to markets and labor opportunities, this apparent difference may also partly be due to the difficulty of (and potential to under-estimating the volumes and values NTFP harvesting, consumption, and marketing). This is supported by the focus group data that indicated a much greater perceived contribution from NTFPs than did the survey. A second major difference between the PSFMPSFM and concession village economies is that, whereas livestock were estimated to be the predominant source of cash income in the PSFM village (60%), labor was clearly the most important source of income for the concession villages (Eucalyptus: 50%, Sugarcane: 51%). Additionally, it should be reiterated that the still-significant income from livestock sales in the concession villages appears to represent the selling off of livestock reserves due to a shortage of forests needed to provide forage, rather than the annual sale of surplus livestock. Therefore, this source of income is expected to be rapidly reduced in the coming years. Both livestock and forest products can be considered as ecosystem values derived from forest land. Across all three villages, at present they provide around 38% of household cash income, ranging from 68% in villages with intact forests, to 23-39% in villages where significant areas have been lost to concessions. The third major point has to be to reiterate the importance of labor in providing cash income to concession village households in particular. From stakeholder interviews, the sale of labor is 34

increasingly replacing the sale of NTFPs as a source for cash required to make everyday purchases. Unfortunately, stakeholders complain that whereas NTFPs used to be relatively plentiful and could be sourced throughout the year, opportunities for labor are severely limited in the area, and may not be available when most needed. Across the three communities, income from forest products appears to be very limited (6%). However, in addition to the fact that NTFP sales provide more money to middle-class and wealthy households, reflecting their closer proximity to major markets, average cash income from NTFPs remains greater in concession villages than in the PSFM village. This would run counter to the neo-Malthusian arguments describing the roles of NTFP consumption and sales as being associated with perpetuation of cycles of poverty. In Lao PDR, this study, which is corroborated by numerous other studies, indicates that the consumption of NTFPs is clearly an integral part of rural economic strategies, and arguably cultures, irrespective of wealth distinctions (Foppes 2011, WFP 2007). Given the significant export of NTFPs from this part of the country to Thailand and Vietnam), this observation can plausibly be extended to neighboring countries.

5.2 Household Non-Cash Income Whereas total cash incomes are higher in concession villages, the reverse is seen in the case of noncash income, which is defined as all products produced on farm or collected from the wild, that are consumed within the household. These are calculated based on replacement values at local market prices (see figure 7 below).

replacement value (kip)

Household Non-Cash Income in three villages 5,000,000 4,000,000 3,000,000 2,000,000 1,000,000 -

PSFM

Eucalyptus

Sugarcane

Rice consumption 4,555,833

62%

3,432,500

53%

2,720,417

55%

Other crops

29,167

0%

201,250

3%

167,917

3%

NTFPs

1,678,833

23%

1,642,875

25%

1,090,625

22%

Timber

784,000

11%

784,000

12%

784,000

16%

Firewood

300,000

4%

417,500

6%

157,500

3%

Figure 9: Overview of household non-cash income across three case studies

The largest source of non-cash income across all three villages is from rice consumption, representing 57% of all non-cash income generated, and the PSFM village households consumes more rice on average than the other villages. In addition, large amounts of NTFPs are consumed in all three villages,

35

representing on average 23% in the PSFM and Eucalyptus concession villages. Thirdly, timber for house building contributes on average 13% to non-cash income per household per year.10 In terms of forest non-cash contributions to different wealth categories, all household classes earned roughly similar non-cash income from forest resources (NTFPs, timber and fuel wood) with a cash equivalent value ranging roughly between $185-300. The great outlier is that of the middle class households in the Eucalyptus village that receive the equivalent of around $530/yr. In terms of the percent contribution to non-cash income, forest resources appear to be most important to the poor households in the Eucalyptus village, while middle and poor households in the PSFM and Eucalyptus villages both gain around half of their non-cash contributions from forest resources, and the rest from own rice production (see figure 8 below). These results suggest that far from being a poverty trap, in those contexts where resources are not limiting, forests provide an important source of nutrients to a wide range of households. Despite the accounts of a shortage of NTFPs in the Eucalyptus village, it would appear that the poorest households (who also own no cattle), have very limited rice production that they can rely on (and/or are forced to sell), and are therefore still highly dependent on forest products to support their livelihoods and food intake.

Annual income from forest products (KIP)

Non-cash (absolute and percent) income from forests by wealth class in three villages 4,500,000 4,000,000 3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 500,000 -

PSFM

% non cash

Sugarcane

% non cash

Eucalyptus

% non cash

Poor

1,832,500

47%

2,381,625

78%

1,856,500

40%

Middle

1,637,750

51%

2,157,625

37%

4,239,125

61%

Wealthy

1,637,750

18%

1,512,750

23%

2,437,500

30%

5.3 Contributions of NTFPs from Focus Groups In addition to household surveys, this study engaged wealth category and gender segregated stakeholder focus groups in participatory ranking exercises of the importance of different sources of

10

It must be noted that due to the infrequency of house construction (on avg. every 7yrs) and the limited numbers of households sampled, this data could not be collected in the household survey, and was rather based on key stakeholder interview. All households are assumed to construct houses on roughly the same periodicity, thereby gaining the same non-cash value.

36

income. All groups attributed a significantly higher importance to the contributions of NTFPs to household cash and non-cash resources than they did in the individual household surveys. The summary tables from the gender and economic class-segregated focus group discussions in one community (the PSFM village) are shown below. All gender and wealth groups indicated that forest products provide them with at least half of their cash income, though women attributed slightly more to NTFPs than men. The largest contribution was from the sale of frogs and snails, followed by mushrooms and insect products. Table 5: NTFP contributions to total household cash income (Source: PSFM village PRA Discussions)

Cash Income Ban Alang A Forest Products 1 Mushrooms 2 Frogs and snails 3 Insect Products 4 Bamboo and rattan shoots 5 Wild vegetables 6 Kisi damar resin 7 herbal medicines 8 Timber for house building

Poor 54% 11% 14% 11% 6% 2% 6% 1% 3%

Middle Wealthy 56% 54% 9% 11% 20% 14% 9% 11% 9% 6% 4% 2% 5% 6% 0% 1% 0% 3%

All 55% 10% 16% 10% 7% 3% 6% 1% 2%

Men Women 51% 57% 10% 12% 11% 18% 10% 11% 4% 7% 0% 4% 9% 2% 3% 0% 3% 3%

As is reflected in the importance attributed to forest products in the household data, participants indicated that forests contribute on average 52% to total non-cash household income, with the poorer households attributing a significantly greater proportion (67%) than do the wealthier households (36%). Firewood, bamboo/rattan shoots, frogs and snails were ranked highest by all household groups, while mushrooms and insect products appear most important to poorer households. Women and men do not seem to value these resources very differently. Table 6: NTFP contributions total household non-cash income (Source: PSFM village PRA Discussions)

A 1 2 3 4 5 6 7

Non-Cash Income Forest Products Mushrooms Frogs and snails Insect Products Bamboo and rattan shoots Wild vegetables Firewood Timber for house building

Poor Middle Wealthy 67% 53% 36% 8% 5% 3% 9% 8% 8% 8% 4% 0% 12% 8% 10% 7% 5% 7% 11% 14% 8% 12% 11% 0%

All 52% 5% 8% 4% 10% 6% 11% 7%

Men Women 56% 47% 7% 4% 9% 8% 5% 3% 11% 9% 8% 4% 10% 12% 7% 8%

5.4 Household Expenditures An overview of household expenditures highlights some commonalities and differences between the three community case studies and wealth categories. On average, poorer households spend around 73% of their cash income (and on the purchase of food, with little left for other uses, while middle income and wealthy households spend between 30-39% of their cash income on food and spend more 37

Expenditures (Kip x 1m.)

significantly on household, production, social, and health costs in particular (see figure 9 below). The marginal condition of poorer households is underscored by their average low availability of cash income per year (4.3m Kip or $541), and their overall greater absolute expenditures on the purchase of food suggests a clear lack of self-sufficiency in food production from either agricultural or forest sources.

Household expenditures (absolute and proportional) by wealth class 4 3 2 1 -

Poor

% of total

Middle

% of total

Wealthy

% of total

3,171,667

73%

2,648,462

30%

2,409,808

39%

Household

401,583

9%

1,750,769

20%

1,196,615

19%

Production

91,667

2%

1,585,385

18%

939,231

15%

Social

326,667

8%

1,615,385

19%

638,462

10%

Health

291,667

7%

707,692

8%

835,385

13%

Schooling

41,667

1%

400,769

5%

234,615

4%

Food

Food

Household

Production

Social

Health

Schooling

Figure 10: Household expenditures by wealth group in three villages

PSFM and Sugarcane village households appear to spend the greatest proportion of their financial resources (both roughly 45%) on the purchase of food. The average Eucalyptus village household spends a little less (38%) of its cash on food but appears spend more on household and productionactivity related expenditures (see figure below). Whereas the PSFM community spends significantly more than the others on educational expenses, the sugarcane village appears to spend the most on social events11.

11

Note – this is not skewed by any marriages, as none took place this last year in any household sampled

38

Household expenditures (absolute and proportional) by village Expenditures (kip x 1m.)

4 3 2 1 -

Village

PSFM

2,619,375

45%

2,354,167

36%

3,677,917

45%

Household

845,417

14%

1,498,000

23%

1,251,167

15%

Production

572,500

10%

1,442,500

22%

811,667

10%

Food

Eucalyptus

Sugarcane

Social

441,667

8%

575,000

9%

1,751,667

21%

Health

1,058,333

18%

350,000

5%

555,000

7%

334,167

6%

262,500

4%

133,333

2%

Schooling

Food

Household

Production

Social

Health

Schooling

Figure 11: Household expenditures by village

5.5 Balancing household cash income and expenditures A review of the total household cash income and expenditures across the three villages and wealth groups is provided in the table below. Overall, based on the household surveys, it would appear that the average household in the PSFM and Sugarcane village earns less than it spends in cash, as do the poorer and middle class households. Table 7: Cash incomes and expenditures in three villages and wealth groups Ba l a nce between a vera ge Income/Expend. Ca s h Income Expendi tures Ba l a nce US$ (yr 2012) $

hous ehol d ca s h i ncome a nd expendi tures over three vi l l a ges PSFM Eucalyptus Sugarcane 3,873,833 8,118,833 8,077,000 5,871,458 6,482,167 8,180,750 (1,997,625) 1,636,667 (103,750) (250) $ 205 $ (13)

Average 6,689,889 $ 836 6,844,792 $ 856 (154,903) $ (19) $ (19)

Ba l a nce between hous ehol d ca s h i ncome a nd expendi tures between wea l th groups i n three vi l l a ges Income/Expend. Poor Medium Rich Average Ca s h Income 3,091,250 5,806,538 9,865,846 6,689,889 Expendi tures 4,324,917 8,708,462 6,254,115 6,844,792 Ba l a nce (1,233,667) (2,901,923) 3,611,731 (154,903) $ $ (154) $ (363) 451.47 $ (19)

% 836 856 (19)

However, as is demonstrated in Annex 4, there are large differences between the indicated contributions to income from various sources depending on the methodology used. The cash contributions of NTFPs were particularly divergent in the PSFM and Sugarcane villages. Additionally, there appeared to be significant divergences in the contributions of income from labor in middle-

39

wealthier households of the Eucalyptus village and across all households of the Sugarcane village. What these two sources of income have in common is that they may pose challenges for stakeholders to recall due to irregularity in timing and income derived. In addition, if any households engage in illegal logging, it is unlikely that they will report this to us. Unfortunately, there are not many studies that have tried to compare such methodologies, and in many household surveys of income, no attempts are made to quantify the values of forest products that are consumed/used rather than sold. This is an area that clearly requires further research.

5.6 Contribution of assets to resilience: Land The three case study villages represent quite different land use and forest endowment contexts. The PSFM village is relatively isolated from major markets and roads, has three times the land area of the other two villages, and retains a significant proportion of its land (62%) in forest cover (see figure below). In this community, each individual has a larger mean area under cultivation (for rice), as well as la ge a ou ts of chap chong din la d that has been reserved by individual households for future expansion) of paddy. In contrast, the two concession villages are located close to the major East-West highway that connects this province directly with the Provincial capital (2.5-3hr away) as well as Vietnamese and Thai markets. Households in these villages have much smaller land holdings (both in terms of paddy and chap chong din). The eucalyptus village retains roughly 62% of its village lands as communal forest and 14% of the land under chap chong din forest cover. In the Sugarcane village, only the 35% of the forest cover remains and most of this in under chap chong din (and is therefore likely to be converted in the future). This study did not allow for a complete analysis of all household assets, however the availability of paddy for rice cultivation, as well as the household claims to land for future agricultural expansion and additional communal forests allow for some general comparisons to be made. The figure below indicates the average amount of paddy available to each household.

40

25

3000

20

2500 2000

15

1500 10

1000

5 0

500 PSFM

Eucalyptus

Sugarcane

Paddy (ha/hh)

3.26

2.04

2.73

Reserved land(ha/hh)

7.43

1.23

1.23

Concession land lost (ha/hh)

0.00

3.69

3.61

Communal Forest(ha/hh)

19.89

5.36

0.28

2709.00

737.05

766.61

Total Village land (ha)

VillageArea (ha)

Area/HH (ha)

Village area per land use category

0

Figure 12: Total land area available per household in three case study communities

The PSFM village households have both the largest average area of paddy under cultivation (3.26ha) as well as the largest amount of forest land reserved as chap chong din (7.43ha). In contrast, the average Eucalyptus village household retains only 2.04ha of paddy and 1.23ha of chap chong din. The Sugarcane village households similarly own an average of 2.73ha of paddy and 1.23ha of reserve forest, although in this case most of this forest has actually been claimed as chap chong din, and can be expected to be converted soon.

5.7 Contribution of assets to resilience: Livestock Livestock assets are the main source of savings for rural households in Lao PDR. Livestock can be converted to cash by selling in times of need (such as in the case of crop failures due to flooding, or severe illnesses). The value of livestock assets can therefore be compared to annual cash income of households, as a measure of self-sufficiency or resilience in times of crisis. When we compare livestock ownership across the three villages, we note significant differences (see table below). Overall, the PSFM village owns roughly 46%, and 80% more livestock than do the villages associated with the Eucalyptus and Sugarcane plantations, respectively. Interestingly, the middle class across the three villages own relatively similar numbers of livestock (the means of each village are within 12% of the overall mean). In contrast the differences between the poor in the three villages is very great, with the Eucalyptus village owning 200% over the mean and the Sugarcane village own less than 5% of the mean.

41

Table 8: Contribution of livestock assets to household resilience

poor livestock assets

middle

wealthy

average

Kip

$

Kip

$

Kip

$

Kip

$

PSFM village

1,487,500

186

6,878,750

860

33,362,500

4,170

13,909,583

1,739

Eucalyptus village

3,695,000

462

7,721,250

965

17,210,000

2,151

9,542,083

1,193

Sugarcane village

81,250

10

6,066,250

758

17,102,500

2,138

7,750,000

969

average 1,754,583

219

6,888,750

861

22,558,333

2,820

10,400,556

1,300

total HH cash income PSFM village

1,041,750

130

2,062,250

258

8,655,000

1,082

3,873,834

484

Eucalyptus village

3,842,500

480

7,037,500

880

13,476,500

1,685

8,118,833

1,015

Sugarcane village

4,302,000

538

5,343,500

668

10,157,500

1,270

8,077,000

1,010

average 3,062,083

383

4,814,417

602

10,763,000

1,345

6,689,889

836

livestock/income PSFM village

143%

334%

385%

359%

Eucalyptus village

96%

110%

128%

118%

Sugarcane village

2%

114%

168%

96%

80%

186%

227%

191%

PSFM village

1.4

3.3

3.9

3.6

Eucalyptus village

1.0

1.1

1.3

1.2

Sugarcane village

0.0

1.1

1.7

1.0

0.8

1.9

2.3

1.9

average HH resilience (years)

average

One means of representing the relative contributions of livestock in ensuring the resilience of households to shocks is to compare the value of livestock owned to the total annual cash income of each household. In the PSFM village, the value of large livestock was on average comparable to 43 months or 3.6 years of total household cash income. In contrast, in the concession villages, where households have a much greater reliance on cash income overall, livestock assets represent between 12 -14 months of cash income. Notably, while there are certainly disparities between household groups, even the poorest households in the PSFM village own more livestock equivalent of cash income (1.4 yrs.) than the mean wealthy households in the Eucalyptus village (1.3 yrs.), and almost as much as the wealthy households in the Sugarcane village (1.7 yrs.). The poorer households in the Sugarcane village appear to have almost no livestock buffer. While the poorer and middle class households in the Eucalyptus village and the middle class households in the Sugarcane village presently do own reasonable livestock buffers, qualitative interviews with stakeholders indicate that availability of forage and water sources in these communities is a significant concern. At present, the Sugarcane village relies to a significant degree on the permeability of village boundaries to enable their livestock to graze in neighboring village areas as the concession has converted all communal forest lands, and the only forage left in their village is on lands that has been claimed for future development/expansion of paddy (chap chong din). In the case of the Eucalyptus village, insecurity over the potential to lose additional village lands to concessions is resulting in a large-scale privatization of the remaining communal forests into chap chong din plots. As chap chong din plots progressively become fenced and converted for other uses, this source of resilience to these two communities is likely to be rapidly eroded in the coming years.

42

5.8 Household Food Security and Nutrition The dietary staple (glutinous rice) of all three villages comes from rain-fed paddy agriculture, and the nationally mandated annual minimum rice consumption is 350kg per person. An analysis of rice production and consumption indicates a common pattern across all three villages (see table below): wealthier households all have a rice surplus, while most poorer and middle class households live within 10% of the recommended 350 kg/per capita for rice consumption. The wealthier households in the PSFM village seem to just meet their consumption needs, the wealthier households in the two concession communities, appear to enjoy on average between 5970% rice surpluses. Additionally, in the concession communities it would appear that the middle class households fare worse than the poor households in terms of meeting the 350kg/year level of rice consumption. However, food security does not appear to depend on greater access to land alone. Across the three villages, access to paddy varies but is not correlated with wealth. In contrast, rice yields per hectare do appear to be correlated with wealth within each village, though in the case of the Eucalyptus village, middle and wealthy households have roughly the same yields. Based on stakeholder discussions as well as the household interviews, it is clear that wealthier households in the concession villages appear to invest greater amounts of capital on agricultural inputs (both labor and fertilizer). In addition to inter-household differences in yield, there are significant differences in rice yields between the villages. Whereas stakeholder discussions in the concession villages point toward an increasingly widespread use of inorganic fertilizers, attempts by some households to raise yields in the PSFM village appear to mainly use collection and application of natural fertilizers. It should be mentioned here that PSFM households on average consume significantly larger amounts of rice than do households in the other villages (see table 12 below), however larger family sizes (due at least in part to lower migration rates) reduce per capita consumption to well below that of the other villages. Table 9: Food Security - Rice consumption across three communities

Village

Wealth group

poor PSFM middle wealthy poor Eucalyptus middle wealthy poor Sugarcane middle wealthy Average

Area of rice paddy 0.36 0.37 0.17 0.53 0.32 0.79 0.70 0.29 0.38 0.44

Rice consumed Actual Norm Rice produced rice bought consumption paddy rice balance to norm kg/ha kg/capita kg/capita kg/capita kg/capita 1,040 18 319 350 -31 1,290 13 332 350 -18 1,317 11 360 350 10 1,547 9 360 350 10 1,640 7 297 350 -53 1,690 18 558 350 208 2,208 11 322 350 -28 2,333 6 290 350 -60 2,500 18 596 350 246 1,391 15 404 350 54

These disparities, if real, are not however, reflected in the consumption of NTFPs, i.e., primarily mushrooms, rattan/bamboo shoots, insects, frogs and snails. Stakeholders in poorer households appear to consume roughly the same amount (expressed in cash equivalent) of NTFPs across the three

43

villages ($21-33/year), and in both the PSFM and Sugarcane villages, the poor consume vastly more NTFPs than do the wealthier households (see table below). Additionally, whereas the middle class households appear to have the greatest rice deficit, they would appear in at least two villages to consume more than twice the amount of NTFPs as other households. In the Sugarcane village, it would appear that NTFP consumption is negatively associated with wealth. If this is representative of the community, this is an interesting result given that the availability of communal forest land is almost reduce to nil, and given that wealth is closely correlated with claims to chap chong din (forest land claimed for future conversion to paddy). This suggests both that very little of this forest has as yet been converted and that these claims do not at represent a hindrance to othe s a ess fo the olle tio of NTFPs. Table 10: Food Security - NTFP consumption across three communities

Village

PSFM

Eucalyptus

Sugarcane

Average

Wealth group

Area of forest (ha/capita)

Consumption of wild foods (kip/capita)

($/capita)

poor

2.59

198,000

$

24.75

middle

2.56

510,581

$

63.82

wealthy

2.45

40,122

$

5.02

poor

0.81

171,429

$

21.43

middle

1.02

495,262

$

61.91

wealthy

1.08

219,769

$

27.47

poor

0.19

264,568

$

33.07

middle

0.20

181,232

$

22.65

wealthy

0.36

101,667

$

12.71

1.05

238,988

$

29.87

On one hand, these results may raise an interesting question of whether the wealthier households that enjoy greater food security in terms of caloric intake have the same levels of nutrition of the poorer and middle class households. However, while food security is not presently of greatest concern in these communities, we can clearly see cause for concern with regard to both caloric intake and nutritional diversity among poorer and middle class households as natural population growth and loss of forests to concessions and paddy will combine to reduce the supply of NTFPs for all. All three communities express concerns regarding the limited availability of land suited for conversion to paddy as well as to support livestock or NTFP consumption needs. According to stakehodlers living in the concession villages, some seasonal nutritional impacts are already being felt, highlighting how annual estimates of cash incomes frequently mask the seasonality of food and an cash insufficiency. Concession stakeholders indicate that they previously paid for small household purchases through cash raised by the sale of NTFPs, which they could collect as needed throughout the year. Unfortunately, with the increasingly limited per capita amount of forest available and the intermittend availability of labor opportunities available to them now, households indicate greater seasonal food insecurity.

44

In addition, when asked what kinds of food they purchase, most stakeholders indicate that they cannot afford to replace the NTFPs that they had previously consumed on a regular basis through purchases. Instead of the diversity of frogs, snails, fish, insects, small birds and mammals that they would previously have consumed throughout the year, households are increasingly reliant on the purchase of chicken, pork and salted tuna. Furthermore, they indicated that they do not purchase any vegetables to replace those no longer available to them. The primary impacts of land use changes on stakeholder resilience in the selected rice farming communities relate to the loss of forest NTFPs and fodder for livestock. Livestock are an asset that households use to enhance their resilience against shocks. Additionally, farmers in these areas attribute rice production largely to the capacity of dry forests to absorb and regulate flows of water (Fujita, 2000). In most DDF area, rice production is mainly practiced in shallow depressions suffering marked drought/flood regimes that are primarily a function of direct rainfall intensity. There seems to be a possible relationship between household rice consumption deficit and NTFP consumption. It is also important to point out that most households do not have enough cash, even with labor opportunities to match their expenditures. Forests play an important role in the resilience of households to major shocks as well to enable them to overcome periodic shortfalls in cash or food supplies. This resilience was clearly higher in the PSFM area than in the concession areas.

6 The values of Dry Dipterocarp Forests Ecosystem Services It should be indicated here that most cost-benefit analyses calculate and compare the economic values of ecosystem service from different land sues by applying differential discount factors over specific time horizons, i.e. the Total Economic Value (TEV). However, in this case study, the limited data available suggests that neither of the concession models as applied to the DDF context appear to be economically viable at any time horizon. Consequently, there does not appear to be any value to calculating costs and benefits in terms of TEV. Instead, as all benefits to households from forest ecosystem services are harvested on an annual basis, this is how they are represented.

6.1 Valuation of provisioning services The optimal stocking density of livestock in the dry forest context of SE Asia has been estimated by Zola and Fraser (2012), as 5.4ha/head. The PSFM village is the closest measurement we have for a sustainable livestock provisioning service in this context. In the other communities, stakeholders specifically allude to allowing their livestock to graze on the lands of neighboring villages, and cannot be used for an accurate calculation of a sustainable ecosystem service. Based on stakeholder focus group activities and discussions, different land categories in the PSFM village were ranked for their relative importance in providing fodder to livestock. Stakeholders indicate that livestock rely to the greatest extent on forest lands, and that rice paddies only provide nutrients during the two months of the year immediately following the rice harvest. In comparing the amount of forest (both communal and current (chap chong din) and the numbers of livestock registered by village leaders, the respective stocking densities in the all three villages have surpassed what would be regarded as a sustainable level (3.5ha/animal): PSFM (2.48ha/animal) , Eucalyptus (1.49ha/animal, from pre-concession 4.3ha) and Sugarcane (0.43ha/animal).

45

Through a multiplication of the total market value of cattle and buffalos sold in the PSFM village last year (KIP 155,660,131, or roughly $19,458) and the relative contribution of each land use type and area in the village in providing forage12, the overall value of dry forest provision services for fodder come out to roughly $64.60/ha (see table below). It should be noted, that while paddy does provide a significant value in fodder as well, this is cannot replace forest as this supply only extends to a short period of time (1-2 month) immediately following the rice harvest. In this community, stakeholders indicate that the mixed deciduous forests have little forage to offer, and are not accessed very much by livestock. Table 11: Values of livestock per hectare for three land use types in the PSFM village

Mixed Livestock values Dry forest Paddy Deciduous Forest Total livestock sold (kip) 129,197,908 26,462,222 Total livestock assets (kip) 745,201,667 152,631,667 Land area (ha) 1,465 778 277 Value of livestock sales per ha (kip) 88,214 95,490 Value of livestock sales per ha ($) $ 11.03 $ $ 11.94 Value of livestock assets per ha (kip) 508,812 550,778 Value of livestock assets per ha ($) $ 63.60 $ $ 68.85

Total 155,660,131 897,833,333 2,520 61,767 $ 7.72 356,268 $ 44.53

A calculation of the annual estimated sale or market equivalent value of all NTFPs consumed and sold indicates that DDFs provide a higher value ($4.19), almost 2/3rds the value of the paddy based on the value of rice sold and consumed (see table below). Table 12: Values of Annual Non-Timber Forest Products sold from three land-use types in the PSFM village

Non-Timber Forest Values NTFPs consumed (kip) NTFPs sold (kip) Total NTFP value (kip) Total land area (ha) Value NTFPs consumed (kip/ha) Value NTFPs sold (kip/ha) Value of all NTFPs per ha (kip/ha) Value NTFPs consumed ($/ha) Value NTFPs sold ($/ha) Value of all NTFPs ($/ha)

DDF 40,356,997 8,722,062 49,079,059 1,465 27,555 5,955 33,510 $ 3.44 $ 0.74 $ 4.19

MDF 13,004,498 2,190,934 15,195,432 778 16,707 2,815 19,521 $ 2.09 $ 0.35 $ 2.44

Paddy 11,206,040 3,580,684 14,786,724 277 40,437 12,921 53,359 $ 5.05 $ 1.62 $ 6.67

Total 64,567,535 14,493,681 79,061,215 2,520 25,621 5,751 31,372 $ 3.20 $ 0.72 $ 3.92

A limited amount of timber is permitted to be harvested from the PSFM forest for house construction purposes. There are two primary types of timber that are harvested from DDFs for house construction: mai koung (Dipterocarpus tuberculatus) for the main supporting posts upon which the house is elevated, and mai chik (Shorea obtusa) for the floor and wall boards. There are several other common species of Dipterocarpacae, each with its own uses, however given the limited data available on the amounts of each tree species found in these forests, an average value of the two species above 12

Villagers in the PSFM village estimate that for grazing, their cattle depend for ten out of twelve months (83%) on the DDF forest and for two out of twelve months (17%) on the paddy fields in the beginning of the dry season.

46

is used to derive an estimated value of the DDF provision of timber ($6.91/ha). When measured in comparison with the average value of a mixed-deciduous forest ($36-145), this low value is a significant basis for why most foresters assess a very limited value to DDFs (see table below). Table 13: Value of timber resources in three land use types in the PSFM village

Timber values Timber volume (m3/ha) Average timber price ($/m3) Total Timber value ($/ha) Annual value ($/ha) * * Based on 110 year rotation cycle

$ $ $

DDF 19 40.00 760 6.91

MDF 20 -80 $ 200.00 $4000-16,000 $36-145

Paddy 5 $ 20.00 $ 100 $ 1

Total $ $ $

12 246 2,953 26.85

The limited value of DDFs for timber is further reduced due to the poor soil conditions that constrain tree growth. Based on a detailed inventory and calculation of growth rates in DDFs in Savannakhet Province by Pukkala (2005), it is estimated to take over a century for the two above-named species to reach a harvestable width (roughly when they attain a DBH13 of 25cm) (see table below). Table 14: Time needed to grow to harvestable size for three dominant species in dry dipterocarp forests in Savannakhet

Key DDF species Lao Name

Scientific Name

Mai koung

Dipterocarpus tuberculatus

Mai chik Mai sat

Average

Growth

No years to

No years

DBH (cm)

cm/ 5yrs

average DBH

to DBH=25

26

1.2

108

104

Shorea obtusa

28.7

1.1

130

114

Dipterocarpus obtusifolius

23.3

1.04

112

120

26

1

117

113

Average Source: Pukkala, 2005.

From key stakeholder interviews in the PSFM village, a rough estimate was obtained for the consumption of timber for household construction (see table 15). The actual off-take represents a value of $3.75 per ha of forest per year. This is much lower than the value that could be obtained if the forest was harvested in a sustainable way as calculated in table 13 above: $6.91 for DDF forest and $36-145 for MDF forest. This would seem to indicate that present levels of legal village harvesting of timber for house construction are very modest. Table 15: Villager's estimate of annual consumption of timber for house construction in the PSFM village.

Actual Timber Off-take

Values

No Households per year Volume of wood needed/hh (m3) Local price kip/m3 Total value kip/year Total value $ Value $/ha of forest

13

$ $

Diameter at Breast Height

47

7 30 320,000 67,200,000 8,400 3.75

In addition to the harvesting of timbers for construction and furniture making, a significant amount of wood is harvested for firewood. Although communities near the major highway appear to be very actively involved in the production of charcoal (particularly for export to Vietnam), the distance from the PSFM village (that this valuation is based upon) to such markets undoubtedly makes the transportation costs economically unviable (for the present). Based on the market values of firewood reported, the average provisioning service value of firewood from DD and mixed-deciduous forests are $1.09/ha, and $2.05, respectively. Table 16: Values of firewood per hectare from three land use types in the PSFM village

Firewood values DDF Annual value of fuelwood consumption (kip) 12,750,000 Land area (ha) 1,465 Value per ha (kip) 8,706 Value per ha ($) $ 1.09

MDF Paddy Total 12,750,000 25,500,000 778 277 2,520 16,380 10,119 $ 2.05 $ $ 1.26

As with most communities located next to a major forest under government management, it can be assumed that there is some under-reporting of both the amount of timber and firewood harvested.

6.2 Valuation of regulating services In addition to the provisioning services that are provided by the DDF, these forests provide a range of services that are less visible. While these forests are not typically located on very steep slopes, in which the erosion and flood control functions of forests may play a larger role, the loss of DDFs would result in some alteration of base hydrological flow and siltation of existing fields. Therefore a conservative valuation of the watershed regulatory services has been calculated at 25% the value of the annual rice production, resulting in a total DD valuation of $26.31/ha. Table 17: Watershed regulatory values of dry forests, calculated as 25% of total rice production value in the PSFM village

Rice values and erosion values

DDF

MDF

Paddy

Rice consumed in whole village ($) Rice sold in whole village ($) Land area

1,465

$

48,406

$

1,771

778

$ 174.67

$

6.39

Total value of rice per ha ($/ha)

$

181.06

$

4.98

$

13.98

1,771

174.67

Value of rice sold per ha ($/ha) 26.31

$

2,520

$

$

$ 48,406

277

Value of rice consumed ($/ha)

Erosion values (25 % of rice production)

Total

$

6.39

$ 181.06 $

45.27

6.3 Valuation of Co-benefits Finally, we must underscore that in addition to the ecosystem services for which tentative valuations of their contributions to local stakeholder resilience are calculated here, there are a host of additional ecosystem services that DDFs provide to a wider range of stakeholders at various scales. One form of potential co-benefit for which tentative markets do already exist (but whose value as applied to Laos s d dipte o a p fo ests ha e ot a tuall ee ealized i the a ket pla e is that of the international market price for the trade in carbon credits. Based on estimations of carbon content fo “a a akhet s DDFs Pukkala 2005), and a conservative carbon credit of $4.8/ton, over 48

the course of their 110 yr. harvest rotation cycle, DDFs could potentially contribute a net annualized value of $3.55/ha. Table 18: Value of DD forests in terms of potential carbon credit

Carbon values

DDF

carbon content (tons C/ha)

MDF 22.20

carbon content (tons CO2/ha)

54.50

81.46

Carbon value, price= $4.8 /ton CO2 ($/ha)

$

390.99

Total Carbon value whole area (1,465 ha)

$

572,636

carbon asset per year (110 yr cycle, $/ha)

$

3.55

Paddy

Total

6.92

0

200.02

25.40

$

0.05

960.07

$ 121.90

$

133.79

$ 747,320

$ 33,782

$

1,353,738

$

$

$

4.46

$

8.73

1.11

In addition, there are a range of ecosystem services for which estimations of monetary values in this context are either not possible given the data available, and many represent areas for potential future research. There are also some cases where the calculation of a monetary value may not be deemed desirable for moral or cultural reasons. Some of these ecosystem series for which no data is presently available, include:  

       

Provision of drinking water for livestock consumption, Provision of habitat for biodiversity in general, as well as the flora and fauna involved in o Natural pest regulation and barriers to the spread of diseases, o Provisioning of honey and pollination services for many plants, both domesticated and not, Soil micronutrient uptake, Microbial diversity for soil health, Local temperature and climate regulation during heat waves that protect humans, livestock and damage to a range of plants used by humans, Cultural and religious values, Recreational and aesthetic values, Genetic diversity values, Medicinal values, As yet unknown values.

6.4 Total value of Dryland Dipterocarp Forest Ecosystem Services To summarize, ecosystem services provide an immediate and measurable source of annual returns to households through sale and/or replacement values for livestock, NTFPs, firewood, construction materials and domestic water supplies. The value of current estimated annual income in the PSFM village from these provisioning services amounts to roughly $20.66/ha (see table below). In addition, when we add a conservative estimation of the contributions of DD forests to regulation of the water supplies and erosion control for agricultural production, the total value of known, quantifiable ecosystem services is estimated to amount to $46.97/ha. Table 19: Economic values of dry forest ecosystem services as annual returns to the PSFM village

Dry Forest Ecosystem Services Annual returns

kip/ha

49

$/ha

Livestock sales NTFP sales and consumption Timber for house construction Firewood consumption Domestic water use Total value provisioning services Erosion control (25 % of rice production) Total value regulatory services

Actual contribution to HH income Potential from timber sales (110 yr. cycle) Potential annual carbon value Total option values Overall potential value

88,214 33,510 24,000 8,706 10,857 165,288 210,455 210,455

$ $ $ $ $ $ $ $

11.03 4.19 3.00 1.09 1.36 20.66 26.31 26.31

375,743

$

46.97

55,273 28,435 83,708 459,451

$ $ $ $

6.91 3.55 10.46 57.43

In the table above, we also indicate some option values, or co-benefits, as described above that result from the conservation of this resource (potential for sustainable off-take on a 110yr rotation length, and the value of the standing carbon on the global carbon credit trading market).

I additio , ho e e , th ee of these e os ste se i es a e ega ded as assets , that o u ities could potentially use and sell or traded in times of emergency. As such the DDF ecosystem provides an important safety net as discussed above. The total of DDF ecosystem asset sale value in the PSFM village amounts to $1,231 per ha (see table below). Table 20: The value of Dry Forest Ecosystem services as livelihood assets in the PSFM village

Dry Forest Ecosystem Services as Assets

kip/ha

Timber Carbon assets Livestock assets Total ecosystem assets value

6,080,000 3,127,897 508,812 9,716,709

$/ha $ $ $ $

760 391 64 1,215

6.5 Substitution Values of Conversion to Plantations Despite requests for interviews and data sharing with the respective concession managers in Savannakhet, the research team, with the support of the PAFO representative, was unable to access primary data on actual economic viability or growth rate in their concession. In their defense, this may be due to the fact that both concessions were only established recently, and have not yet completed a full growing cycle, either for sugarcane (3yrs), or Eucalyptus (7yrs). Additionally, with regard to the Eucalyptus plantation, the significant amount of ongoing conflicts with the local o u it dis ussed a o e a ha e o t i uted to the o pa s elu ta e to speak ith the team. Therefore the valuation had to be based on other sources than those actually being conducted in the case study villages.

50

According to an assessment conducted of the Asian Development Bank funded LPA project of a smallholder eucalyptus plantation in another DDF context in Savannakhet Province, the projections with regard the to the returns on investment of Eucalyptus were well below the range expected, and did not break even when calculated over a 7 year rotation (LPA,2005). In fact, the actual yield was less than 25% of the lower end of the range of projected yield, and less than 10% of the most optimistic projection. F o the tea s qualitative observations of the Birla Lao concession, it would however, appear that this plantation is likely to achieve reasonable yields, at least in line with the lower end of the ADB-LPA projections. At the same time, it must also be underscored that the community received very limited compensation for the loss of these forest lands and consequent decreases in NTFP and livestock fodder supplied or forgone options for future agricultural expansion.

Table 21: Economic values of Eucalyptus plantations on former Dry Dipterocarp forest soils.

Economics of Eucalyptus planted on Dry Dipterocarp soils Mean annual increment m3/y/ha price $/ton gross income value $/ha/yr total income over 7yrs production costs Net profit over 7 years Net income per year Sources: ADB 2005, LPA 2005.

ADB LPA project Eucalyptus yields Birla Lao foreseen foreseen actual foreseen low high yield 10 24 2.28 ? $ 27.00 $ 27.00 $ 27.00 ? $ 270.00 $ 648.00 $ 61.48 ? $ 1,890.00 $ 4,536.00 $ 430.37 ? $ 493.00 $ 493.00 $ 493.00 ? $ 1,397.00 $ 4,043.00 $ (62.63) ? $ 199.57 $ 577.57 $ (8.95) ?

The calculation of the economic viability of sugarcane in DDF soils comes from a combination of data from two land owners from nearby villages in Phalanxay District who had attempted to cultivate Sugarcane, and from data collected from other sugarcane concessions in Laos and Thailand. The difference between small-scale farmer costs was due to the fact that Farmer B did not pay for labor, a d used f ee fa il la o . ‘ega dless, due to poo g o th ates a d op pests, all th ee Lao examples of sugarcane cultivation on DDF soils appear to be underperforming, and have led the small scale land owners to give up entirely. Based on expert inputs regarding what Thai farmers across the border require in order to reach their eak-e e poi t , the ields ould need to at least double for either Lao farmers or concessions to be able to succeed. Given that Lao farmer A and the Lao concessions invested close to the same amount of money in their production system, it would seem that the either the white leaf pest or the soils in the Savannakhet examples are particularly problematic. Additionally, different sugar cane producers may get significantly different prices for their product.

51

Table 22: Economics of sugar cane farming on former Dry Dipterocarp Forest soils.

Sugarcane values yield history

Lao farmer a* yr 1: 70 t/ha; yrs 2 and 3 failed

avg yield/yr over 3 yrs 23 (tons/ha) price baht/ton 609 gross income baht/ha 14,219 average costs (baht/ha) 56,250 -42,031 net profit (baht/ha) net proft (kip/ha) -11,208,333 -1,401 net profit ($/ha) *source: stakeholder interviews. **source: anonymous sugarcane expert.

Lao farmer b* yr 1: 50 t/ha; yr 2: 25 t/ha; yr 3 failed

Lao concessions** problems with "white leaf" disease

Thailand farmers** "break-even point" around 1100 baht/ton

25

33

59

609 15,234 39,375 -24,141 -6,437,500 -805

1000 33000 64900 -31,900 -8,506,667 -1,063

1200 70800 64900 5,900 1,573,333 197

6.6 Summary of impacts of concessions and PSFM programs on the provision of ecosystem services It must be underscored at the outset, that these economic assessments of the profitability of concessions are tentative and illustrative at best. Lacking access to data from the actual concessions in question, it is difficult to assess their economic feasibility with any accuracy. Based on wide ranges of the projected yields and the actual yields achieved in the case studies, we can however, observe that many concessions on DDF soils are inherently uncertain investments. Detailed information regarding payments for the establishment of concessions is also not available (in part as the payments are negotiated with different levels of government, and none of these are recorded centrally). However, one quoted stakeholde s estimate of $6/ha would suggest that the go e e t s low pricing strategies may in fact contribute to risky investments by investors. With regard to the impacts on ecosystem services and the rural livelihoods that rely upon then, the conclusions from these case studies are quite clear. The concessions in these case studies provide very limited compensation to local stakeholders for services forgone, and provide very limited direct economic contributions to the household economy. Due to the loss of DDFs, households lose a significant source of nutrients and income that they had previously gained from the gathering, consumption and sale of NTFPs. This forces households to seek low-skilled labor elsewhere, in an economic context where such opportunities are limited. Due to the loss of DDFs, households also lose their primary source of savings – livestock – which would otherwise help households in weathering crop failures and other shocks. It is difficult to quantify to what degree the losses of DDFs reduces agricultural production and health through loss of reliable hydrological flows and erosion regulation, but the impacts are certainly seen by stakeholders. Additionally, if concession holders abandon their concessions as economically unviable, the slow growth rates in DDF soils ensure that it would take a long time for DFFs to recover. It should be recalled that the average DDF tree in Savannakhet requires over 100 years to attain a minimum, commercially harvestable, diameter. Therefore, the impacts from concessions should be regarded as relatively

52

irreversible, and the negative long-term impacts on ecosystem services and livelihoods should be reflected in any evaluation of the costs and benefits. In any case, even if land were reverted to the community later on, it would appear highly likely that most of the benefits from the recovery of the landscape for the provision of fodder for livestock would accrue mainly to the wealthier households that were able to maintain limited livestock holdings. This is demonstrated in the Sugarcane village, where despite the loss of local forest fodder supplies, wealthier households have been able to maintain some livestock by grazing the o othe illages lands, whereas the poorer households have sold off almost all their livestock. In contrast, DDFs under the protection of the PSFM project provide stakeholders with high value supplies of provisioning services (ex. food, fodder, timber) and regulatory services (watershed hydrological service protection). In addition, there are the potential values from carbon credits that could help to encourage community support for forest conservation programs. Stakeholders in both the PSFM village and the nearby World Conservation Society – supported Elds Deer Sanctuary also indicate an appreciation of the protections that these programs have provided their communities. Specifically they highlight the visible protected area status as reducing pressures by investors and the government to allow concessions to be established on their lands. In the other communities, local stakeholders were largely passive observers of the concession allocation process as they had not been sensitized to their right (which has a vague basis in law) to deny access to their village lands.

53

Dry Forest Contributions to Resilient Agricultural Livelihoods in Laos: A Policy Brief Forest Ecosystem Services support livelihoods in Lao PDR directly and indirectly 73% of the Lao PDR population lives in rural areas (NSC, 2006), where forests provide important ecosystem services to society and the environment. Forests contribute directly to household food security through direct consumption of Non-Timber Forest Products (NTFPs) as well as providing households with income from products that they can sell for cash (harvestable timber and NTFPs). The total annual value of subsistence and commercial use of NTFPs in the country was estimated to be $510 million per year or 9.7% of the national GDP (Foppes and Samontry, 2010). In addition to direct benefits, forest ecosystems support smallholder agricultural and livestock production systems through regulation of water, erosion and climate. Farmers perceive rice production in rain-fed paddy fields to depend largely on the capacity of forests to absorb and regulate flows of water (Fujita, 2000). Similarly, the significant quantity of large livestock (buffaloes and cows) kept by small-holder farmers in Laos depend mainly on natural vegetation occurring in forest lands as Laos has limited pasture lands (Zola and Frazer, 2012). Dry Dipterocarp Forests are unique but poorly understood Dryland Dipterocarp Forests (DDF) cover 13% of the Lao PDR forest estate, and represent a unique vegetative ecotype spanning parts of SE Asia. DDF are characteristic of shallow soils that are prone to flooding in the rainy season, extreme drought in the dry season and frequent exposure to wildfires (Bunyavejchewin, 1983). They support very slow vegetative growth rates with DDF trees taking roughly 100 years to grow to a commercially harvestable diameter (roughly 25cm) (Pukkala, 2005). Though DDFs provide local communities with a range of ecosystem services, the benefits from DDFs remain poorly documented, and consequently DDFs are undervalued by policy makers. DDFs in Laos also provide habitats for several engendered wildlife species, such as the Elds Deer (Panolia eldii). National policy maker concern with the rapid conversion of DDF land to commercial plantations Large swathes of Lao DDF lands are targeted for conversion to commercial agricultural and forest pla tatio s. This fo s pa t of the Lao atio al go e e t s atte pt to le e age la d esou es to attract foreign investment, which is expected to ultimately lead to greater rural development. National policy makers recognize the need for better data to inform decision-making on how to balance the welfare of local communities with national priorities for attracting foreign investment. Following recommendations of the Department of Forestry, this study assessed contributions of DDF ecosystem services to local livelihoods. By law, communities should be compensated for the loss of economic benefits when companies occupy land for concessions. So far, little compensation has been offered. How would the economic viability of concessions be affected, if such compensation was included? This study provides some data to assist policy makers in better land use planning and in shaping DDF management strategies. This study covers three villages in Savannakhet province: 





P“FM illage : Pa t of the illage la d falls ithi the Do g Kapho NPA a d is a aged u de guideli es de eloped the Pa ti ipato “ustai a le Fo est Ma age e t P“FM Project. No concessions took place here, the remaining communal forest covers 19.89 ha/hh. Eu alyptus illage : 45% of village forest land was converted to Eucalyptus plantations for the Birla Lao company, remaining communal forest area covers 5.36 ha/hh. “ugar a e illage : 75% of village forest land was converted to Sugarcane plantations for the Savan Sugar company, hardly any communal forest area remains (0.28 ha/hh).

Key findings - the economic rationales for conserving Dry Dipterocarp Forests

54

(g) Non-timber forest products are important for rural households Forest resources provide the lion-share of cash income across wealth categories in the PSFM village: NTFPs (55%), livestock sales (26%). Among NTFPs, the most important are frogs and snails (16%), mushrooms (10%), insects (10%), bamboo and rattan (7%). Across all three communities, forest resources (NTFPs and livestock) constitute an average of 35% of total income. When comparing the importance of cash vs. non-cash income in the PSFM village, non-cash income represents roughly 68% of total income, however, non-cash income still represents 35-41% of total income in concession communities. (h) Loss of forests reduces livestock resources which are key to resilience Livestock (water buffalo and cattle) are the primary source of savings and provide a safety mechanism during times of crisis (such as crop failure). Villagers estimate that large livestock depend on natural stands of Arundinaria grass in DDFs for 80 % of their grazing requirement. The value of livestock owned by households in the PSFM village was equivalent to 3.6 years of average annual household cash income, compared with 1.2 years in the Eucalyptus village and 1.0 years in the sugarcane village. Elevated livestock sales in the Eucalyptus village may reflect the loss of forest grazing land and diminished water supplies for livestock are a critical concern. Meanwhile, the poorest households in the Sugarcane village have been forced to sell off almost all livestock. Fodder from forests represents a significant source of resilience to shocks, and stakeholder perceptions of vulnerability and food insecurity are significantly higher where DDF has been converted to concessions. (i) Forests are not a poverty trap but a resilience strategy This study concludes that forest-dependent livelihoods in Laos and reliance on NTFPs in particular a ot e ega ded as po e t t aps . O the o t a , DDF e os ste se i es ep ese t iti al resources that are used relatively equally by all wealth groups and average total incomes between these communities do not differ significantly ($1514-1736/year). According to stakeholders, the primary distinction is that in DDF communities, NTFPs contribute to a diversified diet throughout the year, and they are a ready source of cash when needed. Communities that have lost much of their forested land express concerns over limited access to alternative sources of income to provide cash when needed and indicate that they do not purchase the same amount of vegetables or animal-based protein as what they used to gather and consume from DDFs just a few years ago. Concession communities rely to a greater degree on local and migrant labor for cash income (35-46%) than the PSFM village (9%), where no concession occurred. (j) Conversion of communal forest land to concessions increases land conflicts There are numerous documented cases of conflicts between local stakeholders and concessions due to a combination of poor rural land titling, stakeholder engagement and documentation by concession holders, and an absence of conflict resolution platforms. In the two concession communities studied, local stakeholders received little or no compensation for forest resources lost, and gained little or no employment from the concessions themselves. Furthermore, due to insecurity that additional DDF land may be lost to concessions, most of the remaining, formerly communal forests are undergoing privatization by individual households. This further erodes community access to livestock forages, NTFPs for consumption and timber for house construction. (k) Potential impacts from land use change on hydrological services In focus group discussions, communities express a strong belief that DDFs regulate the flow of water to wells and paddy fields. They also recognize that the forests protect the shallow, often sandy soils from erosion. They cannot quantify the impact of deforestation on both these services, but they are quite concerned having witnessed declining river flows due to previous logging activities. (l) Compensation for losses to livelihoods would improve livelihood and conservation outcomes.

55

According to this study, the value of livestock kept by villagers on DDF land is worth $64 per ha or $1,193 per household per year. According to Prime Minister Decree 192, project developers are required to compensate all affected people for any economic losses that result from their project. It must also be underscored that any measure of economic viability should include the costs of meaningful compensation to local communities for DDF ecosystem services lost. This is not the case now, and concessions are granted at almost token rates by the government on the assumption that these concessions will contribute to rural development. If such compensation were internalized into cost of operation, it is questionable whether many such concessions located on the relatively unproductive DDF lands would be considered economically viable. Key findings on the impacts of climate change and land use change An analysis of past and projected climate trends was conducted over a 50 year period (2030-2080) for Savannakhet Province, comparing 16 global climate change scenarios. Temperatures are uniformly expected to increase: 0.6-1.4 degC by 2030, and 1.1-4.7 degC by 2080. The baseline for average annual rainfall was 2061 mm/year, and projections of future rainfall changes from this baseline range between -60.18 and +78.72 mm/year in 2030, and -202.71 to +219.08 mm/year in 2080. The expected impact of climate change on natural vegetative production is a net increase in primary productivity of 1.4-12.2% by 2030, or 14-53% by 2080. These findings are consistent with studies documenting similar increases in expected agricultural production (Lefroy ea 2010; Jintrawet ea, 2012.). However, these projections are based on annual average temperature and precipitation projections and mask the potentially significant risk of increased variability of rainfall at the start of the agricultural growing cycle. However, an analysis of projections of land use change indicates that any gains in forest growth due to climate change will be far outweighed by negative impacts from the policy and economic drivers of land use change. Taken together, the scenarios for forest productivity in Savannakhet project a uniform net decline in primary productivity (-8.46% to -28.64%) through 2030, but the potential for policies to either somewhat improve or greatly reduce forest production levels by 2080 (NPP: +19.06 to -51.03%). On the whole, the conclusions of this analysis reflect stakeholder perceptions: that challenges posed by on-going transformations of the landscape and livelihoods due to development of concessions far outweigh concerns or possible benefits from climate change. Conclusion - Co essio s that ig ore DDF values are aladaptatio strategies Natural DDF are important for managing and maintaining water soil resources that underpin the entire rural economy. At present, many concessions are granted with little or no regard or compensation for the losses in food, income or resilience to local communities. Meanwhile, the concessions contribute little to household income and contribute to the pressure for households to send their young adults to seek labor opportunities elsewhere (particularly Thailand). Furthermore, poor transparency or documentation in concession allocations and limited availability of conflict resolution and adjudication institutions is increasing conflicts between rural stakeholders and concessions. A general sense of land tenure insecurity is resulting in a rush to privatize remaining forest commons. This combination of processes is resulting in reduced perceptions of wellbeing in the short term and measurable increases in vulnerability over the long-term. As practiced currently, most large-scale DDF fo est o e sio s ust e ega ded as a fo of li ate aladaptatio , i.e. se e el u de i i g local stakeholder resilience to climatic (and other) shocks. From a national development point of view, the establishment of concessions in DDF areas must also be recognized as involving significant levels of uncertainty due to the poor growing conditions. Lacking clear evidence to the contrary, and the uncertain experience of other sugarcane concessions in Laos, it is difficult to assess whether some of these concessions are likely to prove economically profitable.

56

Such viability is further brought into question if the concessions actually compensated stakeholders for the income and savings lost as is required by law. We conclude that the economic rationales at both local and national levels to leave DDF forest land intact clearly outweigh those for conversation unless it can be clearly demonstrated that the economic returns to both national and local stakeholders are warranted. In addition, it must be underscored that given the limiting soil conditions and the reduction in soil quality once DDF soils are cleared of native forest cover, large-scale conversions should be recognized as relatively irreversible. If the concession fails for any reason, this land is unlikely to be of significant value for rice production, and nor will it recover to high-quality forest cover. Comprehensive solutions for stakeholder resilience through Ecosystem based Adaptation For small scale farmers, livestock in DDF represent a key source of resilience to all kinds of potential shocks, and help them compensate for the risks of rain-fed rice farming on drought-prone shallow soils. Access to DDF is also impacted by small-scale DDF conversions, partly motivated by local concerns of losing village land to concessions. This pressure to privatize and convert remaining DDFs would be reduced if local communities could gain tenure security over communal forest land. A more comprehensive approach to enhancing resilience proposed here is one of Ecosystem Based Adaptation, whose specifics might include the following elements: (a) Reducing exposure to climate variability and external shocks  Invest in water storage and utilization infrastructures for irrigation and drinking water;  Restore forest cover to rehabilitate hydrological services to streams and wetlands and reduce impacts of prolonged drought on local climate and biodiversity; (b) Reducing sensitivity to climate variability and external shocks  Reduce dependence on ecosystems services for livelihoods, e.g. by creating alternative employment  Apply integrated village land use planning strategy that sets aside sufficient land for the maintenance of livestock herds, and some expansion of small scale agriculture;  Ensure community access to a range of ecosystems within the DDF landscape, ranging from the highest ridges, most sensitive to drought and erosion, to the lower strata, which have deeper soils and are more resilient to drought and erosion;  Enhance tenure security and reduce the rush to privatize remaining forests through allocation of communal forest rights;  Apply a nation-wide assessment of DDF resource quality, to better understand the ecosystem alues of good , ediu a d poo fo ests a d de elop spe ifi a age e t egi es fo them to permit more targeted land conversion;  Require compensation for ecosystem service benefits lost due to concessions. (c) Options for enhancing adaptive capacity  Adopt more intensive systems for animal feeding to reduce area of land needed for livestock;  Adopt more intensive rice cultivation systems, including more intensified use of organic and inorganic fertilizers; Improve access to education to improve the quality and likelihood of gaining off-farm employment. Key policy implications The main conclusion of this study is that values of forest ecosystem services should be incorporated in all decision making systems around conversion of forest land. More effort is needed to improve land use planning, design drought-resilient livestock and agricultural systems and more effective enforcement of rules on forest protection. National partners could explore ways to include measurable values into land use planning processes, e.g.:

57

   

How many ha of paddy land should be allocated per household or per capita to ensure future food self-sufficiency? How many ha of DDF forest should be preserved per household or per capita to allow for a minimum resilience buffer to shocks? What criteria could be applied to ensure that companies can demonstrate the viability of concessions, compared to existing ecosystem values, before they are granted? Where concessions are allocated land in use by communities, what amount and form of compensation would assist communities in replacing the lost forest benefits?

The Forest Investment Program (FIP) could promote sustainable NTFP and livestock production in production forests and common forest lands through technical extension services. It could address gaps in land tenure such as: sensitization on community rights to forest land, community land titling for common forests and establishment of sub-district level arbitration courts for law enforcement and conflict resolution. Additio all , ate is the ke egulati g se i e. FIP could integrate water management with forest management. Finally, illegal logging cannot be stopped by law enforcement alone. The FIP could contribute to conservation and livelihoods objectives by providing villagers with support and incentives to contribute to law enforcement, e.g. through payment for patrolling, fire management, reforestation and other ecosystem management services.

58

7 Literature Cited ACIAR, (2012), Socio economic theme: Southern Laos Farming and Marketing Systems Project Presentation on Rice-based Systems Research. Regional Technical Workshop June 2012 Australian Centre for International Agricultural Research. Barney, K. and K. Canby (2011), EU FLEGT Facility, Baseline Study 2, Lao PDR: Overview of Forest Governance, Markets and Trade. Berkes, F., and C. Folke, (eds), (1998), Linking Social and Ecological Systems: Management Practices and Social Mechanisms for Building Resilience. Cambridge University Press, New York. Bunyawejchewin, S. 1983. Canopy structure of the dry dipterocarp forest of Thailand. Thai. For. Bull. 14: 1-132. CIDSE (2009), Land concession induced livelihood changes: Research on land and livelihood impacts from land concession from mono-culture tree plantation. International Cooperation for Development and Solidarity (CIDSE-Laos). Cle do , K. , The ‘ole of Fo est Food ‘esou es i Village Li elihood “ ste s: A “tud of Th ee Villages i “ala a P o i e, Lao PD‘, IUCN. Dwyer, M.B. (2011), Building the Politics Machine: Tools for Resolving the Global Land. Paper presented at the International Conference on Global Land Grabbing (6-8 April 2011). EIA (2011), Crossroads: The Illicit Timber Trade Between Laos and Vietnam. ENVIRONMENTAL INVESTIGATION AGENCY (EIA) FAO/WFP (2011), Special Report: FAO/WFP crop and food security assessment mission to Lao PDR. Folke, C., (2002), Social-Ecological Resilience and Behavioural Responses. Beijer Discussion Paper. Stockholm, Sweden, Beijer International Institute of Ecological Economics: 19. Foppes, J. and Samontry, D. (2010), Technical Report: Assessment of the values of Non-Timber Forest Products (NTFP) in Lao PDR. With special reference to National Production Forest Areas (PFAs) supported by the Sustainable Forestry through Rural Development (PSFM) project. Department of Forestry, Ministry of Agriculture and Forestry. Fujita, W. (2000), Fa e s Vie s of the Fo est: Pe eptio s of the Fo est a d the Natu al E i o in Northeast Thailand. Southeast Asian Studies, Vol.38, No 1. June 2000.

e t

Gunderson, L. H., and C. S. Holling (eds.), ( 2002), Panarchy: Understanding Transformation in Human and Natural Systems. Washington, DC: Island Press. Hacker, J.B., Novaha, S, and Phengvichith, V. (1998), Some natural and induced grasslands of the Lao PDR. IN: Sturr, W.W. (1998) Proceesings of the Thirds Regional Meeting of the Forages for Smallholders Project held at the Agency for Livestock Services, East Kalimnatan, Indonesia.., CIAT Working Document No 188. Haughton, J. (2008), Situational analysis of human trafficking in the Lao PDR - With emphasis on Savannakhet. A literature based study undertaken for World Vision Lao PDR. 59

Hodgdon, B.D.(2010), Community Forestry in Laos, Journal of Sustainable Forestry, 29: 1, 50-78. Holdridge, L. R. (1947). Determination of world plant formations from simple climatic data. Science (105) 367-368. Holling, C. S. (1973), ‘esilie e a d “ta ilit of E ologi al “ ste s, A Systematics 4: 1–23.

ual ‘e ie of E olog a d

Holling, C. S. ( 1978), Adaptive Evironmental Assessment and Management. Chichester, NY: John Wiley and Sons. ILO-IPEC (2003), Labour Migration Survey in Khammuane, Savannakhet and Champasack of Lao PDR. IPCC, (2007). Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I., M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, 976pp. IUCN and NERI (2011), Report on Economic, Social and Environmental Costs and Benefits of Investments in Savannakhet Province. Poverty-Environment Initiative of Lao PDR. Jintrawet, A. and Chinvanno, S, (2012), Climate Change in Southeast Asia and Assessment on Impact, Vulnerability and Adaptation on Rice Production and Water Resources. Asia-Pacific Network for Global Research, Project Reference Number: CRP2008-03CMY-Jintrawet. Kenney-Lazar, M. (2010). Land Concessions, Land Tenure, and Livelihood Change: Plantation Development in Attapeu Province, Southern Laos. Vientiane: National University of Laos. Kingsada, K. (1998), Summary of the country outlook: Lao PDR. In Asia-Pacific forestry sector outlook study working paper series. FAO. LAC (2012), Lao Agricultural Census 2010-2011. Steering Committee for the Agricultural Census, Agricultural Census Office. Lefroy, R. Collet, L and Grovermann, C. (2010), Study on Potential Impacts of Climate Change on Land Use in the Lao PDR. International Center for Tropical Agriculture (CIAT). Locatelli, B., Herawati, H., Brockhaus, M., Idinoba, M. and Kanninen, Marku (2008), Methods and Tools for Assessing the Vulnerability of Forests and People to Climate Change. CIFOR Working Paper No 43. Lestrelin, G.; Giordano, M.; Keohavong, B. (2005, Whe o se atio leads to la d deg adatio : Lessons from Ban Lak Sip, Laos. Research Report 91. Colombo, Sri Lanka: International Water Management Institute (IWMI). LPA (2005), Draft Lao Plantations Authority Operations Manual. ADB, Manila. MAF (2005), Forestry strategy to the Year 2020 of the Lao PDR. Ministry of Agriculture and Forestry, Lao PDR. McCarthy, J.J., O.F. Canziani, N.A. Leary, D.J. Dokken, and K.S. White (eds.), (2001). Climate Change 2001: Impacts, Adaptation and Vulnerability. Cambridge, U.K.: Cambridge University Press. Millennium Ecosystem Assessment. (2005). Ecosystems and Human Well-being: Synthesis. Washington, DC: Island Press. 60

Molnar, A., K. Barney, M. DeVito, A. Karsenty, D. Elson, M. Benavides, P. Tipula, C. Soria, P. Shearman and M. France (2011). Large Acquisition of Rights on Forest Lands for Tropical Timber Concessions and Commercial Wood Plantations. Washington DC: Rights and Resources Initiative. (http://www.rightsandresources.org/publication_details.php?publicationID=2242). NSC (2006), Results from the Population and Housing Census 2005. National Statistics Centre Steering Committee for Census of Population and Housing, Vientiane, Lao PDR Pei, “.T.

9 , Lao People s De o ati ‘epu li Fo est Outlook “tud . FAO ‘egio al Offi e fo Asia and the Pacific, Working Paper No. APFSOS II/WP/2009/17, Bangkok, Thailand.

Pramova E., B. Locatelli, M. Brockhaus, S. Fohlmeister, 2012b. Ecosystem services in the National Adaptation Programmes of Action. Climate Policy 12(4): 393-409. doi: 10.1080/14693062.2011.647848 Pukkala, T. (2005), Growth Models for the indigenous forests and trees in Savannakhet, Lao PRD. PSFM Technical Report. Rafanoharana, S.C., Locatelli, B. and A.J.M. Russell (2012). An analysis of climate change scenarios and their impacts on ecosystem services in Savannakhet, Lao PDR. Center for International Forestry Research; Draft Technical Report - November 2012 Rigg,J. (2007), Moving Lives: Migration and Livelihoods in the Lao PDR. Population, Space and Place (13) 163–178. Robichaud, W.G., Sinclair, A.R.E., Lanquaye, N.O., and B. Klinkenberg, (2009), Stable Forest Cover under Increasing Populations of Swidden Cultivators in Central Laos: the Roles of Intrinsic Culture and Extrinsic Wildlife Trade. Ecology and Society 14(1): 33. Russell, A.J.M., Locatelli, B., and E. Pramova. (2011). State of Knowledge Report -Forests, Ecosystem Services, and Climate Change Adaptation. PROFOR/World Bank working paper. Sandhu, H. (in press). Ecosystem services valuation framework and economic impact of alternative forest management under different climate scenarios. Schoenweger, O., and A.Uellenberg (2009), Foreign Direct Investment (FDI) in Land in the Lao PDR. GTZ, Eschborn, Germany. Schumann, G, P. Ngaosrivathana, B. Soulivanh, S. Kenpraseuth, K. Onmanivong, K.Vongphansipraseuth, and C. Bounkhong 2006. Study on state land leases and concessions in Lao PDR. Lao-German Land Policy Development Project, Land policy study No. 4. GTZ, Vientiane, 100 pp. SERCAsia (2008), A Comparative Picture of Migration in Laos, Myanmar, Cambodia, Vietnam and Thailand: Summary. (www.SERCASIA.com) MPI (2009). Background Document on Recent Socio-Economic Developments, Opportunities and Outlook. Ministry of Planning and Investment, Round Table Implementation Meeting , 3 November 2009, Vientiane Capital City. http://countryoffice.unfpa.org/lao/drive/RTIM2009BackgroundDocumentOfficialDraftforRelease30Oct09.pdf). UNIFEM (2007), Gender Concerns in Migration in Lao PDR.

61

Vongsiharath, V. (2006), Forest cover and land-use change in LAO PDR according to the national forest reconnaisance survey. Proceedings Of The Regional Land Degradation Assessment In Drylands (Lada) Workshop For Southeast Asia, Bangkok, Thailand, 27 – 30 April 2009, p.75-82. Walker, B., L. Gunderson, A. Kinzig, C. Folke, S. Carpenter, and L. Schultz. (2006). A handful of heuristics and some propositions for understanding resilience in social-ecological systems. Ecology and Society 11(1): 13. [online] URL: www.ecologyandsociety.org/vol11/iss1/art13/. WFP (2008), Comprehensive Food Security and Vulnerability Analysis (CFSVA) - Lao PDR. World Food Program, Rome, Italy. WREA/ UNDP/GEF (2009), Lao PDR - National Adaptation Programme of Action to Climate Change. Zola T. and A. Fraser (2012), Economic Evaluation of Land-Use Options In Lao PDR. Report submitted to RECOFTC.

62

8 Annexes

63

Annex 1: Case study 1: PSFM village A

Land use history

While the government classifies this village as being from the Katang minority, the villagers themselves identify more closely with the Brou ethnicity. According to the community, this village was established well before 1832 (as they refer to a document that mentions their village name from this date). The villagers have lived in this location throughout the last century, with the exception of the US-Indochina war period, 1960-75, when they were forced to live in the nearby forest (within the village boundaries) for fear of bombing by the American Air Force. The PSFM village is the best understood of the three villages due to in part to the fact this is the home village of one of the District Agriculture and Forestry Office (DAFO), and therefore the team was able to more quickly achieve rapport with the leadership. This village has also been involved in the Swedish-Lao Forestry and PSFM projects (see land use map below).

Figure 13: "PSFM village" Swedish-Lao Forestry Project Land Use Map (1999)

A small proportion (346ha) of the total village forest lands (2709ha) falls within the Dong Kapho PFA (supported by the PSFM project). However several impacts of the PSFM, and two preceding projects (JFM and FOMACOP)14 apply to the village lands overall. At the community level stakeholders have a greater understanding of the rationale for sustainable forest management and this applies to a certain deg ee th oughout its fo est a eas. I additio , as the go e e t does t allo la d o essio s i villages associated with the PFAs, and this strengthens the resolve of the community to reject the frequent requests for land from outside investors. In the PSFM community, each household is estimated to have access to an average of 17.23 ha of DDF land (see table below).

14 JFM – Lao-Swedish Joint Forest Management project (1991-1995), Forest Management and Conservation Program (FOMACOP) (1995-2001);

64

Table 23: Available land per Household (HH) in the PSFM village

Area/hh (ha) village use forest pa somsai 833 9.80 land reserved for agriculture din hae kasikam 631 7.43 Total Dry Forest Area pa kok 1,465 17.23 regeneration forest pa feun fu 432 5.08 state production forest pa palit 346 4.08 holy forest pa mahaesak 71 0.84 burial forest pa sa 8 0.09 Total Mixed Deciduous Forest pa pasom 858 10.09 paddy field na deum 277 3.26 Total agricultural land din kasikam 277 3.26 wetland din beung 54 0.64 built up area din beung 27 0.32 river bodies mae nam 22 0.26 roads sen tahng 6 0.08 Total others din eun 110 1.29 Total 2,709 31.87 source: village land use map, Lao-Swedish JFM Project, 1998. Land Use Category

B

Lao Name

Area (ha)

Analysis of household vulnerability

The caveats associated with this analysis of case studies and the vulnerability among households must be (repeatedly) underscored lest it is overlooked. Cognizant of the potential biases involved, all conclusions regarding relative drivers and characterizations of vulnerability or resilience should most appropriately be regarded as hypotheses to be tested. A detailed overview of sources of cash and non-cash income, by household wealth category is provided in the Annex. The discussion below will focus on the over-arching trends and key patterns observed.

C

Household Sources of Cash Income

In the PSFM village, stakeholder assessments of wealth categories appear to be closely aligned with measures of total cash income. While the middle wealth group reports earning on average roughly double the amount of cash of the poorer group, the highest wealth group reports cash earnings over 800% that of the poorer households. Across all PSFM village households surveyed, annual sales of livestock appear to be the overwhelmingly largest source of household cash income (60%) followed by income from labor (24%), and the sale of forest products (8%). However when differentiated by wealth categories, this pattern generally only appears to hold true for middle and wealthy income groups. For poor households, remittances are by far the largest source of income (42%), followed by livestock (29%) and forest product sales (22%). Interestingly, the sale of forest resources appears to contribute less to middle categories than to either the poor or the wealthy, and in absolute terms it provides wealthy households with double the amount of income that it does to poor households. 65

D

Total Household Income (Cash plus Non-Cash)

As was the case with cash incomes, in the PSFM village household wealth categories are closely correlated with absolute values of non-cash income from agriculture. In terms of proportional contributions to income in the PSFM community, non-cash income accounts on average for roughly 2/3rds of the total household income, with 70%, and 30% coming from rice cultivation and forest products gathered, respectively. However, the middle class group consumed a much greater amount of foods from the forest (four times) than did the poor households. Overall, middle and wealthy households consumed roughly the same value of self-cultivated or gathered foods, both consuming more than twice the amount of poorer households. Nevertheless, non-cash income sources represent the vast majority of poorer and middle class households incomes (contributing 75% and 81%, respectively), with roughly 1/3-1/2 coming from forest resources. Wealthier households rely roughly equally on cash versus non-cash income and only 10% of the noncash income comes from forest product consumption. However in absolute values, the amounts of forest products gathered are only roughly 20% below that of the poorer households. Overall, the differences in total household income among the wealth groups is remain quite significant, although less so than the initial assessment of cash income would indicate. With an average total income estimated around $2,131, wealthy households earn almost four times the total income of poor households ($521) and roughly 60% more than the middle class group ($1,340).

E

Household Cash Income vs. Expenditures

Upon asking stakeholders about household expenditures, the three greatest sources of expenditures cited by both middle and wealthier class households were to purchase rice, hire labor, and health expenses. None of the poorer households appear to (be able to) purchase labor, and so their top three expenditures go to buy rice, health expenditures and agricultural inputs. In completing a balance sheet based on reported household cash income and expenses, it would appear that most poorer and middle class households are unable to make ends meet. Based on qualitative data, however, we do not feel that this accurately reflects the quality of life of these households. Meanwhile, the wealthiest households would appear to essentially break even with no savings to show for their efforts. However, in addition to the possibility of human error in data collection, and potential aberrations due to the small sample size, data collected using the PROFOR poverty assessment participatory group exercises suggests that both cash and non-cash incomes from NTFPs and rice harvesting may be greatly underestimated by the surveys (see Annex 4). In the PSFM village the greatest amount of variance related to the poorer and middle class households estimates of the relative contributions of cash income from the sale of forest/livestock products and non-cash income from on-farm resources. The results from the poorest households are shown here for illustration. This study cannot fully explain these differences. However the former might be explained in part due to the difficulty in estimating the values and amounts of NTFPs that may be sold frequently, but intermittently, in differing quantities and at different market values. Nevertheless, given the balance sheet conclusions above, these results may suggest a significantly greater benefit derived from these resources, and therefore a significantly greater ecosystem service value of DDFs. 66

PSFM village, poor households cash and non-cash income estimates 70% 60% 50% 40% 30% 20% 10% 0% Forest and Forest and Farm cash livestock livestock cash non cash

Farm non cash

Labor and other cash

Labor and other non cash

Figure 14: Illustration of divergence between PSFM village data sources (Source: Household survey and PRA)

67

F

PSFM village – Detailed household income and expenditures

SUFORD village Wealth category Labor Livestock Remittances Forest Agriculture Others Total Cash Income US$ (yr 2012)

Cash Income per Household (kip/year) poor middle wealthy average 75,000 552,500 2,125,000 917,500 265,000 936,250 5,630,000 2,306,250 500,000 166,667 201,750 73,500 600,000 291,750 500,000 166,667 300,000 25,000 1,041,750 2,062,250 8,655,000 3,873,833 $ 130 $ 258 $ 1,082 $ 484

SUFORD village Wealth category Forest (NTFP & firewood) Agriculture Total US$ (2102)

Non-Cash Income per Household (kip/year) poor middle wealthy average $ 1,048,500 4,257,000 853,750 1,978,833 $ 247 2,080,000 4,400,000 7,537,500 4,585,000 $ 573 3,128,500 8,657,000 8,391,250 6,563,833 $ 820 $ 391 $ 1,082 $ 1,049 $ 820

Cash Income per Household (%) poor middle wealthy average 34% 49% 10% 30% 66% 51% 90% 70% 100% 100% 100% 100%

SUFORD village Wealth category Cash income Non cash income Total US$ (2102)

Household Cash and Non-Cash Income (kip/year) poor middle wealthy average $ 1,041,750 2,062,250 8,655,000 3,873,833 $ 484 3,128,500 8,657,000 8,391,250 6,563,833 $ 820 4,170,250 10,719,250 17,046,250 10,437,667 $ 1,305 $ 521 $ 1,340 $ 2,131 $ 1,305

Cash/Non-Cash Income(%) poor middle wealthy average 25% 19% 51% 37% 75% 81% 49% 63% 100% 100% 100% 100%

SUFORD village Wealth category Cash Income Cash Expenditures Balance SUFORD village Wealth category Livestock Assets SUFORD village TYPE OF EXPENDITURE Rice buying expenditures Hiring labor Social events Health expenditures Clothes Meat and fish buying Other food buying Electricity bills and appliances Kitchen utensils School expenses Agricultural inputs Marriage price Contribution to savings fund TOTAL HH EXPENDITURES TOTAL HH CASH INCOME BALANCE

$ $ $ $ $ $ $ $

115 288 21 36 21 3 484

Cash Income per Household (%) poor middle wealthy average 7% 27% 25% 24% 25% 45% 65% 60% 48% 0% 0% 4% 19% 4% 7% 8% 0% 24% 0% 4% 0% 0% 3% 1% 100% 100% 100% 100%

Household Cash Income and Expenditures (kip/year) poor middle wealthy average 1,041,750 2,062,250 8,655,000 3,873,833 3,676,250 7,685,833 8,431,458 5,871,458 (2,634,500) (5,623,583) 223,542 (1,997,625)

HH Cash Income/Expenditures ($) poor middle wealthy average $ 130 $ 258 $ 1,082 $ 484 $ 460 $ 961 $ 1,054 $ 734 $ (329) $ (703) $ 28 $ (250)

Livestock Assets per Household (kip) poor middle wealthy average 1,487,500 6,878,750 33,362,500 13,909,583

Livestock Assets/HH ($) poor middle wealthy average $ 186 $ 860 $ 4,170 $ 1,739

Expenditures in kip per Household Category Expenditures in % poor middle wealthy average poor middle wealthy average 1,512,500 1,732,500 2,478,125 1,907,708 41% 23% 29% 32% 1,500,000 1,500,000 392,500 0% 20% 18% 7% 250,000 300,000 300,000 441,667 7% 4% 4% 8% 675,000 1,375,000 1,375,000 1,058,333 18% 18% 16% 18% 161,250 875,000 875,000 645,417 4% 11% 10% 11% 100,000 400,000 400,000 441,667 3% 5% 5% 8% 127,500 150,000 100,000 600,000 -

400,000 400,000 183,333 520,000 -

400,000 400,000 183,333 520,000 -

270,000 200,000 334,167 180,000 -

3% 0% 4% 3% 16% 0%

5% 0% 5% 2% 7% 0%

5% 0% 5% 2% 6% 0%

5% 0% 3% 6% 3% 0%

3,676,250 1,041,750 (2,634,500)

7,685,833 2,062,250 (5,623,583)

8,431,458 8,655,000 223,542

5,871,458 4,302,000 (1,569,458)

0% 100% 0% 0%

0% 100% 0% 0%

0% 100% 0% 0%

0% 100% 0% 0%

68

Annex 2: Case study: Eucalyptus village A

Land use history

The inhabitants of this village belong to the Lao Xouay ethnicity, a population whose language, physical features, and up-land farming settlement locations relate them most closely with the Katang ethnicity. However, this distinction is slightly contentious as some community leaders claim to regard their ethnicity as belonging to the largest national grouping of ethnicities (the Lao loung). According to local and district officials, the village was officially relocated from its original location on the banks of the Xaha River (shown on map) to its present location two years ago. A district official indicated that this village location had been somewhat artificially established by the French colonial ad i ist ato s p essure on communities to locate themselves near rivers where they would be more easily monitored (at the time there were no roads connecting this village to the outside world). The primary reasons for the recent village relocation was indicated as being in order to enable the villagers to access a new power line brought in by the government, and to provide for more room for population and paddy expansion. However, the utility line route suggests that little was gained (for the villagers) from this move. Indeed, while a significant part of the new village site does not have power, a newly established sawmill located in the middle of the village does. The government has i di ated the i te sio to esta lish a fo al a ea a ket at this site. In addition to the changes to livelihoods due to the village relocation, in 2006, the Indian-owned Birla pulp ood o pa e ei ed the o essio to o e t oughl ha o e thi d of this illage s territory into eucalyptus plantation. Unfortunately, the plantation management did not adequately assess or respect existing land use patterns in the delineation of its concession boundaries, resulting in significant adverse impacts on local livelihoods. An estimated 68ha of existing paddy fields were appropriated by the company, representing 28% of the total village paddy in use. Additionally, an u k o a ou t of la d that as fo all egiste ed illage s fo futu e ge e atio s padd e pa sio k o as chap chong din 15), as well as non-claimed standing (predominantly) DDFs were bulldozed for inclusion in the plantation. In total, 246ha of forest were converted, or roughly 33% of the total village forest lands.

15

These are forested lands that are claimed by individuals for future development into paddy, upon which the individuals pay land taxes to the district government.

69

Table 24: Paddy and Forest converted to Eucalyptus

Land Use Category Total land area of the village Total land remaining for villagers Paddy Original paddy area Paddy converted to Eucalyptus Remaining paddy area Share of paddy lost Forest Original forest area Forest converted to Eucalyptus Total area planted to Eucalyptus Remaining forest area Share of forest lost

area (ha) % of total 977.45 100% 664.01 68% 241.09 67.58 173.51

25% 7% 18% 28%

736.36 245.86 313.44 490.5

75% 25% 32% 50% 33%

Source: NAFRI

This has resulted in a range of conflicts between the villagers and the company, based on which the local law enforcement agencies have documented over 40 individual disputes. Unfortunately, the police lack a formal institutional framework for mediation and arbitration in local disputes over land claims as there are significant inconsistencies between the policies permitting and regulating land concession titling and local land registration.

B

Analysis of household vulnerability

The caveats associated with this analysis of case studies and the vulnerability among households must be (repeatedly) underscored lest it is overlooked. Cognizant of the potential biases involved, all conclusions regarding relative drivers and characterizations of vulnerability or resilience should most appropriately be regarded as hypotheses to be tested. A detailed overview of sources of cash and non-cash income, by household wealth category is provided in the Annex. The discussion below will focus on the over-arching trends and key patterns observed.

C

Household Sources of Cash Income

This o u it s lassifi atio of household ealth atego ies also appea to e o elated ith measures total cash income (see Annex 2). The mean levels of cash income across these three groups appea s ha e a e o al dist i utio ith the iddle lass earning ($899) on average 87% more than the poor ($480), and 85% less than the wealthy households ($1,663). If we look at the overall sources of cash income across the Eu al ptus illage households surveyed, income from labor appears to be the largest source of household cash income (50%) followed by income from the sale of livestock (35%) and remittances (9%). As we look at the distribution of sources of cash by wealth category we note the correlation between increasing wealth and income from labor and livestock sales, but an inverse relationship with importance of income from remittance and overall diversity of sources of cash income. A major difference between middle and wealthy class households is the a mount of income received from the sale of livestock, as wealthy households earn between nine and sixteen times the amount earned by

70

poor and middle income households, respectively. Conversely, though not comparable in absolute terms, middle income households earn 8% of their cash income from the sale of forest products, roughly four times the absolute value earned by poorer and wealthier households.

D

Total Household Income (Cash plus Non-Cash)

This o u it s household ealth atego ies appea to also correspond closely with the overall amounts of non- ash i o e e ei ed f o fo ests a d fa s. Ho e e , i o i g the iddle lass s greater sale of NTFPs, this class also consumes two to three times the amount of NTFPs consumed by wealthier or poorer household, respectively. Interestingly, when we combine the values of NTFPs sold and consumed, the poor also appear to gain significantly less value ($154) than do the wealthy ($225). In addition, poorer and middle class households consume roughly half the amount of rice consumed by wealthier households. Overall, the contributions of non-cash income have a slightly leveling effect on overall household i o e dist i utio s a oss the o u it , ith the iddle lass ea i g $ , % o e tha the poor ($959), and 65% less than the wealthier households ($2,552).

E

Household Cash Income vs. Expenditures

In this village, the amount of expenditures for the purchase of rice, while the single largest expenditure overall (29%) appears to decline with wealth category, and the wealthiest households appear to be fully rice self-sufficient. This would make sense given the much higher consumption of agricultural produce by the wealthiest households, and their greater investment in agricultural inputs. We cannot interpret why the lowest and highest wealth group households would have roughly the same amount of expenditures, while the middle wealth group had twice their level of expenditures and conclude that this is due to human error in the data collection process with regard to this middle category. If the data for the poorer and wealthier household categories is correct, the poorer households balance appears to be roughly 18% below income, while the wealthy households would appear to earn/consume twice the amount of resources that they expend. As was the case in for the PSFM community, these results are somewhat confounded by significant differences between perceived contributions from different livelihood activities between the household survey and PRA exercises. As with the PSFM village poorer household estimates of noncash income from forests and farms, these display a significant divergence between data sets, while the middle income group display the greatest divergence in across cash and non-cash sources of income (see table below). These differences cannot be adequately resolved with the available data.

71

Eucalyptus village medium households cash and non-cash income estimates 70% 60%

50% 40% 30% 20% 10% 0%

Forest and Forest and livestock livestock cash non cash

Farm cash

Farm non cash

Labor and other cash

Labor and other non cash

Figure 15: Illustration of divergence between Eucalyptus village data sources (Source: Household survey and PRA)

72

F

Eucalyptus village – Detailed household income and expenditures

Eucalyptus village Wea l th ca tegory La bor Li ves tock Remi tta nces Fores t Agri cul ture Others Total Cash Income US$ (yr 2012)

Cash Income per Household (kip/year) poor mi ddl e wea l thy a vera ge 2,125,000 4,650,000 5,300,000 4,025,000 455,000 830,000 7,300,000 2,861,667 1,100,000 707,500 450,000 752,500 162,500 601,500 150,000 304,667 105,000 105,000 75,000 0 300000 0 100000 3,842,500 7,194,000 13,305,000 8,118,833 $ 480 $ 899 $ 1,663 $ 1,015

Eucalyptus village Wea l th ca tegory Fores t (NTFP & fi rewood) Agri cul ture Tota l US$ (2102)

Non-Cash Income per Household (kip/year) poor mi ddl e wea l thy 1,856,500 4,239,125 2,437,500 2,760,000 2,683,750 5,457,500 4,616,500 6,922,875 7,895,000 $ 577 $ 865 $ 987

Eucalyptus village Wea l th ca tegory Ca s h i ncome Non ca s h i ncome Tota l US$ (2102)

Household Cash and Non-Cash Income (kip/year) poor mi ddl e wea l thy a vera ge 3,842,500 7,194,000 13,305,000 8,118,833 4,616,500 6,922,875 7,895,000 6,478,125 8,459,000 14,116,875 21,200,000 14,596,958 $ 1,057 $ 1,765 $ 2,650 $ 1,825

Eucalyptus village Wea l th ca tegory Ca s h Income Ca s h Expendi tures Ba l a nce

Household Cash Income-Expenditures (kip/year) poor mi ddl e wea l thy a vera ge 3,842,500 7,194,000 13,305,000 8,118,833 4,675,000 10,350,000 4,421,500 6,482,167 (832,500) (3,156,000) 8,883,500 1,636,667

HH Cash Income/Expenditures ($) poor mi ddl e wea l thy a vera ge $ 480 $ 899 $ 1,663 $ 1,015 $ 584 $ 1,294 $ 553 $ 810 $ (104) $ (395) $ 1,110 $ 205

Eucalyptus village Wea l th ca tegory Li ves tock As s ets

Livestock Assets per Household (kip) poor mi ddl e wea l thy 3,695,000 7,721,250 17,210,000

Livestock Assets/HH ($) poor mi ddl e wea l thy a vera ge $ 462 $ 965 $ 2,151 $ 1,193

Eucalyptus village TYPE OF EXPENDITURE Ri ce buyi ng expendi tures Hi ri ng l a bor Soci a l events Hea l th expendi tures Cl othes

Expenditures in kip per Household Category Expenditures in % poor mi ddl e wea l thy a vera ge poor mi ddl e wea l thy a vera ge 3,295,000 1,937,500 1,744,167 1 19% 0% 27% 3,600,000 725,000 1,441,667 35% 16% 22% 400,000 425,000 900,000 575,000 0 4% 20% 9% 75,000 500,000 475,000 350,000 0 5% 11% 5% 375,000 650,000 525,000 516,667 0 6% 12% 8%

Mea t a nd fi s h buyi ng Other food buyi ng El ectri ci ty bi l l s a nd a ppl i a nce Ki tchen utens i l s School expens es Agri cul tura l i nputs Ma rri a ge pri ce Contri buti on to s a vi ngs fund TOTAL HH EXPENDITURES TOTAL HH CASH INCOME BALANCE

a vera ge 2,844,375 3,633,750 6,478,125 $ 810

Cash Income per Household (%) poor mi ddl e wea l thy 55% 65% 40% 12% 12% 55% 29% 10% 3% 4% 8% 1% 0% 1% 1% 0% 4% 0% 100% 100% 100%

a vera ge 50% 35% 9% 4% 1% 1% 100%

$ $ $

$ 356 454 810

poor mi ddl e wea l thy 40% 61% 31% 60% 39% 69% 100% 100% 100%

a vera ge 44% 56% 100%

$ $ 1,015 $ 810 $ 1,825

Cash/Non-Cash Income(%) poor mi ddl e wea l thy 45% 51% 63% 55% 49% 37% 100% 100% 100%

a vera ge 56% 44% 100%

a vera ge 9,542,083

180,000 175,000 100,000 75,000 -

380,000 270,000 1,540,000 660,000 387,500 -

650,000 175,000 209,000 435,000 325,000 2,500

403,333 206,667 583,000 398,333 262,500 833

4,675,000 3,842,500 (832,500)

10,350,000 7,037,500 (3,312,500)

4,421,500 13,476,500 9,055,000

6,482,167 (6,482,167)

73

$ $ $ $ $ $ $

$ 503 358 94 38 9 13 1,015

0 0 0 0 1 1 (0)

4% 3% 15% 6% 4% 0%

15% 4% 5% 10% 7% 0%

6% 3% 9% 6% 4% 0%

0% 0% 100% 68% -32%

0% 0% 100% 305% 205%

0% 0% 100% 0% -100%

Annex 3: Case study: Sugarcane village A

Overall land use history

The inhabitants of this village belong to the Katang ethnicity, an ethnicity that is closely related in language, physical features, and up-land farming settlement locations to the Xouay of the Eucalyptus village. However, this village has undergone a significant reorganization during recent years, as five villages were merged into two new administrative units. In addition to the demographic changes, livelihoods have been significantly altered with the establishment of a sugar cane concession on its lands. In this village, the Savan Sugar Company engaged actively with the community in order to avoid expropriating either paddies currently in use or the lands identified by the community as being egiste ed as chap chong din fo est la d pla ed i ese e for future paddy expansion needs). Therefore this concession has not been burdened by any significant amount of conflicts with the community. Nevertheless, the impact on livelihoods of this concession can be regarded as greater than in the Eucalyptus village as the sugar cane concession has taken up almost all the remaining forest lands available to the community, leaving villagers with very little forest for livestock to graze in, or for the collection of NTFPs. No village map or map of the concession was available to us. However, based on the land holdings under chap chong din registry and number of cattle owned by the sampled households, the average area of grazing land that available per head of livestock appears to be around 0.43ha, which clearly suggests an enormous pressure the resources base, translating into pressure on households to sell their stock or hope that cattle will be able graze sufficiently in neighboring village lands. Stakeholders also indicate that the availability of NTFPs for consumption has plummeted.

B

Analysis of household vulnerability

The caveats associated with this analysis of case studies and the vulnerability among households must be (repeatedly) underscored lest it is overlooked. Cognizant of the potential biases involved, all conclusions regarding relative drivers and characterizations of vulnerability or resilience should most appropriately be regarded as hypotheses to be tested. A detailed overview of sources of cash and non-cash income, by household wealth category is provided in the Annex. The discussion below will focus on the over-arching trends and key patterns observed.

C

Household Sources of Cash Income

Stakeholder assessments of wealth categories in this community appear to be closely aligned with measures of total cash income, and the distribution of cash income is falls between the other case studies in terms of its skewedness. As was the case in the Eucalyptus village, income from labor appears to be the largest source of household cash income (51%), however whereas the reliance on labor income increased with wealth category in the Eucalyptus village, this relationship is reversed in the sugarcane village where fully 89% of cash income comes from labor. Across the village, labor is followed by income from remittances (20%), and the sale of livestock (17%). Diversification of sources of cash income is closely correlated with wealth. Among the poorer households, sale of labor accounts for 89% of cash income with the remainder coming from the sale 74

of fo est p odu ts. Mea hile the iddle ealth g oup s i o e f o la o a d li esto k sales a e quite close (46%, 40% respectively), followed by forest product sales (12%). Among the wealthier households, remittances, labor and livestock provide the top three sources of cash income (38%, 29%, 19%, respectively) and indeed, only these wealthier households received any income from remittances. In this village the poorer households own very few cattle and do not report selling any. Although middle and wealthier households earn roughly the same amount of income from cattle sales ($170), this represents almost a third of the total middle class cattle holdings, while it only represents around 11% of the livestock owned by the wealthier households suggesting either that these households had a greater need for capital this last year, or that there is some disproportionate pressure on middle class household to sell off livestock due to a shortage of forage overall. The loss of ost of this o u it s fo est a ea a a ts the i t odu tio of se e al a eats i the interpretation of this data. The relatively low contribution of forest product sales to household cash income across wealth groups (6-10%) is described by villagers as being due to an absolute shortage of forest available and the over-exploitation of most fauna species, however interestingly in absolute terms the wealthiest households earned roughly twice the amount of NTFPs as did the poor where most forest products gathered appear to be consumed within the home.

D

Overall Household Income (Cash plus Non-Cash)

I additio to ea i g a st o g elatio ship ith ash i o e, it ould appea that this o u it s classification of wealth groups also bears a strong correlation with increased rice consumption and reduced consumption of forest foods. However, in contrast with the other case studies, there is no clear relationship between wealth and relative contribution of cash vs. non-cash income with the wealthier households receiving a greater proportion (67%) of their total income from non-cash sources than did the poor (63%), and middle group (51%). In this community, consumption of forest products appears to be of greatest importance to the poorest households, where this contributes 70% of non-cash income (roughly twice in actual volume of forest foods consumed by the wealthy), although in absolute terms the middle class households consume roughly 85% the amount of poorer households. Conversely, wealthier and middle class households appear to consume over 5-6 times more rice than poorer households. Upon inclusion of non-cash income sources, the total household income increases to: wealthy ($1,931) and middle ($1,281), poor ($778).

E

Household Cash Income vs. Expenditures

Upon asking stakeholders about household expenditures, the single greatest source of expenditures for each of the three wealth classes was for the purchase of rice, amounting to roughly 95% (poor), 68% (middle), and 15% (wealthy) of the total cash annual household cash resources. Among the middle and poorer households, social events were cited second, while the wealthy cited the purchase of other foods almost as highly as their purchase of rice. Third for each group were purchase of clothing (poorer), utility bills (middle), and only the wealthy cited any significant cost incurred to the hiring of labor.

75

In completing a balance sheet based on reported household cash income and expenses, it would appear that poorer and middle class households expenditures outweigh sources of cash and non-cash income, while wealthier households reported an average 12% surplus. Most remarkably, however, focus group discussions in all wealth categories indicated particularly large disparities in the perceived contributions from the sale of forest products in addition to cash income from labor (see annex, and illustration in table below).

Sugarcane village, wealthy households cash and non-cash income estimates 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Forest and Forest and livestock livestock cash non cash

Farm cash

Farm non cash

Labor and other cash

Labor and other non cash

Figure 16: Illustration of divergence between Sugarcane village data sources (Source: Household survey and PRA)

76

F

Sugarcane village – Detailed household income and expenditures

Sugarcane village Wealth category Labor Livestock Remittances Forest Agriculture Others Total Cash Income US$ (yr 2012)

Sugarcane village Wealth category Forest (NTFP & firewood) Agriculture Total US$ (2102)

Cash Income per Household (kip/year) poor middle wealthy average 3,493,750 2,393,750 3,025,000 4,206,250 1,925,000 1,962,500 1,362,500 3,900,000 1,625,000 308,250 524,750 632,500 469,083 637,500 247,500 500,000 500,000 166,667 3,944,500 5,253,500 10,352,500 8,234,500 $ 493 $ 657 $ 1,294 $ 1,029

$ $ 526 $ 170 $ 203 $ 59 $ 31 $ 21 $ 1,029

Cash Income per Household (%) poor middle wealthy average 89% 46% 29% 51% 0% 37% 19% 17% 0% 0% 38% 20% 8% 10% 6% 6% 0% 0% 6% 3% 13% 10% 0% 2% 100% 100% 100% 100%

Non-Cash Income per Household (kip/year) poor middle wealthy average $ 1,597,625 1,373,625 728,750 1,248,125 $ 156 680,000 3,621,250 4,363,750 2,888,333 $ 361 2,277,625 4,994,875 5,092,500 4,136,458 $ 517 $ 285 $ 624 $ 637 $ 517

Cash Income per Household (%) poor middle wealthy average 70% 28% 14% 30% 30% 72% 86% 70% 100% 100% 100% 100%

Sugarcane village Wealth category Cash income Non cash income Total US$ (2102)

Household Cash and Non-Cash Income (kip/year) poor middle wealthy average $ 3,944,500 5,253,500 10,352,500 8,234,500 $ 1,029 2,277,625 4,994,875 5,092,500 4,136,458 $ 517 6,222,125 10,248,375 15,445,000 12,370,958 $ 1,546 $ 778 $ 1,281 $ 1,931 $ 1,546

Cash/Non-Cash Income(%) poor middle wealthy average 63% 51% 67% 67% 37% 49% 33% 33% 100% 100% 100% 100%

Sugarcane village Wealth category Cash Income Cash Expenditures Balance

Household Cash Income-Expenditures (kip/year) poor middle wealthy average 3,944,500 5,253,500 10,352,500 8,234,500 5,223,500 5,578,500 8,973,750 8,180,750 (1,279,000) (325,000) 1,378,750 53,750

HH Cash Income/Expenditures ($) poor middle wealthy average $ 493 $ 657 $ 1,294 $ 1,029 $ 653 $ 697 $ 1,122 $ 1,023 $ (160) $ (41) $ 172 $ 7

Livestock Assets per Household (kip) poor middle wealthy average 81,250 6,066,250 17,102,500 7,750,000

Livestock Assets/HH ($) poor middle wealthy average $ 10 $ 758 $ 2,138 $ 969

Sugarcane village Wealth category Livestock Assets Sugarcane village TYPE OF EXPENDITURE Rice buying expenditures Hiring labor Social events Health expenditures Clothes Meat and fish buying Other food buying Electricity bills and appliances Kitchen utensils School expenses Agricultural inputs Marriage price Contribution to savings fund TOTAL HH EXPENDITURES TOTAL HH CASH INCOME

BALANCE

Expenditures in kip per Household Category Expenditures in % poor middle wealthy average poor middle wealthy average 4,065,000 3,640,000 1,482,500 2,367,500 78% 65% 17% 29% 25,000 1,125,000 458,333 0% 0% 13% 6% 330,000 850,000 875,000 1,251,667 6% 15% 10% 15% 125,000 200,000 865,000 555,000 2% 4% 10% 7% 250,000 200,000 600,000 516,667 5% 4% 7% 6% 135,000 103,500 65,000 25,000 125,000 5,223,500 4,302,000

60,000 85,000 278,500 65,000 50,000 125,000 5,578,500 5,343,500

831,250 1,415,000 877,500 167,500 300,000 435,000 8,973,750 10,157,500

573,750 736,667 540,333 194,167 133,333 353,333 500,000 8,180,750 3,842,500

0% 3% 2% 1% 0% 2% 0% 0% 100% 82%

1% 2% 5% 1% 1% 2% 0% 0% 100% 96%

9% 16% 10% 2% 3% 5% 0% 0% 100% 113%

7% 9% 7% 2% 2% 4% 6% 0% 100% 47%

(921,500)

(235,000)

1,183,750

(4,338,250)

-18%

-4%

13%

-53%

77

Annex 4. Range of estimated cash and non-cash income values (Sources: survey and PRA exercises)

78

Annex 5. Market prices for primary NTFPs collected in case study communities, Savannakhet Province (April, 2012)

Lao Name

Scientific Name

Price (kip/kg)

A: Mushrooms het bot

Lentinus polychrous

het poh

Astraeus hygrometricus

50,000

het langoh

Amanita spp.

25,000

het pouak

Termitomyces spp.

30,000

B: Frogs, snails and other water animals kob khakam

Rana sp.

20,000

kiet na

Rana lymnocharis

10,000

eung

Kaloula pulchra

20,000

meng kaeng

Stink bugs of Tesseratoma papillosa

50,000

khai mot daeng

red ant eggs of Oecophylla smaragdina

30,000

chak chang

cicads of Orientopsaltria sp.

50,000

C: Insect products

D: Bamboo and rattan shoots nor lan

2,000

nor san

2,000

nor kasen

1,000

vai

rattan: Calamus spp.

500

D: Wild vegetables and fruits pak wan

Melientha suavis

50,000

dok kachiao

Curcuma sp.

10,000

pak kadone

Barringtonia spp.

5,000

mak kamphom

Emblica officinalis

8,000

dammar resin from Shorea obtusa

4,000

E: Tree exudates ki si

F: Timber and firewood species koung

Dipterocarpus tuberculatus

sat

Dipterocarpus obtusifolius

hang

Shorea siamensis

chik

Shorea obtusa

tiu

Cratoxylon formosum

daeng

Xylia xylocarpa var. kerrii

seuak

Terminalia spp.

79

Annex 6: Key policies impacting forests and concessions Policy Forestry 2020

Strategy

Implementing agency

Goals relating to DDF

Department Forestry

Restoring forest cover to 65% by 2015, 70% by 2020

of

Subdivides forest estate into three categories:   

Seventh National Socio-Economic Development Plan (2011-2015)

Ministry of Planning and Investment, Ministry of Industry and Commerce, Ministry of Energy and Mines

Conservation in PAs, Sustainable Forest Management in PFAs, Plantations elsewhere.

National 8% economic growth annually   

GDP per capita at least USD 1,700 Exit Least Developed Country (LDC) by 2020, Reduce poverty to 19% of pop. and 11% of households by 2015

Foreign investment in land: $8-8.75 billion between 2011-2015, incl. establishing 500,000ha tree plantations by 2020 Develop hydropower: 10 additional large dams to produce 5,015MW; Develop mining/processing: copper plates (86,200 t/yr.), gold bars (6 t/yr.), coal (728,000 t/yr.), copper ore (298,000 t/yr.), and gypsum (600,000 t/yr.).

Labor Employment Strategy

and

Agricultural Sector Strategy

Provide labor: agriculture sector (2.29 million), processing, industry and construction (226,000), service (739,000).

Ministry of Agriculture and Forestry

Food security: To produce rice equivalent to 4 million tons, at 3.9 tons per hectare on average; Cash crop production for income generation: expand maize production to 150,000 hectares, produce coffee at more than 553,000 tons. Livestock Production for consumer markets: produce meat at 32 kg/person/year, aquatic products (fish, frogs, shrimp) at 22 kg/person/year, produce 120,000 cattle for export.

Land Policy

Land Department in Ministry of Natural Resources and Environment

Equitable Land Use Planning: To create detailed land management plans at macro and micro levels , and land use plans in villages and village-clusters (Kumbans) across the country

National Strategy on Climate Change (NSCC)

Ministry of Resources Environment

Reduce carbon emissions (NAMAs): only one NAMA developed for the transportation sector: target reduction of 440,000 t CO2 for 2013-2020.

Decentralization Policy

Prime Minister's Office, Ministry of the Interior

Decentralization of services: through establishment, resettlement around rural trade focal areas

Renewable Energy Development Strategy 2020

Ministry of Energy and Mines

Production of 205 million liters of biodiesel by 2020, roughly requires approx. 400,000 ha of jatropha

Natural and

Land Titling for tenure security and tax collection: issuing one million land titles, achieve a three-fold increase in land revenue (5% of the national revenue).

Climate Adaptation (NAPAs): not yet developed

80