A method for the economic valuation of non-timber tropical forest products

A method for the economic valuation of non-timber tropical forest products
of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A Method for the Economic Valuation of Non-Timber Tropical Forest ProductsAuthor(s): Ricardo Godoy, Ruben Lubowski and Anil MarkandyaSource: Economic Botany, Vol. 47, No. 3 (Jul. - Sep., 1993), pp. 220-233Published by: Springer  on behalf of New York Botanical Garden Press Stable URL: . Accessed: 15/02/2014 12:04 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at  .  . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact  .  New York Botanical Garden Press  and Springer   are collaborating with JSTOR to digitize, preserve and extendaccess to  Economic Botany. This content downloaded from on Sat, 15 Feb 2014 12:04:04 PMAll use subject to JSTOR Terms and Conditions  A METHOD FOR THE ECONOMIC VALUATION OF NON-TIMBER TROPICAL FOREST PRODUCTS' RICARDO GODOY, RUBEN LUBOWSKI, AND ANIL MARKANDYA Godoy, Ricardo, Ruben Lubowski, and Anil Markandya (Harvard nstitute or International Development, Harvard University, Cambridge, MA 02138, U.S.A.). A METHOD FOR THE ECONOM- IC VALUATION OF NON-TIMBER TROPICAL OREST PRoDucTs. Economic Botany 47(3):220-233. 1993. By drawing on quantitative tudies in social anthropology, oology, ethnobotany, and economics we present a method or conducting an economic valuation of non-timber orest products. A review of 24 studies suggests hat the median value or non-timber orest products s about 50/ha/year. We discuss problems with past studies and suggest ways o get better stimates of output quantities, marginal costs, and prices. Key Words: non-timber orest products; ropical forests; economic valuation; methods; sus- tainability. Policy-makers often assume tropical forests have no economic value unless they are logged or farmed (Dove 1983; Hecht, Anderson, and May 1988). Besides timber, tropical forests pro- duce food (Arnold 1991; FAO 1989), construc- tion materials, medicinal plants, fodder, and fire- wood, all of which villagers use daily (Panayotou and Ashton 1992). As a proxy for the opportunity cost of the forest-the maximum value derivable from the forest before it is put to new uses (Go- doy 1992)-the economic value of non-timber tropical forest products s conservative because the forest also produces other benefits, such as biological diversity and environmental services (Panayotou, pers. comm.). Policy-makers and development organizations need an accurate es- timate of the opportunity cost of the forest to evaluate proposed projects and filter out eco- nomically disadvantageous ones. Under some conditions, eaving the forest unlogged and using it to get non-timber forest goods and environ- mental services may be socially and economi- cally optimal (Panayotou, pers. comm.). In this article we review studies estimating he economic value of non-timber tropical forest products. Drawing on anthropology and eco- nomics, we present a method for future valuation ' Received 8 April 1993; accepted 5 May 1993. studies. We limit our discussion to non-timber forest goods because methods for valuing other goods and services from the tropical forest (e.g., timber, biodiversity) are outlined elsewhere Go- doy 1992; Panayotou, pers. comm.) and because the valuation of non-timber products s the area of tropical orestry n which we are ikely to make the greatest contribution. Ethnographers re fa- miliar with the social uses of many forest goods through heir research on hunting and gathering and swidden agriculture. Ethnobotanists have provided detailed inventories of useful forest plants. Social anthropologists have the classic ethnographic ieldwork ools, such as participant observation, time allocation methods, house- hold diaries, direct records of the goods extracted from the forest, and long-term ield research, hat give them a comparative advantage n studies of forest products. We focus on tropical orests or brevity, though the valuation method outlined here could be ap- plied to any forest type. Lastly, we focus on val- uation at the forest gate, the point at which vil- lagers irst sell or consume the goods, rather han assessing he value of forest goods to the regional, national, or world economy. We omit studies that present too little information o permit cal- culation of yearly profits per hectare e.g., Fearn- side 1989; Hart 1978; Hecht, Anderson, and May 1988; Jessup and Vayda 1989; Peluso and Pof- Economic Botany 47(3) pp. 220-233. 1993 ? 1993, by The New York Botanical Garden, Bronx, NY 10458 U.S.A. This content downloaded from on Sat, 15 Feb 2014 12:04:04 PMAll use subject to JSTOR Terms and Conditions  1993] GODOY ET AL.: METHOD FOR VALUATION 221 fenberger 1989). In assessing he studies and pro- ducing our methods we draw on the anthropo- logical iterature n hunters and foragers, urveys of methods in social anthropology e.g., Bernard et al. 1984, 1986; Borgerhoff and Caro 1985; Gregory and Altman 1989; Gross 1984), con- versations with some of the leading researchers in the study of tropical-forest plant and animal use, and our own fieldwork experience in Asia and Latin America. Until recently few scholars had studied the economic value of non-timber tropical-forest products (e.g., Dunn 1975). In 1988 the Inter- national Tropical Timber Organization pub- lished a report calling for the rigorous study of these products (Panayotou and Ashton 1992). Since then many researchers ave done such val- uations (e.g., de Beer and McDermott 1989; Go- doy and Feaw 1989; Padoch and de Jong 1989; Peters, Gentry, and Mendelsohn 1989; Plotkin and Famolare 1992; Schwartzman 1989). Table 1 contains a summary of these studies and points up four problems with them: incompatibility of results, a tendency to examine flora (mainly) or fauna but not both, a lack of attention to sus- tainability, and a disproportionate attention to Latin America. The net values vary widely, ranging rom about 1 to about 420/ha/year. The variation can be explained by the biological and economic diver- sity of the different tudy sites sampled, the dif- ference in the methods and assumptions used, and the different products studied. Even when studying he same goods, independent valuations conducted at nearly the same time have pro- duced different results (e.g., Padoch and de Jong 1989; Peters, Gentry, and Mendelsohn 1989). Even a single plot may yield different values de- pending on the valuation technique used (Abey- gunawardena and Wickramasinghe 1992). Fu- ture valuations should use a common method and reporting procedure to facilitate the com- parison of results (Borgerhoff and Caro 1985; Johnson 1978; NSF/USAID 1989:12; Redford and Robinson 1987). Though hunting s reputed o produce benefits comparable to those from converting forest to farmland (Panayotou and Ashton 1988; Paucar and Gardner 1981), few researchers have esti- mated the economic value of the fauna extracted, and, as far as we know, no one has measured he combined economic value of plants and animals. A complete economic valuation of non-timber tropical-forest roducts must include both fauna and flora. Ethnographies f hunting and fishing are of limited use in estimating he value of wild- life per hectare because their authors do not re- port both the size of the catchment area and the monetary value of the game caught. Few researchers discuss the extent to which extractive practices are sustainable (Schwartz- man 1989). Lastly, most of the studies come from Latin America, mainly the Amazon basin. We need studies from the tropical forests of Africa and Asia, where different harvesting methods, forest types, societies, and economies may yield different values (Anderson 1990, citing Nair 1985). Besides addressing hese issues, research- ers must discuss the reasons for their choice of study site, the sample of people studied, and the representativeness f the study period so other researchers an assess the typicality of their find- ings. They will need to measure the quantity, location, price, and marginal cost of the goods brought from the forest to determine their net value. For non-marginal se or extraction ates, where extraction is sustainable, the value of the non- timber products may be measured as Qi(Pi - Ci), where Qi = the quantity of good extracted, Pi = the forest gate price of the good (which may be equal to its price under competitive market conditions with no externalities), C, = cost of extraction and i = set of non-timber tropical products. If the extraction rates are so large as to change the prices of the products concerned, or supply a significant hare of the market, hen allowance needs to be made for the intra-mar- ginal values, by measuring the consumer and producer surplus hat is created by the supply of the goods. Furthermore, f the extraction rates are non-sustainable, adjustment hould be made for the eventual depletion of the products by adding to Ci, a depletion premium based on the expected date of extraction. Markandya and Pearce (1987) discuss how to calculate this pre- mium. In this context, researchers hould eval- uate directly the sustainability of plant and an- imal extraction, rather than assume that local people necessarily use the forest in an ecologi- cally sound manner. Even after allowing for all these corrections, this method of valuation underestimates he val- ue of non-timber ropical orest products because it does not fully capture he benefits of its services in lean years (Arnold 1991; FAO 1989). This content downloaded from on Sat, 15 Feb 2014 12:04:04 PMAll use subject to JSTOR Terms and Conditions  222 ECONOMIC BOTANY [VOL. 47 THE CONTEXT In few of the studies surveyed do the research- ers discuss the biological representativeness of their research ites, so it remains unclear o what extent the sites resemble he forest around hem. Since tropical orests are heterogeneous, different study sites will yield different nformation. Un- less researchers discuss the biophysical charac- teristics of their sites, it is difficult o generalize or to compare the results of different tudies. To make sure that the study site represents he sur- rounding orest and to get a measure of variance one should where possible choose a sample of study sites. Researchers may often be unable to select representative orest sites, but they should always discuss the reasons for their choices to enable other scholars to compare and evaluate their results. Proper ampling of the population s necessary for a study to be generalizable Gregory and Alt- man 1989; Gross 1984). The main attributes of forest-product extractors which influence har- vesting should first be identified. Important at- tributes may include age, technology, and in- come: men under 17 years of age or over 50 may hunt ess frequently Hill and Hawkes 1983), een- age hunters may be less skilled (Yost and Kelley 1983), shotguns are better than blowguns and bows for certain types of kills (Stearman 1990; Yost and Kelley 1983), and richer villagers may rely less on forest goods (FAO 1989). After den- tifying the major variables, he researcher hould conduct a census, noting the distribution of the attributes of interest, and then select a stratified random sample to ensure that the study group mirrors he entire population in these attributes (Borgerhoff and Caro 1985). Most anthropolo- gists working with extractors of forest-products do not use stratified random samples because these often include people who are difficult to reach. Anthropological amples are usually too small for conventional statistical esting Behrens 1990). Reviewing ime-allocation tudies, Minge- Klevana (1980) finds a median sample of about 50 households or persons. Unless the researcher ollects information for 12 consecutive months, sampling must cover the seasonal variations in a year. Hurtado and Hill (1990), Henley (1982:47-49), and Nietschmann (1972, 1973) show that the amount of non-tim- ber tropical forest products collected varies with the season because of climatic changes, govern- ment regulations Hart 1978), or seasonal work in agriculture. To avoid skewed results, research- ers should randomly select for observation the same number of weeks within each month of the year and the same number of days within each week (Gross 1984). Even a continuous yearlong study may not reflect he typical composition of goods removed from the forest because the amounts of game and plants vary from year to year (Ashton 1989; Leighton and Leighton 1982; Vickers 1979). Contraction or expansion of the larger economy or a crisis in the household may also influence he degree o which people depend on forest resources. Redford and Robinson 1987) show that only 3 of the 21 studies they review on hunting n the neotropics asted more than a year. Researchers can begin to decide whether the natural supply of non-timber tropical forest products during their study period is represen- tative by comparing regional climatic informa- tion during the period of the study with infor- mation from past years. QUANTITY Researchers should distinguish between two types of quantities: the inventory (the stock quantity n the forest) and the flow (the quantity actually used by people). Some researchers alue the inventory (e.g., Caballero, Toledo, and Ar- gueta 1978; Peters, Gentry, and Mendelsohn 1989; Prance et al. 1987; Toledo, Caballero, and Argueta 1978), others the flow (e.g., Godoy and Feaw 1989; Schwartzman 1989), and still others value both (e.g., Padoch and de Jong 1989). For most purposes, the value of the inventory is a meaningless oncept related neither o present or to sustainable use. The difference n value be- tween the flow and the inventory can be large. Padoch and de Jong (1989) measured he poten- tial and the actual yearly profits of non-timber tropical forest products per hectare in two sep- arate plots in Iquitos, Peru, and found that the value of the goods extracted was only 2.5-3.5% of the value of the inventory. If we assume this relation between flow and inventory, the value reported by Peters and his coworkers or Iquitos would correspond o about 15/ha/year for flow, comparable o the flow value reported by Padoch and de Jong (1989) for a neighboring ite. The most accurate method of valuing the products extracted s to identify, count, weigh, and measure hem as they enter the village each day (e.g., Bergman 1986; Dufour 1983; Hill and This content downloaded from on Sat, 15 Feb 2014 12:04:04 PMAll use subject to JSTOR Terms and Conditions  1993] GODOY ET AL.: METHOD FOR VALUATION 223 Hawkes 1983; Stearman 1990; Vickers 1980; Wilkie and Curran 1991). This technique may be difficult f extractors ive scattered n different villages, if they consume goods in the forest (Nietschmann 1973:150), or if they extract goods at night or before dawn. To address he problem of scattered users, researchers must observe a random sample of villages and households. To estimate the type and amount of goods con- sumed n the forest, hey can either ask extractors or randomly follow them and record their con- sumption. We know of only one scholar who has recorded extraction of products from the forest at night (Scaglione 1986). In weighing oods, researchers hould use scales or balances suited to the different ypes of goods collected (Hart 1978; Hurtado and Hill 1987, 1990). Weighing the entire daily yield may be time-consuming, but informants can be trained to help. If it is impossible to weigh each animal, the average weight for the species can be deter- mined and used instead. For instance, Yost and Kelley (1983) approximated he weight of adult animals hunted using the average weight for the species but reduced he weight by 60 f hunters classified he animal as young. Redford and Rob- inson (1990) recommended hat researchers eigh a sample of at least five adults to establish the average weight of a species. If such a sample is weighed only once, it may not reflect he seasonal variation in the weight of the animals during a year. Again, if the researcher annot count the entire catch, average weights for aggregate uan- tities (e.g., a basket of fish) should be estimated (Gregory nd Altman 1989:148). If extractors at part of the animal in the forest, the weight of the animal should only be estimated for the parts brought back to the village. We should not value what is eaten because that is part of the cost of hunting. When researchers o not see the animal, they can estimate ts weight rom hunters' reports of its size (Gregory and Altman 1989:146). When counting and weighing extracted goods, researchers must also identify the species in- volved. Redford and Robinson (1987) report hat the irregular use of scientific names hinders the comparison of hunting studies in the Amazon. Since species dentification may be difficult, hey recommended elsewhere 1990) that researchers take photographs and collect skulls from which, in combination with information on the weight and the location of the kill, taxonomists can later identify them. Researchers hould also note the sex and the age of the animals and the number of fetuses of pregnant ones; such information helps in assessing the sustainability of hunting and fishing (Redford and Robinson 1990). Through direct observations researchers can measure the extraction area. They can map the terrain covered by extractors using aerial pho- tographs (Bergman 1980:32) or pedometers (Stearman 1990). Vickers (1980) calculated dis- tance indirectly by measuring he time hunters were out in the forest; he found that hunters on average traveled 2-3 km per hour. Researchers can also follow extractors as they forage or ques- tion them about the location of their activities, preferably t the end of each day (Vickers 1983: 457). They can also ask them to pinpoint the locations of their kills on maps; Vickers found an informant who gave him a detailed map of local hunting rails. A new device for measuring the size of the extraction area is the Global Po- sitioning System (GPS) receiver, which picks up satellite signals to give its users their ground co- ordinates Baksh 199 1; Behrens and Sever 199 1). Researchers re limited in the number of per- sons they can directly observe, but they can in- crease he amount of information hey can gather by questioning extractors about their activities and by training hem to help gather nformation. With literate populations, researchers an train extractors o keep diaries on the results and lo- cations of extractive expeditions and on the al- location of their time in the forest. White (1976), Bergman 1986), Alcorn (1981), Dufour (1983), Padoch and de Jong (1989), and Stephen Beck- erman (pers. comm.) trained nformants o keep diaries, and extractors have been taught o weigh and measure the daily catch (Bergman 1986:32; Nietschmann 1973:21; Stocks 1983; Wilkie 1989). Although extractors may give accurate nfor- mation, one should avoid relying oo heavily on their accounts. Distortions arise because people may be careless n their record keeping, ie to the researcher Nacham 1984), or forget past events. The accuracy of recall information drops when people are asked to remember events far in the past (Bernard et al. 1984). Besides errors from carelessness, ying, and faulty memory, people bias information by reporting culturally signifi- cant events or information they believe inter- viewers want to hear. For instance, Dufour (pers. comm.) says that the Tucano of Colombia do not report all the game and fish caught because they This content downloaded from on Sat, 15 Feb 2014 12:04:04 PMAll use subject to JSTOR Terms and Conditions
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks