Cataloging Life on Earth: Implications for International Trade

Erik Lichtenberga,b, Lars J. Olsona, and Chad Lawleya

a

Department of Agricultural and Resource Economics, 2200 Symons Hall, University of Maryland, College Park, MD 20742-5535 b

Corresponding author. Email: [email protected]. Telephone: 301-405-1279. Fax: 1-

301-314-9091.

November 13, 2008

 

1

Abstract While systematics is understood to be important, its value has not been quantified and the field remains subject to severe funding pressures. We develop a model of optimal investment in systematics in the context of preventing errors in screening imports for invasive pests. The model is used to derive an expression for the annualized value of knowledge stocks which is then quantified using data from screening of imports by the US Department of Agriculture (USDA). The resulting estimates indicate that each dollar spent on systematics and quarantine support services generates at least $4.87 in benefits to the US economy.

Glossary of Acronyms

 

SPS

Sanitary and phytosanitary

USDA

US Department of Agriculture

APHIS

Animal and Plant Health Inspection Service, USDA

CBP

Customs and Border Protection, Department of Homeland Security

EAN

Emergency Action Notification

APHIS/PPQ

Plant Protection and Quarantine Division, APHIS

2

I.

Introduction  Systematics—the branch of biology that deals with the identification and classification of

organisms and the description of their life histories—is essential for humanity’s ability to manage the world’s genetic resources. Understanding threats to biodiversity from habitat loss, climate change, invasive species and human exploitation depends on our ability to discover, classify, and describe the species that exist on Earth. Bioprospecting for pharmaceuticals and other useful compounds presupposes a stock of knowledge about the organisms from which those compounds are derived. The use of germplasm from seed banks in crop breeding presupposes knowledge about life histories of plant varieties and about interspecific compatibility. Sanitary and phytosanitary (SPS) policy toward invasive pests presupposes that potentially hazardous organisms can be identified accurately and that we understand how to prevent their entry or control them (Mayr 1968; Davies, King, and Smith 2004; McAusland and Costello 2004; Canadian Food Inspection Agency 2006). Despite an awareness that systematics is fundamental to all organismal biology the field has been in decline for some time. The spectacular successes achieved in molecular and cell biology and in genetics in recent years have eroded both funding for and scientific interest in systematics (Wilson 1985). Over the past decade the U.S. has experienced a shift in doctoral degrees from organismal to molecular biology and funding for natural history collections has declined (Gropp 2003; Dalton 2003). Likewise, a recent report by the British House of Lords (2002) found decreases in funding at major UK research institutions on the order of 15-25% over the preceding decade, lack of interest in the field among young biologists, and a shift toward molecular and genetic approaches and away from traditional classification methods, all of which were found to threaten the long term viability of the discipline in the UK. This deterioration of  

1

the stock of human capital in systematics may have detrimental long term effects on our ability to tackle biological resource problems that are growing in scope and importance (Wilson 1985; Davies, King, and Smith 2004). The Conference of the Parties of the Convention on Biological Diversity has called for a global capacity building in systematics and has established the Global Taxonomy Initiative, designed to facilitate international cooperation and coordination with regard to taxonomic information (Conference of the Parties 2008). The contradiction between the growing importance of systematics in high visibility policy issues and the decline in its organizational and financial status suggests that society has not fully appreciated its value. Indeed, it is difficult to value scientific knowledge that is not directly used in the marketplace. 1 This paper attempts such a valuation exercise by investigating the economic value of systematics knowledge embodied in scientists and physical reference collections housed in museums and research institutions in the context of enforcement of sanitary and phytosanitary (SPS) trade regulations. Systematics knowledge reduces two kinds of errors that can occur in screening imports for invasive pests: (1) incorrectly determining that a harmful organism detected in an import shipment is safe and can be allowed to enter (or, equivalently, failing to detect a harmful exotic organism due to lack of knowledge about where to look for it); and (2) incorrectly determining that a harmless organism detected in an import shipment is harmful and either forbidding entry or requiring unnecessary expensive treatment prior to entry.                                                              1

The economic literature on managing genetic resources presupposes the existence of a stock of knowledge that can be used to ascertain the value of genetic material but does not examine the accumulation of that stock of knowledge or its value. The literature on germplasm concentrates on the acquisition, maintenance, and efficient search of genetic material in collections like seed banks (Evenson and Kislev 1976; Evenson and Gollin 1997; Gollin, Smale, and Skovmand 2000; Woo and Wright 2000; Pardey et al. 2001; Smale and Day-Rubinstein 2002; Zohrabian et al. 2003; Koo, Pardey, and Wright 2003). The literature on bioprospecting concentrates on preservation of habitat modeled as a stock of hitherto undiscovered genetic material (Polasky and Solow 1995; Simpson, Sedjo, and Reid 1996; Rausser and Small 2000; Kassar and Lasserre 2004; Costello and Ward 2004). A second, but smaller, branch of the literature focuses on the use of tariffs and inspections in phytosanitary trade policy given a stock of knowledge sufficient to identify potential invasive pests (McAusland and Costello 2004).

 

2

We use data from screening of imports by the US Department of Agriculture’s (USDA’s) Animal and Plant Health Inspection Service (APHIS) to estimate the value of taxonomic knowledge in preventing potentially harmful organisms that can accompany international trade. The estimated annualized value of systematics for these services is over $180 million, or $4.87 for each dollar USDA spends on systematics and quarantine support services combined. The rules that govern the application of food safety and animal and plant health regulations in international trade are defined in the SPS Agreement entered into force with the establishment of the World Trade Organization in 1995 following the Uruguay Round of the General Agreement on Tariffs and Trade. SPS regulations are among the most contentious components of international trade agreements, due largely to widespread suspicion that they are used as instruments of protectionism in the face of limits on direct import restrictions imposed by international trade agreements. For that reason, World Trade Organization rules require all SPS regulations to have a defensible scientific basis. This paper speaks to that scientific basis. As such, it is relevant to the literature on the economics of SPS regulations (e.g. Roberts 1999; Beghin and Bureau 2001; Mumford 2002). In the following section we describe the process of screening for invasive pests in shipments imported into the US. Section III describes the data used to estimate the annualized value of existing knowledge stocks. The results of this estimation procedure, as well as qualifications, are presented in section V. The final section provides concluding remarks.

II.

Screening for Invasive Pests in Import Shipments  APHIS regulates imports of plants, plant materials, soils, biological control organisms,

animals, and packaging materials under the authority of the Plant Protection Act of 2000. This act consolidated and harmonized 11 earlier statutes governing SPS regulations, including the  

3

Plant Quarantine Act (first enacted in 1912), the Plant Pest Act (first enacted in 1957), and the Federal Noxious Weed Act (enacted in 1974). The Plant Protection Act of 2000 gives APHIS the authority to inspect all incoming import shipments containing perishable materials (including packing materials) for potentially harmful exotic organisms, including plant and animal pests, weeds, and diseases. Actual performance of most inspections was delegated to Customs and Border Protection (CBP) with the establishment of the Department of Homeland Security in 2003; APHIS continues to inspect plants imported for propagation. Surveillance inspections conducted by CBP and APHIS personnel utilize targeted sampling of roughly 2% of each shipment, adjusted according to inspectors’ perception of the likelihood that an invasive pest organism is present. The adequacy of surveillance inspections is verified by a separate sampling program, Agricultural Quarantine Inspection Monitoring, which utilizes intensive sampling of selected cargoes based on a hypergeometric sampling scheme (Venette, Moon, and Hutchison 2002). Shipments from areas that are certified pest-free, have implemented accepted precautionary measures, or conduct inspections prior to shipment (and are thus pre-cleared), and shipments in which commodities are transported under conditions known to achieve effective preventive control (and are thus considered pre-treated) are exempt from universal surveillance inspection and are inspected only periodically to monitor compliance with and effectiveness of pre-clearance and pre-treatment protocols (Follett and Neven 2006) Between 1984 and 2000 approximately 750,000 nonindigenous pests comprising over 2,300 species were intercepted in baggage or cargo by APHIS inspectors at 160 points of entry into the U.S. Among these were over 565,000 insects, 6,000 mites, 11,500 mollusks, 440 nematodes, 95,840 pathogens and 50,000 weeds. About 22% of the pests originated in Central or South America, 18% from the Caribbean, 16% from Mexico, 16% from Asia, 9% each from

 

4

Europe and the Pacific (excluding Asia), with the bulk of the remainder from Africa and the Middle East (McCullough et al. 2006). When organisms are detected in cargo, CBP puts the shipment on hold. Samples of the organisms and the host material in which they were found are brought to APHIS identifiers housed in the largest ports of entry, who make initial identifications whenever possible. Identifications are then used to determine quarantine status. There are two general types of quarantine status. Cargoes in which no exotic organisms have been detected or in which organisms detected are known to pose no risk of harm are considered non-actionable and are allowed to enter the US untreated. Cargoes in which exotic organisms are detected that do pose risk of harm are considered actionable. If the potential pest can be eradicated by fumigation, exposure to cold, or other forms of treatment, the cargo is allowed to enter the US after suitable treatment. Cargoes with actionable pests for which no reliable treatment methods are available are not allowed to enter the US; they may be diverted to other countries or destroyed, at the discretion of the shipper. Figure 1 depicts the decision tree involved in this process. Final identifications are made either by port identifiers, by national level identifiers located at APHIS national headquarters in Riverdale, Maryland, or by scientists in the systematics laboratories of the US Department of Agriculture’s Agricultural Research Service, located in Beltsville, Maryland and at the Smithsonian National Museum of Natural History in Washington, DC. Most identifications are made by APHIS port identifiers who have been certified to determine quarantine status of specific organisms (i.e., given authority to discard or require treatment of cargoes) after exhibiting sufficient expertise. Scientists in the systematics laboratories are responsible for conducting that certification process.

 

5

Scientists in the systematics laboratories and at APHIS headquarters handle the most difficult identifications. Samples of organisms detected are shipped by overnight express to these locations. Identifications must be made by the close of business the following day (i.e., the day the sample arrives at the lab). Digital images may be sent when believed to be sufficient for identification, in which case determination of quarantine status may be made the same day. Scientists in the systematics labs perform these identification services in addition to their regular research duties. Identifiers can make two kinds of mistakes in making decisions about exotic species found in shipments of imports. Identifiers may decide that an organism poses no risk of causing damage in the US and allow shipments containing the organism to enter. If that decision is wrong (a false negative), the US will suffer some damage and incur some control costs from the ensuing pest invasion. Alternatively, identifiers may decide that an organism does pose a risk of causing damage in the US. When identifiers decide that an organism is a quarantine pest, then the exporter’s agent chooses whether to treat the shipment, destroy it, or re-export to another country. If that decision is wrong (a false positive), the importer and shipper combined will suffer losses due to unnecessary treatment costs or foregone benefits from rejection of the shipment. In Section I of the Supplementary Information we present a formal optimization model of this decision process. The model assumes that the objective of identifiers is to minimize the social cost of these errors, which consists of the expected costs of damage due to entry of organisms in shipments that should not have been allowed plus expected foregone benefits or treatment costs of shipments that were improperly rejected or fumigated. APHIS port identifiers and scientists in the USDA’s systematics laboratories are able to influence these expected costs

 

6

by expending effort to ensure that identifications are made with greater accuracy. Greater accuracy of identification comes at a cost, however, in terms of work time expended by identifiers. Accuracy of identification also depends on the existing stock of knowledge about organisms, embodied in the expertise of scientists making identifications and in the resources they can fall back on when their expertise alone is insufficient, such as reference collections of specimens and the scholarly literature available to them in libraries and online. 2 We use the model to derive an expression for the annualized value of existing stocks of systematics knowledge. We use that expression as the basis for deriving an empirical estimate of the value of those knowledge stocks.

III.

Data and Estimation Method  We use data collected by APHIS during its screening of imports of plants and plant

materials in combination with data from Foreign Agricultural Trade of the United States (FASOnline) to estimate the value of knowledge embodied in the scientists who identify                                                              2

The stock of systematics knowledge can be augmented through independent research and learning-by-doing. In the case of screening import shipments for harmful invasives, independent research involves observing organisms around the world in their natural habitats, recording their life histories, collecting specimens for reference collections, etc. while learning-by-doing occurs because experience reduces the time needed to identify an organism as well as broadening identifiers’ personal knowledge of organisms. Additionally, specimens of newly encountered organisms are added to reference collections, thereby enhancing scientists’ ability to identify those organisms in the future. At the same time, portions of the existing stock of knowledge become obsolete over time. In the case of screening import shipments for potentially damaging organisms, portions of the existing literature may become obsolete as classification systems and nomenclature change, as the composition of imports changes (rendering some knowledge irrelevant), and as further investigation overturns older beliefs about life histories. Additionally, the personal knowledge of experienced individual systematists may be lost as they retire without training replacement scientists. The latter phenomenon is increasingly likely as the number of systematists has been shrinking over time. In 2003, for example, Wilson (2003) estimated the number of biologists working in systematics at about 6,000, compared to estimates less than two decades earlier of 8-10,000 in North America alone (Gaston and May 1992). Shortages of systematists are especially acute for the phyla to which most invasives belong, which tend to have few systematists per species. Extrapolating from Australian data, Gaston and May (1992) estimate that there are 840 known species and an additional 400 unrecorded species for each taxonomist working on insects and spiders. They cite US data indicating that only about 2% of taxonomists work on microorganisms, of which there are roughly 100,000 known species and millions of unrecorded species (Wilson 2003).

 

7

potential pests and in the reference materials available to them. The decision tree depicted in Figure 1 provides the basis for our analysis. The annualized value of current knowledge stocks can be estimated by considering the costs of the errors that would occur if systematics knowledge did not exist. Three kinds of errors may occur while screening imports, two of which occur during identification. First, some shipments will contain organisms that are detected and erroneously determined to be actionable (see Figure 1). Second, some shipments will contain organisms that are correctly determined to be actionable but are incorrectly determined to be untreatable. Third, some shipments will contain organisms that actually are hazardous but go undetected. Our estimate is based on the assumption that all organisms detected in import shipments would automatically be categorized as actionable and untreatable and would therefore be re-routed or destroyed. Under this assumption, detected non-actionable organisms and actionable but treatable organisms alike would be categorized erroneously as actionable and untreatable, so that any shipment containing either would either be rejected for entry into the US or destroyed. 3 Let πj denote the frequency with which at least one organism is detected in import cargoes of commodity j, σj the frequency with which detected organisms are determined to be actionable, and τj the frequency with which detected, actionable organisms are determined to be treatable. Let Vj denote the value of imports of commodity j and Tj the cost of treating commodity j. If imports in which at least one organism is detected constitute a small share of

                                                             3

An alternative approach would be to assume that in the complete absence of accumulated systematics knowledge all organisms detected in shipments would be categorized as non-actionable and thus allowed entry. As we discuss in detail in Section III of the Supplementary Information, this approach is difficult to implement and the fragmentary data available (detailed in Tables S3 and S4) suggest that it would yield a substantially higher value than the approach used here.

 

8

total imports of any commodity, the expected cost of errors from erroneous rejection of safe and safe-if-treatable cargoes can be estimated as:

∑ j π j {(1 − σ j )V j + σ jτ j (V j − T j )} . We show in Section I of the Supplementary Information that this expression can be considered an approximation of the true annualized value of existing stocks of systematics knowledge. CBP and APHIS record all shipments of plants and plant materials on APHIS/PPQ Forms 264 and 280, which respectively record all propagatable and non-propagatable cargoes entering the US in all ports except certain ones on the US-Canada and US-Mexico borders. We use these data to calculate the volume of imports of each commodity Nj. Details of these and all other calculations used in this analysis are given in Section II of the Supplementary Information. CBP and APHIS inspectors record information about all shipments in which at least one organism has been detected on APHIS/PPQ Form 309; since discovery of a single actionable pest is sufficient for regulatory action, additional inspection and/or record-keeping is not required and often not undertaken (McCullough et al. 2006). Information about all organisms suspected of being quarantine pests is then entered into APHIS’ PEST-ID database. Reporting of detected organisms that are not quarantine pests is incomplete because information about nonactionable pest detections may not be entered into electronic databases even when a paper Form 309 is filled out. The PEST-ID data are used to estimate a lower bound of the volume of imports of each commodity in which at least one organism of potential concern has been detected, Πj. If an organism is determined to be a quarantine pest, APHIS issues an Emergency Action Notification, APHIS/PPQ Form 523 (EAN), specifying the disposition of the cargo (refusal of entry, entry after appropriate treatment, etc.). Beginning September 2006, the EAN data also include the identification number from APHIS/PPQ Form 309, permitting reliable merging of the  

9

EAN and PEST-ID data. Data from the last quarter of 2007 are not yet available. We thus used data for the period September 1, 2006 through August 31, 2007 to estimate the value of systematics knowledge stocks in screening for potential invasives in import shipments. All detections of actionable pests and whether they are treatable are reported in the combined PEST-ID/EAN data. This provides a complete accounting of the volumes of imports of each commodity containing actionable, non-treatable organisms, Aj, and actionable, treatable organisms, Zj. (See Section II of the Supplementary Information for details.) The combined PEST-ID/EAN data are used to estimate the probability that actionable, treatable organisms are detected in an import shipment, πjσjτj = Zj/Nj, and the probability that non-actionable organisms are detected in an import shipment, πj(1-σj) = (Πj- Zj -Aj)/Nj. Shipments are aggregated into commodity groups corresponding to those specified in the Harmonized Tariff Schedules of the US. The unit value of each import shipment j, vj, is estimated by dividing the total customs value of imports of the commodity contained in the shipment by the volume of imports of the commodity contained in the shipment, both as reported in Foreign Agricultural Trade of the US (FASOnline). The total value of each import shipment j, Vj, is then estimated as vjNj, the product of its unit value (derived from trade statistics) and volume (derived from APHIS/PPQ 280/264 data). Estimates of treatment costs Tj were obtained from fumigation price lists provided by firms in New York, Philadelphia, Long Beach, San Pedro/San Diego, and Nogales. The results of these calculations are shown in detail in Table S1 of the Supplementary Information.

 

10

IV.

Results  As indicated by the detailed figures in Table S1, detection rates of organisms of potential

concern tend to be relatively low (under 5% of all shipments overall), suggesting that ignoring potential price effects of erroneous rejections is reasonable. 4 Organisms of potential concern are detected most frequently in bay leaves and thyme (about 32% of shipment volume), pineapple, miscellaneous cut flowers, other herbs, and coriander (between 6 and 11% of shipment volume each). Other commodities with detection rates of 4 to 5% include legumes, celery, alstromeria 5 , avocados, citrus, and dates. In contrast, many commodities have detection rates under 1% (bulbs, nuts, snapdragons, orchids, melons, grapes, coconuts, strawberries, eggplants, cranberries, flax seed, watermelons, carrots, turnips, edible roots, kiwi, tomatoes, lilies, asparagus, apples, stone fruits, cucumbers, potatoes, and roses). Actionable pests are detected in relatively small shares of most commodities. Commodities with significant shares of actionable pest detections include basil and thyme (18% of shipment volume), pineapples (7% of shipment volume), and miscellaneous cut flowers (4% of shipment volume). Actionable pests were detected in 1-2% of imports (by volume) of coriander, alstromeria, other herbs, ginger, cassava, celery, chrysanthemum, spinach, durian, legumes, Brassica species (cabbages, broccoli, etc.), dasheen, mushrooms, gypsophila, and dates. Actionable pests were detected in well under 1% of import volumes of most other commodities. In the overwhelming majority of cases, actionable pests detected were treatable. Exceptions include dates, mangoes, mushrooms, pears, papayas, and spinach: Virtually all                                                              4

 The low rates at which exotic species are detected further suggests that our implicit assumption that the volume and composition of trade would be the same in the absence of any ability to distinguish harmful from harmless exotic species is also reasonable.

5

 

Commonly known as the Peruvian lily, this species is widely marketed as a cut flower.

11

actionable pests found on these commodities were classified as untreatable. Other commodities for which significant shares of actionable pests detected were non-treatable include coriander, bananas, cranberries, carrots and turnips, avocados, orchids, and Brassica species. The annualized value of current knowledge stocks, estimated as the annual cost of erroneous rejections of safe and safe-if-treatable imports that would occur in the absence of existing systematics knowledge, is substantial at just over $180 million. A handful of commodities account for the majority of this cost, notably pineapples (24% of the total), miscellaneous cut flowers (16% of the total), avocados (13% of the total), citrus (11% of the total), bananas (5% of the total), and peppers and legumes (3% of the total each). Other commodities with a cost of erroneous rejection on the order of 1-2% of the total include other herbs, cuttings/plants/roots, other vegetables, pears, raspberries, mangoes, alstromeria, chrysanthemums, onions, and Brassica species. Table S1 of the Supplementary Information presents the details of these results. This estimate suggests that the return on investment in systematics at the US Department of Agriculture is extremely high. USDA operates four systematics laboratories at a total annual cost of under $10 million: the Systematic Entomology Laboratory, with an annual budget of about $4.1 million; the Systematic Mycology and Microbiology Laboratory, with an annual budget of about $2.3 million; the Nematology Laboratory, with an annual budget of about $2.4 million; and the Animal Parasitic Disease Laboratory, whose systematics component has an annual budget of about $1 million. Direct expenditures on systematics and collections at these laboratories amount to under $2 million annually. APHIS’ National Identification Services branch, which houses a number of identification specialists, handles determination of quarantine pests, and manages pest surveillance data, has an annual budget of about $5 million. The total

 

12

annual federal appropriation for agricultural quarantine inspection (net of user fees levied on importers to cover the cost of inspection) is $27 million. The annualized value of knowledge created by expenditures explicitly earmarked for systematics is about 18 times their annual cost. More realistically, the annualized value of the knowledge created by all systematics activities at USDA (including research activities that contribute indirectly to systematics) is 4.87 times the roughly $37 million annual cost of USDA’s systematics laboratories and quarantine support services combined, i.e., each dollar spent on systematics and quarantine support services combined generates $4.87 in avoided costs of screening errors. For comparison, rates of return on R&D in agriculture, which are generally considered quite high, are typically estimated to run on the order of $0.40-0.50 per dollar invested (Alston et al. 2000). Our estimation method is subject to both upward and downward biases. Our use of average rather than marginal error rates likely biases our estimates upwards. 6 There are two significant sources of downward bias: (1) under-reporting of non-actionable organism detections and (2) cargoes that were pre-cleared or received precautionary treatment. 7 A significant share of detections of non-actionable organisms are not reported in PESTID, even when inspectors fill out paper versions of APHIS PPQ Form 309. Our estimate of Πj                                                              6

As we show formally in Section I of the Supplementary Information, the value of systematics knowledge stocks equals the marginal avoided cost of errors attributable to those stocks at current levels, that is, the reduction in the costs of errors due to a small increment in current levels of those stocks. The method we use can be thought of as using the average value of existing systematics knowledge stocks as an approximate estimate of the price or marginal value of current knowledge. Existing knowledge stocks are sufficiently large that marginal costs of errors should decrease as stock levels increase, i.e., reductions in the costs of errors due to increments in knowledge become smaller as existing knowledge grows. Decreasing marginal costs mean that the marginal value of existing knowledge stocks is less than the average value, so that the average value of knowledge implicitly used in this procedure overstates the true value. 

7

 A third source of potential downward bias is our assumption that detection rates would be the same in the absence of existing systematics knowledge. It seems likely that detection rates would be lower without that knowledge, especially for such items as pathogens, weed seeds, and insect eggs. Ignoring changes in detection rates is in keeping with our focus on the impact of marginal changes in systemastics knowledge, however.

 

13

therefore understates the true number of shipments of each commodity in which at least one organism has been detected. The degree of underreporting is difficult to assess. It appears to remain significant even though it has declined in recent years due to a transition to web-based reporting, more thorough record-keeping by CBP, and other factors (Cavey 2008, Lee 2008). Under-reporting of detections of non-actionable pests means that our estimates of the costs of erroneous rejections of safe import shipments are biased downward. Until the past year or two, detections of non-actionable organisms that failed to be recorded in PEST-ID and its predecessors amounted to as much as 30-40% of total detections (Lee 2008). Increased emphasis on record-keeping by CBP, increased use of web-based record-keeping, and other factors appear to have reduced the degree to which non-actionable detections go unrecorded in electronic databases. Even so, the PEST-ID data likely continue to underestimate the frequency with which non-actionable pests are detected to a significant extent. Our treatment of cargoes that were not inspected because they were pre-cleared or because they received precautionary treatment (such as exposure to cold or fumigation) prior to arrival in the US is another source of downward bias. We treat these cargoes as if they were inspected but no organisms were detected, hence as if they would not be subject to erroneous rejection in the absence of systematics knowledge. Yet neither pre-clearance nor precautionary treatment would be feasible in the complete absence of systematics knowledge. Pre-clearance programs presume knowledge about the prevalence of potential exotic invaders, effective sanitation measures, and ability to detect invasive pests prior to export. Reliance on precautionary treatment presumes knowledge about treatment effectiveness (see for example Follett and Neven 2006). The use of these programs is extremely high for some commodities. As shown in detail in Table S2 of the Supplementary Information, almost all stone fruits, apples,

 

14

pears, and cranberries are either pre-cleared or, to a lesser extent, subject to precautionary treatment while sizeable shares of asparagus, kiwis, mangoes, grapes, and tubers are subject to precautionary treatment or, to a lesser extent, pre-clearance. As a result, the downward bias introduced by our treatment of these commodities could be substantial.

V.

Summary and Conclusions  Systematics, the branch of biology that deals with the identification and classification of

organisms and the description of their life histories, is essential to our ability to manage biological resources like biodiversity, germplasm, and the threat of invasions by exotic pests. It is one of the oldest branches of biology and constitutes the foundation on which organismal biology stands. Arguably because of its venerability and the magnitude of its existing stock of human capital, it has fallen into disfavor among emerging scientists and is declining at a rapid rate. This paper investigates the value of systematics knowledge in the context of screening imports to prevent invasions of potential exotic pests. We use data from screening of imports by APHIS to estimate the annualized value of systematics knowledge under the assumption that in the absence of those knowledge stocks import shipments in which non-actionable or actionable but treatable organisms were detected would erroneously be denied entry. Our conservative treatment of pre-cleared and precautionarily treated cargoes and limitations in data recording suggest that the estimates we derive are biased downward, most likely significantly. We compare this estimated annualized value of systematics knowledge to the annual costs of running USDA’s systematics laboratories as a means of estimating the return on investment in systematics. Our estimates, which are likely conservative, indicate that the return to all systematics activities at USDA is extremely high, on the order of $4.87 for each dollar  

15

spent on systematics and quarantine support services combined, a figure that far exceeds the rate of return on agricultural research in general (itself believed to be quite high). Thus, the services rendered by USDA’s systematics activities appear to be extremely valuable, especially relative to their cost.

 

16

Acknowledgements This research was funded by a grant from the Beltsville Agricultural Research Center, Agricultural Research Service, US Department of Agriculture. We are grateful for the assistance of Mike Schauff, Alma Solis, Joseph Cavey, Danny Lee, Pete Touhey, and Ethan Kane and for comments at seminars at the University of Arizona, the University of Maryland, and the University of California-San Diego. Responsibility for the content is ours alone.

 

17

References Alston, J.M, C. Chan-Kang, M.C. Marra, P.G. Pardey, and TJ Wyatt. 2000. A Meta-Analysis of Rates of Return to Agricultural R&D, International Food Policy Research Institute, Washington, DC. Beghin, J.C. and J.C. Bureau. 2001. Quantitative Policy Analysis of Sanitary, Phytosanitary and Technical Barriers to Trade, Economie Internationale 87, 107-130. Canadian Food Inspection Agency. 2006. Action Plan for Invasive Alien Terrestrial Plants and Plant Pests, Phase 1, http://www.inspection.gc.ca/english/plaveg/invenv/action/phase1e.shtml#3.2.3. Cavey, J. 2008. Personal communication. Conference of the Parties. Convention on Biological Diversity. 2008. http://www.cbd.int/convention/cops.shtml. Costello, C.J. and M.B. Ward. 2006. Search, Bioprospecting, and Biodiversity Conservation, Journal of Environmental Economics and Management 52, 615-626. Dalton, R. 2003. Natural History Collections in Crisis as Funding is Slashed, Nature 423, 575. Davies, H., N. King, and R. Smith (eds.). 2004. Taxonomy: Targeting Invasives, BioNETINTERNATIONAL. Evenson, R.E. and D. Gollin. 1997. Genetic Resources, International Organizations, and Improvement in Rice Varieties, Economic Development and Cultural Change, 471-500. Evenson, R.E. and Y. Kislev. 1976. A Stochastic Model of Applied Research, Journal of Political Economy 84, 265-282. FASOnline: US Trade Internet System, http://www.fas.usda.gov/ustrade. Follett, P.A., and L.G. Neven. 2006. Current Trends in Quarantine Entomology, Annual Review of Entomology 51, 359-385. Gaston, K.J. and R.M. May. 1992. Taxonomy of Taxonomists, Nature 356, 281-282. Gollin, D., M. Smale, and B. Skovmand. 2000. Searching an Ex Situ Collection of Wheat Genetic Resources, American Journal of Agricultural Economics 82, 812-827. Gropp, R.E. 2003. Are University Natural Science Collections Going Extinct? BioScience 53(6), 550. House of Lords. 2002. What on Earth? The Threat to Science Underpinning Conservation, Third Report, Select Committee on Science and Technology, London.

 

18

Kassar, I. and P. Lasserre. 2004. Species Preservation and Biodiversity Value: A Real Options Approach. Journal of Environmental Economics and Management 48, 857-879. Koo, B., P.G. Pardey, and B.D. Wright. 2003. The Economic Costs of Conserving Genetic Resources at CGIAR Centres, Agricultural Economics 29, 287-297. Koo, B. and B.D. Wright. 2000. The Optimal Timing of Evaluation of Genebank Accessions and the Effects of Biotechnology, American Journal of Agricultural Economics 82, 797-811. Lee, D. 2008. Personal communication. Mayr, E. 1968. The Role of Systematics in Biology, Science 159, 595-599. McAusland, C. and C. Costello. 2004. Avoiding Invasives: Trade-Related Policies for Controlling Unintentional Exotic Species Introductions, Journal of Environmental Economics and Management 48, 954-977. McCullough, D.G., T.T. Work, J.F. Cavey, A.M Liebhold and D. Marshall. 2006. Interceptions of Nonindigenous Plant Pests at US Ports of Entry and Border Crossings over a 17-Year Period, Biological Invasions 8, 611-630. Mumford, J.D. 2002. Economic Issues Related to Quarantine in International Trade, European Review of Agricultural Economics, 29, 329 48. Pardey, P.G., B. Koo, B.D. Wright, M.E. Van Dusen, B. Skovmand, and S. Taba. 2001. Costing the Conservation of Genetic Resources: CIMMYT’s Ex Situ Maize and Wheat Collection, Crop Science 41, 1286-1299. Polasky, S. and A.R. Solow. 1995. On the Value of a Collection of Species. Journal of Environmental Economics and Management 29, 298-303. Rausser, G.C. and A.A. Small. 2000. Valuing Research Leads: Bioprospecting and the Conservation of Genetic Resources, Journal of Political Economy 108, 173-206. Roberts, D. 1999. Analyzing Technical Trade Barriers in Agricultural Markets: Challenges and Priorities, Agribusiness 15, 335-54. Simpson, R.D., R.A. Sedjo, and J.W. Reid. 1996. Valuing Biodiversity for Use in Pharmaceutical Research, Journal of Political Economy 104, 163-185. Smale, M. and K. Day-Rubenstein. 2002. The Demand for Crop Genetic Resources: International Use of the US National Plant Germplasm System, World Development 30, 1369-1655. Venette, R.C., R.D. Moon, and W.D. Hutchison. 2002. Strategies and Statistics of Sampling for Rare Individuals, Annual Review of Entomology 47, 143-174. Wilson, E.O. 2003. The Encyclopedia of Life, Trends in Ecology and Evolution 18, 77-80.  

19

Wilson, E.O. 1985. Time to Revive Systematics, Science 230, 1227. Woo. B. and B.D. Wright. 2000. The Optimal Timing of Evaluation of Genebank Accessions and the Effects of Biotechnology, American Journal of Agricultural Economics 82, 797811. World Trade Organization. 2007. Documents of the SPS Committee. http://www.wto.org/English/tratop_e/sps_e/sps_e#documents Zohrabian, A., G. Traxler, S. Caudill, and M. Smale. 2003. Valuing Pre-Commercial Genetic Resources: A Maximum Entropy Approach, American Journal of Agricultural Economics 85, 429-436.

 

20

Shipment j arrives at entry port and is  inspected

No organisms detected  with frequency 1‐πj

At least one organism is detected  with frequency πj Organism is identified as actionable with  frequency σj

Organism is identified as  untreatable with frequency 1‐τj

Organism is identified as not actionable with frequency 1‐σj

Organism is identified as  treatable with frequency τj

Treatment Rejection

Entry

Figure 1. Decision Tree of Screening Imports for Potentially Harmful Exotic Organisms

 

21